How much lab power supply can you get for $450?
Triple isolated outputs, PC control, 0.05% accuracy, is it too good to be true?
http://www.attenelectronics.com/Products/Power_Supply/Programmable_power/2012/0728/92.html
NOTE: Yes, it's a long review, and I spend most of the time complaining about something. I wasn't going to spend the time to try and edit it down. Complaints on a post card please.
Forum Topic: http://www.eevblog.com/forum/blog/eevblog-439-atten-pps3205t-3s-triple-output-power-supply-review'>http://www.eevblog.com/forum/blog/eevblog-439-atten-pps3205t-3s-triple-output-power-supply-review
EEVblog Main Web Site:
http://www.eevblog.com
EEVblog Amazon Store:
http://astore.amazon.com/eevblogstore-20
Donations:
http://www.eevblog.com/donations/
Projects:
http://www.eevblog.com/projects/
Electronics Info Wiki:
http://www.eevblog.com/wiki/

Hi, it's Power Supply review time again. And We've got a pretty high-end bit of kit today. Not A big name brand, one of the cheapies,, but that's the key. It is a precision triple-output power supply.

It's the Atten, let me get this right, PPS3205T-3S and 3 independent isolated variable outputs, zero to 32 volts at 5 amps on two of the channels, and zero to 6 volts at 3 amps on the third channel down here. Completely There it is, PC interface,. there's the USB and serial interface, and it's a precision supply. Forgive me, I won't-- yeah, I'll hold it up.

come on, I'm strong enough, geeze.. This only weighs ten kilos. Linear power supply, of course. precision output, 0.05%.

So they claim. Plus/minus, you know,, 10 millivolts or something like that,. so real precision bit of kit, and a big graphic display on the thing, as we'll see. So The key to this is that it is half the price of probably the next one on the market that's equivalent.

kind of, you know,, triple output high-power linear precision one, would be the Rigol unit, and it's half the price, I Got this for about 450 Australian dollars! So Yeah,, pretty darn cheap for this class of instrument. Is it any good? Smells A bit cheap, but eh, could be good value for money. Let's check it out. And Here it is, one of the first things you notice is that it is actually a quite compact unit, I mean for the power we're talking about here.

I Mean, we're talking about 160 watts per channel, much less on the third channel. But You know,, really quite a high-power linear supply. As I Said,, it does weigh about 10 kilos, but really, it's only 200 millimetres wide here, so it doesn't take up much space on your bench, really,. you know,, compare it with a multimeter.

It's only like two multimeters width, something like that.. It is quite small. It's about 350 millimeters deep, but yeah,, not too bad at all.. Now Yeah,, it's got a carry handle on top, Enh,.

we'll have to take-- crack the thing open to see if it's actually sturdy or not. It's built like a cheap-ish power supply, not as bad-- doesn't feel as bad as many of the cheapies, but it's certainly not like an Agilent or something like that, so. You Know, really? eh,. you get what you pay for.

And On the back here, nothing much happening, standard IEC mains input, 5 amp fused, A fan which is really annoying, we'll get into that,. it is quite loud, it rattles, I don't like it at all. And USB and RS232 interfaces for the PC control. And I believe they would be, presumably, isolated as well.

We'll check on that. And on the bottom of the unit, 4 big, sturdy rubber feet on the thing,, and it looks like big mounts for the huge monster transformer in there, as we'll no doubt see when we take this thing apart. I Assume that they've got shakeproof-- they don't have shakeproof washers on the bottom,, but they might have them on the top side inside. So Build quality is certainly adequate for the price, that's for sure.
Alright, let's power this thing up, and clunk.. Atten Rev 1.0 Software: Mmm, be afraid, be very afraid. But There it is, it's got a triple output display here, Channel 1, Channel 2 Channel 3 One Of the things I don't like is that like, Channel 1 Channel 2 Channel 3 are identical. It doesn't tell you what the you know, like, it doesn't say this one's zero to 32 volts, this is zero to 32, and this is zero to 6.

Why Doesn't it tell you that? I Mean,, you've got to-- you've got to know, if you just walked up to this, you wouldn't know that this one only goes up to 6 volts. You Would have to know when you actually go in and try to type in a value in there to go up to 30 volts, and Channel 3 can't do it. So, you know,, labeling, I Mean geeze,, it's not rocket science. Now As for the display itself,, it's not bad.

Nice Bright-- nice bright backlight, if you're looking directly straight on the thing. If You turn it to the side, it does sort of,. you know,. it does wash out, almost completely, if you move it out like that.

Or If you go up, then it washes out of course. And It does pretty good from below, so if you've got it up high on your bench, you can actually read it reasonably well. Ahh, you know,, it could be better. It's not the best display.

What I'm going to do is peel off this protective coating here,. it does make it look a bit a bit daggy. There We go. Ahh, much better.

Check Out the quality control on this one, folks.. Look, it looks like someone's, like, eyelash or something has been embedded behind the display in there. Someone At the Chinese Factory. Hi, how are you doing if you're watching? Now You can probably hear that, One of the first things you notice is that it, the sound of the fan is incredibly annoying.

Yes,, it does have a variable speed fan, which is great, but listen to it! Clunk, clunk, clunk. Not Sure if you can hear that, but it really is not that great,. listen. And There you go, it just sounds awful, it sounds sick, like there's something wrong with the thing.

I, just the bearings are shit, they use some cheap-ass fan in it. Argh! Anyway, we'll be able to fix that up later, no doubt with a fan retrofit. Umm, but you know,, it's not too bad. They Could have done more with the graphic display, like for example, if you only have one channel turned on, why not, you know,, have an option for a larger single display or something like that, But no,, it's fixed,.

there's no power display. It's great that it shows the voltage to 10 millivolt resolution, current to an excellent 1 milliamp resolution, which is great, but they don't display the power per channel. That would have been kind of nice. I, you know, it's not the best laid out screen at all.

So, I don't know, personal opinion. Big Output button here, I Like it. If You notice-- whoa, overvoltage-- what?! Overvoltage Protection! Fail! What's going on here folks? Haven't seen this before. Overvoltage Protection in the Chann-- "in the channel two", please push shift OVR-- OVPRESET Key to clear the OVP status and reset.
Ah Well, there you go. I've never seen that before. Shift OVPRESET. There We go.

I Gue-- ah,, 34, there we go. Because Channel 2, I Don't know why, is set to 34 volts. Maybe I was playing around with it, but why set it to 34 volts if it's not actually capable of doing that? Crazy! So Feature-wise on the unit here, we've got. it's actually quite, you know,, quite well-equipped, It's got overcurrent protection of course,, we'll take a look at, the OVP reset there is to reset the overvoltage protection.

If it trips, the overvoltage protection is down here, so you can set all each-- maximum output voltage for each independent channel,. it'll switch it off if it goes over, like if the power supply fails,. fantastic, you know,, protection stuff you don't normally get on low-end supplies. We've got independent, parallel and serial mode here,.

auto is. it doesn't make sense, right,? What's auto on a power supply.? It's actually automatic sequence, so you can set up a sequence of events in this thing, and then you can run it by pressing the auto button,. you can get into the comms, it's got recall and preset stuff, so various front panel settings, you can recall and preset those. This Mysterious button called FC down here actually sets the digit, believe it or not,, sets the digit that you're using when you adjust the voltages here with the knob and stuff like that.

Step is just,. it actually just displays what your stepping at, you can't actually set it. So, you know, I Don't know why they dedicated a key to that, it's bloody ridiculous. You've got your output on/off button which is combined as we'll take a look at.

You can actually switch the beep off if you find that beep really annoying. You can go boom like that, and it shouldn't, hopefully, beep at you anymore. It doesn't. Silent.

Ahh. And W resets to factory defaults. You can adjust the contrast, and you've got some cursor keys to do stuff here. A Lot of functionality in this thing, but convoluted layout.

Now When I first got this thing, it took me ages, because of course I didn't read the manual of course,, you know. It's a power supply, you should just be able to walk up and use the damn thing. And It took me ages to figure out how to adjust voltages with the knob and set the on off-- set the outputs on and off and independent like that and, well,, let me show you so you don't have to go through the same grief. You Can read the manual, the manual is okay-ish, but, you know.

Uhh, well, it could be a lot better, that's for sure. And The user interface could be a hell of a lot better, that's for sure. Now If you want to adjust the voltage, at the moment, all cha-- none of the channels are selected, so if we press VSET over here , it, you know, we can't set the voltage, we can't set the current, we can't type in values, we can't use the knob. So It's almost like an automatic key lock kind of thing..
Which is actually quite a good feature,. you can't just bu-- accidentally bump it and, you know, adjust your outputs accidently and blow up your circuit. That's pretty good. Let's say you want to adjust the voltage and current on Channel 1.

You've actually gotta press shift, Channel 1 here, and Channel 1 is now selected. So Now, no,, you can't just go in and use your knob like that, you've actually got to go. VSET And you still can't use your knob like that, folks! Hate Not being able to use your knob, it's ridiculous! Took me ages to figure out you've got to actually press it, and then you're in knob mode like that,. And yes,, the output does change live even though you just-- you press enter to sort of lock it in and go out of that mode, but it does actually change live and I'll show you that..

So VSET There it is.. Jump In and you can adjust the least significant digit,. as I said this, F/C button here allows you to jump over a digit like that,. And like that, you can't jump to the 10 volt digit though, it doesn't let you do that, presumably because it's too big a step.

and ooh,, you might blow up your circuit because you don't know what you're doing. Dummy Mode. Anyway, you know.. As Long as you-- once you know it's not too bad, you know,, you can get used to that.

Now As I said, you can't just push the buttons like this, it beeps at you,. if you want to type in a voltage, you've got to set 10 point one. two volts enter, and it jumps to 10.12. Not A problem.

And It's the same deal with the current as well,. you go into current like that, you press that, and you can adjust your knob. And Then you can go up in your settings like that. So It's not too bad.

Or you can just type in the value and . By The way,: really annoying, like you can't just go point two three. Right, it automatically jumps to two, like that. So You've actually got to go, same with the voltage.

You've got to go naught point two three, geeze. Thank you very much, okay. Hopeless. You'd think it would just assume the zero.

It's just little firmware annoying issues like that which just ruin the experience on this thing. And Look what happens if we actually go into V here, and we go into knob mode. Oh, it didn't recognise that press, and we're goin-- jumping around like that, and we just start typing in a value, look,, it's just put in 50 over there. You know, five point zero zero, so you can type on top, yet it's left all the garbage underneath.

I mean, it's just look,, it didn't even recognise that! It Didn't even recognise that at all. It's just, why let you even do that?! It's just, it's just madness, it's pure madness. You know, ten point one, eh, two, look, enter, like, argh, now it accepted it, what?! Aww man. And Yes,, I'll state what everyone's thinking.
It's typical of these cheap Chinese designed units. Nothing against China, it's just. you know,, they can produce good products if they really want to. But These companies, they, you know,, get it working, and they don't put any thought into the user interface at all..

And They build them down to a price. They, you know, functionality-wise, spec-wise it's probably quite good value for money, and it's probably reasonably designed and built inside.. But They always lack the user interface, they're just awful. So Really,, you know,, they could have dumped the recall key, the independent key, the auto key, the overcurrent protection key, the overvoltage protection set, this F/C button, and you know,, put those as shift functions somewhere else, and had more usable dedicated keys on the thing.

And I'll show you that this knob does actually adjust the output live because you may not believe it unless you actually try it. It doesn't, like, tell you stuff like that, you know,, in the manual, you've just got to sort of try it for yourself. So Channel One selected, let's go in, umm, voltage select Boom, and we can see it. There We go.

We Can see it jump up there on the meter. And Of course it's bang-on, as we'll check, but yeah. There You go, it does update live and the enter is just to get you out of that mode, and effectively go back into sort of key lock mode. Let's just try the parallel and serial mode, I've got it set to Ohms here, as you can see, totally isolated between all the channels, you know, not a problem at all.

So If we go into-- it's not too bad that you've got just that dedicated independent key. Just To go back in and, you know,, ahh,, just give me all independent channels, but on its own, you know it's a wasted key, they could have used it for something else. But Anyway,, let's go into independent mode here, and you think you'd be able to use the buttons, right? You Think you'd be able to use these cursor keys here. No, you can't.

Right? Like, stupid. Like, crazy. And You've got to-- and the knob only goes one way, right.? It just goes, you know, I-- I... I Give up.

This Thing's hopeless. Anyway, enter, let's go parallel, okay? So These two paral-- by parallel and mode-- parallel and serial mode, only operate on channel one and channel two. And It doesn't tell you that. So If we switch our parallel mode here,, it should parallel these two channels up.

This One is always going to be an independent-- this zero to six volt at 3 amp, always going to be an indepdendent these two. So It only operate on channel 1 and Channel 2, would have been nice to tell you, But anyway. Parallel, we should see them parallel up. And They haven't! Why haven't they paralleled up? We're in auto mode, No,, hang on.
Parallel out. Set. What The hell? There You go,, you've got to actually go out of bloody auto mode here and you've got to go into independent mode, and then let's try that one more time for the dummies. Parallel, boom, It's switched the two outputs in parallel.

And If we try the series mode, we'll find that it actually shorts the positive and negative here as you traditionally would do, so that it can give you a positive and negative common ground output supply for Op-amps and stuff like that,. Really handy. So We'll go down here, we'll go series and Bingo, there it goes.. It Just shorted those two pins.

So That parallel/serial mode makes up for the fact that you don't actually get one of those traditional shorting bars and the fact that the positive and negative terminals aren't next to each other like that, because that's how you traditionally do it before,. you slide a shorting bar across and you can short out the two. This One does it with relays inside, nice and handy. I Like it..

And They've got two earth terminals, which they've marked ground. Yeah, whatever you want to call it. So, you know,, it's handy to have two-- I Mean,, obviously you could strap, say,, that one, if you had them in series mode, then you could strap the two like that,. or you could strap them side-by-side, They don't give you a shorting bar to actually do that, but most of the time you're going to use these things as independent floating outputs.

And Yes,, they are standard 19 millimetre spacing. Not That handy on a power supply,, but eh,, they've done that. And they've even done that for the ground as well,. they've just standardized right across the front of that on the spacing.

Now As for the binding post input,, you saw that they do have standard banana plug input, and at first glance they do look like reasonable quality binding posts, if a little small of course,, but you get that with a standard 19 millimetre spacing. But If you open them up, yeah, they do come all the way off, great. So You can put, you know,, ring lugs over them and stuff like that. And They do have a hole in there to put your wire through like that.

But One thing I don't like is that they don't have the full metal base on them, or on the back of the binding post like this. It's just a little small thing, but yeah, they're not that-- they're not the highest quality binding posts I've seen, but they're reasonable for the price. It Could have been worse,. at least they do have the hole that you can put your wire through anyway.

Now Of course,, what you pay the big money for on this thing is not only the PC programmability, but the precision of the thing. As I Said,, the specs in this thing are very impressive. Let me point them out. You Can have a look at the datasheet for yourself,, but the voltage set accuracy, specified at 25 degrees, is 0.05% plus ten millivolts, or plus 20 millivolts when the rated output voltage is greater than 36 volts? Huh? That Doesn't sound right, it can't go to 36 volts anyway!.
Crazy. Ugh, these Chenglish specs, I Don't know. But Anyway,, the current is also less than or equal to 0.01% plus five milliamps, or plus 10 milliamps when the rated current is greater than three amps.. And Of course, Ripple's pretty good, it's less than one millivolt RMS over the 20 megahertz range for voltage and less than 3 milliamps RMS over that same 20 megahertz range for the current.

As Well,, temperature coefficient: 100 ppm plus three millivolts. Here We go, I May as well show you here on the sheet. Readback Resolution: 10 millivolts, as you saw resolution on the display, and one milliamp current resolution there. Two milliamps if the rated current is greater than three, so maybe it switches range there-- switches, current measurement range, or something there.

The Response time from 10 to 90 percent less than 100 milliseconds, for rise and fall is less than 10 milliseconds if it's greater than 10% rated load. And Temperature readback coefficient less than 100ppm. So, You know, you're talking about a serious bit of kit when it's 0.05%. And That's what you're paying for.

You're paying for the ability when you set this thing to 10.00 volts, that's what you damn well want, and that's what you expect, so let's check it out.. The Fluke Multimeters are similar order, of course. I Could get out my higher-precision bench meter, but let's not gild the lily here,, shall we.? Channel 1, let's switch it on. In Fact,, let's-- oh, I Switched all of them on.

Look, you can accidentally do that sort of thing with this. If You're not careful, I Just switched all 3 channels on at once, because I didn't deselect it. By The way,, if you want to deselect? It's not obvious. We're selected There we go,.

we selected Channel 1 there, so now we only operate on channel-- no,, no, look, it just switched them all off! What The Frig? Are You kidding me? It Just switch-- I Had channel 1 selected, you saw it there, and the output should just work on channel 1. But No, look, it switched them all on. What The hell is going on here? What, because it's in series mode , maybe? Sorry, let's just hit that ma-- magic independent button there. Let's try it again.

There We go, now it seems to be working. Let's try that series/parallel mode again. This-- that's a bad bug if that's the case because it switched on Channel 3. I Can understand switching on 1 and 2, it has to, but to switch on 3? that's got to be a bug.

So Let's tu-- Okay,, series mode, Okay, so we're in series mode now, and let's turn on Channel 1. Output on/off. There We go, it's a bug. So When you're in series mode, and presumably it'd do that in parallel mode as well,, it switches on Channel 3.
That is ridiculous folks.. So Let's select Channel 3. Nope. Unbelievable! And That's the thing,.

like, ever since I started using this thing, I'm always finding something new and stupid and something wrong with the firmware, that doesn't operate the way you expect, or it's buggy, or just, you know,, beeps at you and does nothing because you're in the wrong stupid mode or something like that. It's just really incredibly annoying. Ah, God. Anyway, let's switch them all on.

And Of course,, here you go, there we go.. We're basically bang-on 10 volts there. So The good thing is, its spec is actually better than the least-significant digit there, so when you see 10.00 there,, you can be pretty darn sure you're going to get 10.00 out of this thing. So Let's go into Channel 1 and set the voltage.

One Volt. No, look, it didn't even do that. Unb-- look. One Volt.

Thank you. Look, it assumes that it's 10 volts. It's stupid, why do I have to go 1.0-- look, what?! 1.00 volts. You've got to be shitting me.

Nope, it's not letting me set-- change this voltage on Channel 1 when we're in series mode. It's not letting me. I should be able to change Channel 1 or Channel 2, and they should-- it sh-- they should track. They should just track each other.

It's not letting me change this. Why?! Unbelievable. Let's go to Channel 2, right? I'm selecting Channel 2 now. And Let's go.

Voltage Set. One Volt. It's not letting me set it. You Kidding me? Channel 3, can I change the voltage on channel-- on Channel 3? Yeah, I can change the voltage on Channel 3, but I can't change it on channel 1, or 2 because they're in series mode.

How The hell do you change the voltage, I haven't done this before. Bloody Hell. Now If you read the bloody Manual, it says that Channel 2 is the master. There We go, in series and parallel mode.

Channel 2 is the master, Channel 1 output automatically follows Channel 2 output. Fine. Let's go to Channel 2. Channel 2 There we go.

We should be able to VSET, push, There we go. We Can now adjust it. But You saw it before, it didn't bloody well do it. And Look,, it's not automatically changing the digits on Channel 1 there.

So Even though it knows it's in series mode,, it's not going to change that. I Mean, is that a feature or a bug? You Decide. Alright, 10.25 and look,, right,, it hasn't changed it! In Series mode, it's not a live output like it was-- is it? Yeah, look folks. We're in independent mod-- sorry, series mode, okay? These are supposed to be tracking.

It is not live tracking. Okay? Look It's live tracking on Channel 2, which is the master, but it doesn't automatically live track Channel 1. That is ridiculous, look. If We switch it over if we press enter.

I bet you it tracks it. No, no, it didn't track it. It's supposed to be series mode. shitting me.
And I Just used the keypad to do it, and it switched. Watch This, right? I'm on Channel 2, Okay, so channel 2's selected, I Voltage set 10 volts , not a problem. It changes and it tracks, right? But If you use the knob, it doesn't let you do it! It allows yo-- it sets them independently. And-- but-- that's got to be a bug.

Because It's not a feature, because you can't change Channel 1 unless you go back into independent mode, change it first,, go back in, and la la la. It's just-- it's stupid! It's retarded! Alright, let's try the same thing in parallel mode, and of course,, there we go, we're bang-on 10 volts there, alright. On Both, because they're actually tied together, they're shorted by the relays in parallel. And They would load-share as well, I'm sure.

You Know,, there's internal load-balancing resistor in there and it balances out. Not a problem. So Let's go to Channel 2,, which is our master channel. Nice If you, know, know, actually-- you know, look,, you wouldn't be able to just walk up to this thing on the bench and start using it if you hadn't seen it before.

It'd jus-- you'd just toss it in the bin, because it's just-- you wouldn't be able to figure out how to use it. It'd take you half an hour to figure out how to get the damn thing working if you didn't have the manual. Unbelievable! So Anyway,, let's get in there, set up-- what do we want? VSET, right? Let's go 20 volts. There we go.

Of Course, it's not a problem at all,. it's working fine there, because it's parallel mode. So that independent mode-- let's try, hang on. No,, that's the keypad.

sorry. Let's do this. There We go, it's following. Okay, it's live following now.

But It's just reading back, so of course it's going to be accurate when it reads back. There's a bit of lag there on the-- on that channel. 20.59. Okay, seems to be working.

So Let's check out the accuracy difference in parallel mode as opposed to independent mode. I've set it for one volt there, and we're 0.9963 I Mean,, we're getting more down into the 0.05% plus 10 millivolts down in there. But Let's now switch that to independent. Annoying.

Channel 1, voltage set, 1 Enter. One volt, 0.9964. Okay, so it's the same. Channel 3 is within spec, there we go, 0.88.

It's just within the spec there. Channel 2. Abso-- absolutely tracks Channel 1, we're in independent mode here too, by the way,. and yes, I have double checked this with a second meter and it's spot-on of course, I Do keep my meters within Cal here.

Up At 30 volts there,, yes, we are within spec.. Not A problem. Now One thing that does work well on this is, is that the fact that when you're actually have the output on, it displays the actual output current. But Of course, when you switch all those outputs-- well,, if you switch the one output or all of them off, any combination, then it displays the set current, 1.5 amps.
So Let's check the accuracy of this. So Let's connect it and you should-- to the current mode, short out the output and you should hear the fan go up . Relay Switching there. There We go, 0.9993.

It's display 1.001 and it's set to 1.000. So that is within the current spec of 0.1% plus 5 milliamps, not a problem. And The second channel there, yep. Not A drama.

And Let's try the third channel. There We go. Yep, all within spec.. And We can get it to spark here, folks.

Because We're up at 30 volts. Yeah, now we're talking. Let's just abuse the hell out of this sucker. See if it survives.

Woohoo! Not A problem. And There we are, we're bang-on. Well, on the meter, and a little bit off on the displayed current there,, but certainly well within published specs at 5 amp current there. I've switched it down to 1 volt so we're not getting a huge massive spark when we do that.

There We go, 5 amps. And Channel 3 should be 3 amps. Maximum. 2.999, 3.016, all well within spec..

Nice. And If we try it right down at 10 milliamps current,, there it is.. It's drawing 10 milliamps there, I've got it set to constant current, and of course,, you know,, it's going to be near enough. There We go, that's certainly well within spec right down at that.

And The display is spot on to the least-significant digit. This one's four least significant digits out. But There you go,, it's drawing close enough to 10 milliamps. So, eh,, that's not bad, I mean you're right down in the noise and-- right down in the resolution and the noise of the spec, so to speak.

Right down at that bottom end of the range. But Certainly not a problem. Now What I'm going to do here: I'm going to measure the noise performance of this thing, and I've got my Agilent scope here, bandwidth limited, by the way, to 20 megahertz, because that's the spec. It's very important, the spec says less than one millivolt RMS over the range.

There It is, 20 hertz to 20 megahertz, okay? So It's important that you put that bandwidth limit in. If You don't, you'll see the noise increase there. The Peak-to-peak noise. So You've got to have that bandwidth limit on.

20 megahertz, the RM-- we're looking for less than 1 millivolt RMS, and RMS is the key. The spec is not peak-to-peak noise, it's RMS noise. And There it is,. it's less than 1 millivolt.

So It's doing the business. But Look at this horrible output here!. Look at this switching! We're at 5 microseconds per division, 3 divisions there, and we're looking at, you know,, around about 66 kilohertz switching frequency there. What is the issue here? Well, I'm not actually going to tell you! I'm going to leave that for a quick little separate video coming up.

Okay? So That'll be a little aside video. I Won't explain it here. But This noise is not coming from this power supply. And Here we go,: if we have a look at our noise performance here,, we'll use our Rigol.
We're on 1 millivolt per division there, and our RMS, like the RMS figure, is the one we're actually looking at there, 700 microvolts or thereabouts. This is only for an 8 megahertz bandwidth, the probe is on times 1. Let's try on times ten. And There we go,.

we can only go down to 5 millivolts per division on the times 10, but you know, we're down in the millivolt region, I Mean,, you know. I'm not going to muck around any further, eh.. It's good enough. That's at no load by the way,, that's on the channel.

3 output at 6 volts. And It's about the same there on channel 1 on the 30 volt range. And Just for kicks, we'll try this on the Agilent as well. There are the RMS noise figures there for a times 1 probe there with an 8 megahertz bandwidth.

So, eh, not too bad. And Here it is with a 30 volt 1 amp load. It's not switched on yet, so that's no load. And Now load.

There we go, not a problem. So We are drawing 1 amp down there, 30 volts. Yeah, it is basically bang-on to our power supply there. And Our noise doesn't really change.

Let's bump it all the way up. And Our noise doesn't increase at all at 30 volts at 4 amps. But I Set that 5 amps! Why Is it 4? Something is going on here folks. We have been diddled.

I Have got this set-- the current limit is set to five amps on this channel here, and you can see it if I switch it off. There it is, 5 amps. But It just ain't-- well, now it's delivering 5 amps, but it's gone into constant current mode. So Maybe we're a smidgen over.

Let's tweak that down. There We go, we're going up. And Now it's delivering-- it's trying to deliver 5 volts at 4 amps-- 5 amps. Sorry, 30 volts at 5 amps.

But, it's just there's something happening there. So Anyway,, I Am not sure what's going on there, but let's check the 5 amp ripple. And There we go, it's bugger all really. So Not a problem in constant current mode, but this thing has gone into constant current, so it can't deliver-- I don't-- I Can't see how it can deliver.

its rated 30 volts, 5 amps. And that's only with one channel connected. Let's try channel 2. Nope, we seem to be getting the same performance-- oop, there we go.

There we go, 4.9. No, see, it just dropped down to 4. What The hell is going on here?! I'm setting this to 5 amp constant current and it does not like that at all. There we go, I Just switched it down to 20 volts and I heard the relay inside click, so it obviously went to another transformer tap, and not a problem.

It's able to deliver the 5 amps. Not A problem at 20 volts. But It certainly can't do it at 30. So There's a limitation-- there's performance limitations here, which they're not telling you in the specs..

I Mean, there's no,, you know,, power load curves on this thing or anything.. No Power performance curves, really.? So, yeah,, don't take it at face value. And I've got Channel 3 hooked up and I'm trying to deliver 6 volts at 3 amps, but it's not having a bar of that. There it is, 6 volts, 3 amps, but it can't do that.
Let's see if it can do it at 5 volts. Hmmm. Nope. If We have a look at our waveform over here, that's at 3-- that's at 3 amps, constant current mode.

And that's 50 millivolts per division. So Yeah,, that's at 3 amps. Let's try 2.9 amps. There We go, not a problem.

So It really craps out if you try and get right near the limits. Right, so I've got my current up here set to, say, 2.9 amps. Okay? So It's trying to draw-- it is drawing 2.9 amps, and it is doing it at 6 volts. So It is able to do that.

And My current's set down here-- oops., 3 amps current limit down there. So Yeah,, it's able to do very close to its limit there. And Of course,, there appears to be a bit of discrepancy here,. 5.91 volts, I mean this is a precision load this BK Precision 8500.

It's 0.05% class as well. But It's drop-- that's just the leads I've got here. The Display itself is actually measuring right on the terminals, and if we-- we can confirm that with the meter over here. There it is, 5.997.

So That's just the drop on the leads. No Worries. Alright, now let's test it's switch-on performance. Going On Channel 3 6 volts, 3 amps.

See if it overshoots at all. I've got the scope set up to single-shot capture, so we'll switch it on. Boom, look at that transition. Ramps Perfect linear ramp.

No overshoot at all, very clean. Nice. And That's with a full load too, by the way.. So Let's try it again with no load.

So We'll switch our load off here. Let's reset this . Switch our output off, switch our output on. And Boom.

Look at that. There We go, with no load. it's much quicker. of course, I didn't change the timebase there at all.

And ramps up, there's no overshoot in there. Aww Look,, there's a little-- there you go,. you didn't see that when you're-- There you go, you zoom in,. there's a bit of funny business happening there.

That's with no load, so you know, that might cause an issue with your system if that was a problem. So, there you go,. That's noteworthy, definitely.. So Let's try, say,, an amp..

So We'll set this to 1 amp, and is it? No. I Set 1 amp. Boom. Okay, now we're set for an amp.

Let's try that again. Output. The Output is on. Alright.

And There you go,. that's for a 1 amp constant current load, by the way,, which isn't the best. We'll do it again with a-- we'll do it again with a resistive-- pure resistive load to see if that makes a difference. But Of course, the no-load had that funny business, so even with the constant current load, it's, you know,, it's pretty perfect.

Okay, let's try that again.. I've got a pure resistive load here. It's not a wire-- wirewound resistor. It's a 10 ohm load.
So At 6 volts here,, we're only going to get 0.6 amps. But Let's give that a shot. Shall We? And See what we get. There We go,.

we're getting that same-- getting that same business there happening at 0.6 amps with-- at 6 volts. But Of course. The most important thing is, there's no overshoot-- virtually no overshoot there.. I Mean,, there's a little bit, a little half a bee's dick of overshoot there, with a little bit of recovery.

But, yeah,, that's, yep,, that passes. Alright, let's try the same thing again.. Got Our 10 amp resistive load, 20 volts output on Channel 1 5 amp current limit, so it's only going to draw 40 volts, it'll only give us 2 amps on the output. But Let's ramp that up.

And, well, we'll have to change that to 5 volts per division. Okay, we'll give it a go, single shot mode, output off, and output on.. Bingo, look at that. Nice and clean.

I Like that. Not A problem. That's a switch-on for a 40 watt load. That's at 2 milliseconds per division, so that's taking-- 1 millisecond, so that's taking about 2 milliseconds to switch on there, for a-- 5, 10,, 15, 20 volts of course, for 40 watts at 2 amps output.

Nice. Okay, let's try the switch-off, shall we? Let's give it a go. Output-- let's move our trigger up a bit. There we go.

Our output--hey! There we go. Let's try that one more time for the dummies. We're on. And There we go, clean switch-off.

Not A problem whatsoever. Now I Know what you're thinking, this BK Precision Electronic Load has a bit of a reputation as a power supply killer, so let's muck around with the constant power and constant resistance modes.. Here we go, I've got constant power mode set to 90 watts there. I've got 20 volts output, so it should give us just under the 5 amp current limit.

So Let's switch that on. No worries, 4.5 amps. Not a problem. Let's see what the switch-on performance is.

And There we go, that's our load switch-on. We're at 10 millivolts per division there. 40-- no,, sorry, 90 watts constant power mode,. it only drooped about 15 millivolts or thereabouts, and recovered pretty quickly.

Nice. And Let's do that again, and switch off. Boom. There we go, not bad at all..

And Let's do the same thing again, 20 volts, but with my 10 amp resistive load. And bang. Put it in there, yeah, we've got some bounce there, that's my contact bounce. But Not a problem.

And I've got constant resistance mode set to 4.1 ohms and it's, you know,, handling the switch-on of that. No problems whatsoever. If I Capture that on the scope, there it is.. Not A problem.

Let's go. Really nasty and constant resistance mode, 1 Ohm. So It's definitely, you know, it's not going to try and draw-- well,, it's going to try and draw 20 amps, but it's going to be limited to-- current limited to the 5 amps there. Whoa! Look at that, it's oscillating! Woohoo! Will it kill it? Will it kill it? I Love it! Listen to that! How long should we leave that going for? And Of course, it won't do that.
if you simple short the output, it'll just go into 5 amp current limit. So That's a unique quirk of the constant resistance mode when the thing's trying to hunt to do the constant resistance. That Load is quickly cycling, and the power supply is going "ooh, I Don't know what to do, don't know what to do! Quick, switch relays, switch relays, panic, panic!" But There you go. To Its credit, it survived that and it's surviving.

Just shorts on the output as well. 5 amps constant current. So It lives to power another day. Not A problem.

It's not like that Korad rubbish. There We go, I've got it on constant resistance mode, 1 Ohm on the channel 3 this time. 6 volts, there's no relays clicking It's not changing transformer taps. So It's-- there we go.

It's just jumping in and out of that 3 amp. It doesn't know what to do. 6 volts, 0.1, wah. And You can hear the fan just revving up and revving down each time it does it.

Alright, now what I want to do is capture it. jumping a transformer tap. So I've got it on 5 volts at the moment, so there it is, 5 volts per division, and we'll jump up to say 20 volts, and we'll try and capture that. So The trigger point is just above 5 volts.

So We'll go voltage set, and we'll go 20 , and see what happens when it jumps. Nice linear ramp up there, constant current mode. And No overshoot, nothing when it changes taps. Very nice.

And We'll capture that again with a 10 ohm resistive load. So We'll go from 5 up to 20. Beautiful, look at that. Hey, little bit of a-- little bit of funny business happening there.

But, yeah, it doesn't overshoot, that's the main thing. Let's try that again. Just up to 30 volts. See What happens.

Woohoo! Look at that. But It does actually recover. There We go. Two step process as it goes through, maybe, a couple of taps there.

Through Two taps, and then up to the final 30 volt.. So that was with a 10 ohm load, so at 30 volts, that's jumping up to a 90 watt load. So Let's try that again, but with no load, shall we? So We'll go back down to 5 volts. And We'll jump up to 30 volts.

Bang. Look at that. Not A problem whatsoever. And What happens when we go back to 20 volts under load? Well, let's set our falling edge trigger there, and we're at 20 volts and we've got our 10 ohm resistive load on there, drawing 2 amps.

And Let's go down to our 5 volts and see what happens. There We go. Nice Ramp down. Beautiful.

Think I'm getting the hang of using this thing. There We go, look at that. Alright, let's see what happens when we go from no load into constant current mode. So Let's capture that.
Whoa, look at that. There We go. Let's try that again. Boom, look at that.

So We had some bounce there. So Let's look at that. Yeah, that's probably some bounce from our contacts there. It's jumped down-- that's 2 volts per division, so it's jumped down in voltage of course, and then ramped back up.

Let's try that again. but in DC coupled mode, 1 volt per division so we can actually see the real drop in voltage. And Here we go. Boom.

No, . And Let's see how accurate the constant current mode is.. Let's-- our constant current is set for 3 amps, okay? We're currently drawing 2.995, 2.996, and of course the voltage hasn't dropped. And Let's see if we get very close to that 3 amps.

It should switch into constant current mode. Yep-- hang on. Yep, there we go. Bang.

Right on target. Now I've had this thing going on a decent load for about thirty-odd minutes now. I've got 40 watts load on Channel 1, I've got a hundred and twenty watts load on Channel 2,, and I've got a 18 watt load on Channel 3 here. And I've got my temperature probe, and the temperature around back, it's not-- coming out of the fan, I mean, the fan is going full bore and we're only getting, you know,, 37, 38 degrees or thereabouts out of the fan.

So The thing really isn't getting that hot at all. Now What we're going to do is try and find where the tap voltage on-- for the 30 volts is. So I'm ramping down 30 volts, once again. I've got that full load happening on there.

So, let's wait for the relay click, and... Oop, there we go.. So About 23 volts. Let's go up.

Yep, 23.2 volts is where that tap kicks in. And We should see that temperature go up, and we do because it's now dissipating more, being a linear regulator, at 24.4 volts. of course, it's dissipating more power in the heat sinks. There We go, we're over 40 now, and I've just had it running for a minute.

So that's going to climb up. So It-- the temperature of the heatsinks and its final capacity before any sort of thermal cut off comes in,. if it does actually have that, then we should get that-- you would have to stress it at the lowest point of each tap range. So It's dissipating the most amount of voltage drop-- well,, it's got the biggest voltage drop across the pass transistor..

So There you go, but it did actually go up in temperature as expected. And The next tap here seems to be about-- just under 15. Yep, 14.8 volts there, and jumps back up at 15.2. And the bottom tap, about 7.6 volts.

By The way,: there's one other thing that's noteworthy is that after all these load tests,, it's still within spec on its output, voltage and output current. So Yeah,, drift doesn't seem to be a problem. Now, yet another confusing thing on this is the store/recall mode.. Now, you might think that that stores and recalls voltage and current settings for the 3 channels, but it's not necessarily the case, it has to do with this auto programming mode.
And It's the manual really doesn't explain it. It's ridiculous. Now What it actually is, if you go into store mode here, then it gives you these various channels. Oh, we can't use the cursor keys there, we've just got to scroll through, Okay.

And It has to do with Channel 1, Channel 2, and Channel 3. So the different settings-- So let's go, say we want the auto-sequence-- set up the auto-sequence for Channel 1 output. Let's go into that, Channel 1, and then it has these various storage options. Whoa! We Can finally use our cursor keys.

Look at this folks. We Can scroll around and yes,, it's got up to I Think 15, no,, what is it.? Heaps. 30 Uh, program steps, effectively, per channel. Go up here, and this is the voltage and the current, and then the delay time in seconds.

So We can set our voltage. You Know,, we can-- I've already set it there, 1. And Then we can set it to 1 amp there. For 1 second.

So We want that to stay in that mode for 1 second and then sequence through to the next one. And If you scroll through the list here, and you go and highlight, say the third one and you press auto, then that puts that little cursor down there and I'm presuming it will stop when it gets to that point in the program sequence. So It'll go through those two steps. But It doesn't tell you! I'm going to set-- this is Channel 1 only.

So We're going to cycle through 1 volt at 1 amp -- not that the current limit matters, we won't be using it -- but you can set it up anyway, for one second. So Then it'll jump to 2 volts for one sec, and 3 volts for one second, and should stop when it gets to the end of that point. So It should cycle through all 3 of those. Let's see if we can capture it.

Alright, so I've stored our program There It is,, it's in there. Let's get out of that, and-- no,, you can't recall it, okay. What You've got to do is you just-- all you do is you run it, Okay? So You put it into auto mode. Like This,, there we go..

And Then you press the output button, and all three channels will turn on,. you don't have to individually turn them on, stuff like that. So You press it, and we'll now find that channel 1 is cycling 1, 2,, 3 volts. You Can't see the set current because the Se-- this is the display current, but bingo, there you go.

And if you capture that on the scope. Tada, folks. There it is,. that's what we have.

There We go, that's 1 volt per division, so it goes. 1 volt, for a second -- let's horizontal position that -- there we go, it's reasonably accurate. So It jumps up to 1-- 1 volt, 2 volts, 3 volts, and then that ramp down is because we've got no load on there, so it doesn't ramp down very quick at all,. that's the output storage cap.

But There you go, it is cycling through, and it's continuously cycling through. And That's a good use for having this oscilloscope on roll mode. There You go, you can actually see it. That's 1 second per division now, and you can see it cycling through.
And It's actually doing that in real-time. So It works, huh!? There You go.. That could actually be quite useful. There We go, I've actually plugged on a 10 ohm load onto that, so it's drawing 100 milliamps, 200 milliamps, 300 milliamps.

As You can see now, when it falls back down, it falls back down very quickly. There We go, and boom. And Now to set up a sequence on all 3 channel, Woohoo!! I'm an advanced programmer. Channel 1 is counting up 1, 2, 3 volts, channel 2 is counting up 4, 5, 6 volts, all with 1 second intervals, but you can change the intervals to anything you like with 1 second resolution.

So Channel 3 We've got that cycling 3, 2, 1, so that one's cycling downwards. So There you go, That's actually a pretty useful feature. And You can program it, be it a bit archaically, through the front panel user interface. So Once you know, not bad at all..

But Try and figure it out on your own. Ugh. And For the life of me. I Can not find a use for the actual recall button.

Yeah, shift-store Not a problem,, but the actual recall button. Whether it's in auto mode, doing something,, whether it's in out of auto mode, it's just I Don't know. I Give up, and the manual doesn't help at all. And It really is a shame that it doesn't let you store and recall popular settings for the front panel! I Mean,, for the channels, you know.

Yeah This auto sequence mode is great, but geeze,, where's some just basic functionality to, you know,, preset some voltages? Now One of the good features is overvoltage protection set. So You press that, and you can set up the max-- absolute maximum voltage for any 1 of your channels. So Let's say you had a real critical device under test, real critical board on here,. it's, you know,, 3.3 volt rail.

and if you went over that, you're going to blow the ass out of your, you know,, your $10000 prototype board or something. You Don't want that to happen, could ruin your 10 million dollar project-- the schedule for your 10 million dollar project, Aww! Right? And You'd probably be-- get the boot straight out the door. So You don't want that to happen. So You can go in here and you can set, you know, I Don't want that thing to be over 3.4 volts.

And that is separate protection circuitry, which ensures that there's no way that output can go over 3.4 volts. Or Presumably there's extra circuitry in there that allows that, if the power supply fails. It's supposed to be an independent-- the whole idea is it's an independent thing that which kicks in. But may or may not be, probably just done in software, really..

But Still,, it doesn't-- it means that you can't do anything stupid to the front panel. Let's say, we don't want that channel 2 to go over 3 volts. I've already got that auto sequence programming, let's see if there's a bug in there, alright? So Overvoltage protection on Channel 2. Yeah, there we go.
Notice: overvoltage protection on Channel 2, please push shift overvoltage protection reset key to clear overvoltage protection status and reset. So There you go. We Have to clear that. Oop.

There We go, we've cleared it and Channel 2. It's got 6 volts there, but let's see if we can switch on the output and actually get 6 volts. Alright, let's press the output on. And Nope,, it protects it.

Not A problem and I'm checking on the scope and no,, it limited it and it didn't jump up at all to 6 volts. Brilliant. But One really annoying feature is you can't just turn on channel 1 like that, and have overvoltage protection set on another channel, because it just won't let you turn on any of the channels. It's stupid! Protect That channel, great, but let me do whatever I want with the other channels, please! Now What OCP does, or overcurrent protection here does, is that instead of going into current limiting mode,, if you hit the current limit, it'll just switch off the output.

So Let's give that a go, right?? It's-- OCP is off at the moment, right? So it-- I've got channel 1 set to 3 volts at half an amp, it'll actually attempt to draw more than that. So Let's go, output like that. There We go, it's gone into current limiting mode, okay? So It's current limited it to 500 milliamps, but it's just continuously, like a normal power supply, continuously current limiting. But If you don't want that,, if it excee-- if you want to shut off when it exceeds that current limit,, easy..

Turn on OCP. Bingo, it just turned Channel 1 off. Because We were overcurrent there. So If we switch all of the channels-- See? I switch it on-- I Briefly, try and switch it on, it goes overcurrent and just switches the channel off.

That Can be very useful depending on your circumstances. And I Tried to get the USB serial interface working on this thing, yes, it is just a USB serial converter. It Rec-- I insta-- you know, I-- It recognised it when I plugged it in and installed the FTDI driver and stuff like that,. But the software, well,, let's just say it's as shit as the user interface on this thing.

Couldn't get it to work, couldn't figure out how to set the parameters to-- wouldn't connect to it. Complete Fail. Not Going to waste any more time on it. And It does actually come with a document which lists all of the serial-- serial protocols to talk to this thing and control it.

So I Guess the idea,, write your own software if you want to do that. I'm not spending another second on it. So The verdict on the Atten PPS3205T 3 channel Precision Power Supply? Well, eh, probably a thumbs sideways. Not A thumbs down because, well,, you know, I couldn't kill it, it does represent good value for money.
I think. And, Well,, It ultimately does work, even though it's a pain in the ass to drive the thing. So Because of that, the user interface stuff, can not give it a thumbs up at all.. But It does represent reasonable value for a precision 0.05% class.

It met it's specs. Didn't quite meet the performance-- the power performance on the highest tap on the transformer, but, you know,, hey. Eh, it still work reasonably well. It's quirky.

I Would actually buy some of these for the lab as a joke, right? Stick them in the lab and everyone will come up to it-- all the engineers will come up "oh look at this new funky triple output power supply!" and just sit back and giggle as you watch them try to use the thing, and just give up and toss it across the room in frustration! Because It really is frustrating to use! But, eh, once you get the hang of it I Guess it's going to be okay. I'm going to keep it here in the lab, use it, it'll probably become my main lab supply, because it is so versatile. You Know,, dual 30 volt outputs, separate 6 volts, precision. I Love the precision in it, it's programmable, and that sequence mode, pretty handy.

Might come in useful for sequencing some things, so. Without The need to actually hook it up to a PC and actually control the thing. So Not bad at all. Yeah, thumbs sideways.

I don't know. Cheap Chinese user interface. Bloody Hell. Anyway, if you want to discuss it,, jump on over to the EEVblog forum.

If You like these reviews, please give them a big thumbs up. Catch you next time! Captioned by Sen.

Avatar photo

By YTB

21 thoughts on “Eevblog #439 – atten pps3205t-3s triple output power supply review”
  1. Avataaar/Circle Created with python_avatars Martin Nicolas Mena Mena.- says:

    Hello EEVlog: I really like all your videos that you upload and publish on YouTube, very complete on your part in terms of all your explanations and clear, simple and concise steps (a clarification: to make my comment easier for you, I went to the translator from Google I wrote my comment in Spanish and I was translating it into the English language and then I copied the translated language into English and pasted it in the comment bar on YouTube and there I would send it to you in your language), I hope you have liked my comment, receive your answers as soon as possible, greetings and congratulations to you and everyone around you. From Lujan, Buenos Aires Argentina.-

  2. Avataaar/Circle Created with python_avatars David Kuqani says:

    What is the best power supply?

  3. Avataaar/Circle Created with python_avatars pilatomic says:

    Dave,

    "OVP" has nothing to do with protectiong with failures.
    Like "OCP" turns off the output when current limit is reached, "OVP" simply turns out the outputs when voltage limit is reached.
    Very handy to charge batteries (automatically cuts power when end of charge voltage is reached)

  4. Avataaar/Circle Created with python_avatars david galeski says:

    .230 mach shop

  5. Avataaar/Circle Created with python_avatars wither8 says:

    These companies should just give you like 20k AUD for a few weeks of consulting to fix up the firmware. It's a pretty precision bit of kit especially for the money but the shitty interface totally dissuades hobbyists such as myself from buying it. Hell, someone should just fix it themselves, rebadge it and sell it for +100. Edit: hahaha dave you should have left the relays in oscillation to see time to failure!!

  6. Avataaar/Circle Created with python_avatars fdp says:

    Gawd. Bit on the noisy side.

  7. Avataaar/Circle Created with python_avatars David Bonner says:

    Please do not say the R-word.

  8. Avataaar/Circle Created with python_avatars Wes Z says:

    Great review. Lots of good info and test checks. Thanks for the thoroughness. It appeared to me that the actual circuitry worked very well and responded nicely except for the limit crossover concerns. The biggest problem seemed to be the Functional Logic. Typical Asian mentality.

  9. Avataaar/Circle Created with python_avatars Kasun Nanayakkara says:

    Yes. It's one of the cheapest…

  10. Avataaar/Circle Created with python_avatars Kasun Nanayakkara says:

    Yep. As you said one of the cheapest, it really works as that…

  11. Avataaar/Circle Created with python_avatars hobbified says:

    Dave hates when he can't use his knob.

  12. Avataaar/Circle Created with python_avatars kenny gaia says:

    Awesome

  13. Avataaar/Circle Created with python_avatars m abraham says:

    what about pc control, can you make a separate review?

  14. Avataaar/Circle Created with python_avatars TheExileFox says:

    Haha, @ 41:00 – sounds like a sewing machine

  15. Avataaar/Circle Created with python_avatars BMW 3 Series says:

    who does…….for a PSU 🙂

  16. Avataaar/Circle Created with python_avatars whiskeyify says:

    I just bought an ATTEN mulitmeter, for about 25 dollars. the quality seems quite good for the money. At work I always used a fluke, but at home electronics is just a hobby for me so I didn't want to spend a lot. I was hoping Dave would take it apart, I wanted to see the insides.

  17. Avataaar/Circle Created with python_avatars gonzik says:

    New unit "multimeter widths" =)

  18. Avataaar/Circle Created with python_avatars vitalsfading says:

    "it's only 200 mm wide" ….. thats what she said

  19. Avataaar/Circle Created with python_avatars Brendan says:

    Haha, that fan sounds like a lorry!

  20. Avataaar/Circle Created with python_avatars marinedalek says:

    You know when you're trying to do something in a dream, like search the web or call the police and whatever you do it has unexpected consequences? Like dialing 999 puts you through to a takeaway or the phone crashes. That's what this power supply reminds me of, especially at 24:15

  21. Avataaar/Circle Created with python_avatars IGBeTix - Apprendre l'électronique - Expériences says:

    Funny, 1 volt ohhh no it's 10 volts !!

Leave a Reply

Your email address will not be published. Required fields are marked *