Inside the Atten PPS3205T-3S 3 output precision laboratory bench power supply.
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/

Welcome to Tear down Tuesday And yes, we're going to tear apart this Attn PPS 3205 T-3 Lovely! Name Pro Precision Programmable Power supply so no McKing around. You know what we say here on the EV blog? Don't turn it on, take it apart and here we go. We, uh, expect inside this thing a big ass. uh Transformer Of course, because this thing a couple hundred Watts worth so it's going to dominate the thing.

um, hopefully we get some uh, metal, um, you know, rods going through so it's all nice and sturdy. I Suspect not. um though, although it's a hope for this kind of cost. I Don't expect.

um, real super duper metal, shazzy and all that sort of stuff. but anyway, could all be plastic attachments up the front like it was on. um, say the co- red, uh Supply and others. but uh, we'll see now because this is a Precision uh power supply.

um, you know, 05% class uh instrument? I Um, expect there to be a decent or decent uh Dack per channel in this thing and uh, a decent ADC none of this uh, resistor base Dack rubbish we saw in the co-ed. Um, it should be quite well done. well, at least reasonable. Like using a maybe a, you know, a microchip brand 12bit deck or something like that.

So that's what I expect to find in this thing. Uh, multi- boards. There'll be multiple boards. There'll be one maybe up the back near the heat sink of course for the uh power supply.

Uh, for the main uh power supply. Maybe a shared heat sink between all the Uh channels or something like that? We'll find out. So um, but to, uh, yep, we're out there. We go.

and uh, the lid? There we go. We've got uh, no. uh Shake proof washes under there. but uh, at least it's um, you know it's reasonably solid.

and yeah, we have metal bars going across here, but they go into a plastic surround. uh, plastic. uh, front panel here. So yeah, not the sturdiest uh thing in the world.

but yeah, really? I was just a bit. I Was hoping that we'd get a bit more than just the folded front panel with the plastic. you know, metal bars into the plastic front. but eh, you know this thing's built down to a price.

This is 450 bucks worth. um of precision power supply. So let's take a look in here. I Instantly see some temperature sensors on the heat sink.

There we go. thermal, uh, switches, thermal cutouts. So there you go. It does have, um, thermal protection on both of these heat sinks.

It looks like they are separate. We'll get in there and have a better look, but uh, they do look like as a separate one. I Mean they're very close so you know it's going to be, um, a lot of heat. uh, shared between those.

but they are physically separate so it looks like aha, Here we go the third. Channel Let me check over here. Yep, there we go. There's our three power transistor.

Let's call that Channel One down in there. So Channel one has its own heat sink with its own thermal cutout. and over here Channel two. there's the three devices as well.
three power transistors we'll try and get in or Bridge rectifier and uh, Power transistors. but there looks like they've tacked on the third 5vt channel. So the third channel here this, um I mean it's only low power. It's only uh, six volts at uh, three amps.

so it's only 18. Watts But it does share the heat sink for, say Channel 2 here. So really, um, if you're using both of these at the same time, H you may not get. You know if you're loading down channel 3, you may not get full capacity on Channel 2 as you would for Channel 1 over here.

So that's just something to remember. We're going have a look at the big caps here. But yeah, look at our big beefy power transformer here. It's absolutely enormous.

That's where most of the weight goes in. We'll have a look at the sticker on there. but yeah, we've got one main baseboard down here for all the power supply. Looks like we've got our uh DAC um/ ADC board one of those for each.

Channel We've got our Um serial interface board over here. We'll take a look at all these. There's our Reay switch in. um, we've got measurement circuitry on the front.

We've got another board. Oh yeah, that's probably uh, the channel 3 uh D/ ADC board. We've got our display board up there and let's check out our Transformer it is uh, 240 volt one cuz this is an Australian version. It is labeled specifically for this particular Uh model.

It would be different for the different models of course. and um, it's going to have lots of multiple Taps Look, we've got 0 volts, 7.6 volts, 19, Vols, and 34 Vols and they would correspond to the Taps that we uh measured in the Uh review of this thing as well. So um, they're all rated for uh, each tap. uh rated for 5 amp.

so it would select those with the Uh reays and uh, so that yeah, it matches the you know, the 5 amp output capability I guess um and is going to have two of those. So going to have one for channel one, one for channel two. Then they've got an auxiliary tap of uh, plus - 172 volts here. only low power for the control circuitry 0.35 amps.

Um, so there's the second channel. It's exactly the same as that channel there, both 5 amps. Um, and it looks like they got two. Yeah, each channel has two 72, 7, 12.35 and then there's another one at 11 Vols at 3 amps and that's for your 5V uh output.

And there's another tap 11.5 Vol at 0.8 amps for more miscellaneous circuitry. Woo man, what a beast of a transformer! So let's look at the one individual Channel we've got here. Here's our um, Uh, three input: uh Taps So here's our common and we would have our 7.6 volts, our 19 volts, and our 34 volt Taps And there's the Uh two relays down in there to uh, switch between those uh, various Taps The reason that you have uh Taps like that in one of these linear power supplies is because well, if you got one at 34 Volts for your max, some output voltage capability and you got 5 amp output current, you're going to be dissipating a huge amount of power in your heat sink. That's why they can get away with making this heat sink so small for like 100 what is 150 odd Watts You know, um per channel on this thing.
That's I mean you know this is not a big enough heat sink to dissipate, you know, 150 watts. So they keep the thing uh low. Um, they way they keep it small is to have the multiple caps just to ensure and hopefully if you designed it right, the worst case, um, dissipation capability in your heat sink will be the lowest uh, the highest voltage differential across from What tap you're using to your output voltage. so at a certain current therefore Ohms law, that power has to be dissipated in the Uh past transistor down there and then down in this section we have another small Reay couple of heat sunk uh, little regular ators I'm assuming there's our other Transformer uh tap coming in there.

that plusus 17 Vols for the control circuitry. and then they've got another connector here which uh goes off to um, the digital Uh board with all the Uh control circuitry. And then we have this vertical uh Riser board here which presumably has the Uh DAC control on it for Um, setting the output Uh voltage and output current. And if we take a look at the Uh Power Trans down in there, bingo, we have a 2sd 1047 Npn Power transistor specifically designed for these applications.

There it is. This is the St data sheet, but uh, there's many manufacturers of this one assuming that they're genuine. I'm not sure who the manufacturer I think it had like a V on it or something like that, but it's specifically designed for linear power supply applications. 140 volt uh 12 amp job and if we have a look at the SEO data sheet here, interestingly, it's says check it out um on chip Bast resistance and well, that's ideally what you want on here.

When you uh, combine the multiple Uh channels, you don't have to have an external balance resistance. You can just parallel the transistors, takes care of it for you. Very nice apologies for interrupting the tear down, but I'm pissed off. You know why? Because I just spent hours shooting this bloody tear down, and then I just spent hours trying to recover the data from my pain in the ass piece of Canon Hfg10 camera somehow screwed up all the video files.

There were like 80 clips that I shot for this thing or most of them completely screwed there. only a couple that seem to actually recover properly. They wouldn't load into my bloody video editing software. They wouldn't do anything.

wasted freaking hours doing it. and now they're They're useless. So I've got to re-shoot the whole bloody thing. I only started with like what the Clips you saw before this cuz I shoot the thing in sequence.

So now I've got to re-shoot the whole bloody thing the rest of it and I'm I'm pissed off. so if I'm a bit tur or I'm not my usual self or I miss things or whatever From now on for the rest of this review, you'll know why. Breathe in, breathe out. All right, let's look at the mains wiring, shall we see if it's nice and safe? Deja Vu Here we go: Um yeah, here's the main switch real clunk and Main switch on the front.
We've got heat shrink uh, tubing here on the um, on the actual connections themselves and we've actually got um, extra um insulated sleeve in over the wires because they have to go under the board over here so that's a nice little touch. They know what they're doing there and they come up over to the I connector at the back. They got heat shrink under that. Not a problem.

Don't really like the way they've done the Earth this is H fairly typical of these kind of products. They've just got it going down to the uh lug uh down on the board here I don't like the ones that just go to the board but anyway, uh, it'll do I'm not happy with all the residue and uh, crap around the board here, that's you know it's not good at all. and uh, we'll see that further on on the SMD assembled Parts but uh, the all this uh stuff is actually uh, on the side of the connectors here is actually glued to hold the connectors in place to stop them uh, vibrating out? they shouldn't because they're usually uh, friction and or these are only friction ones I think or they might have a small um interlocking uh tab in there. but anyway, they've G on the extra step to glue them in so that's actually okay.

and the cabling from the power boards at the back through to the uh front panel binding post board all 188 um AWG cable so that's not bad at all. So they're all are cable tied in place which is good to to see. So we've got uh. here's the 5V uh Channel and the second uh Channel going over there down to the main board and also we've got the Uh, there it, it's tucked under there somewhere.

where's the other board? Where's the other channel? There it is. Yep, it's tucked up under there comes across here and there it is going down to the front panel binding post board which will take a look at. And here's the B metal thermostat switch and there are KSD 302 um type really? uh, you know, common as mud um and really cheap. Very simple and very reliable.

Now Um I haven't looked up the data sheet for these I don't recall but uh, it's got 80 on there. So I'm going to presume that that's Uh 80. um Celsius cut off for that and there's one on each. uh heat sink here.

they've actually got uh. two heat sinks which uh, they? This one over here is shared between Channel 2 and Channel 3 and this one's just dedicated to channel one. Now these are are actually a Uh 10 amp rated uh contact in these so they could actually um, use these to directly switch the output, but they, uh, don't. Of course they've already got relays that do that and you can tell they don't do it by the little piss an uh wiring they got going into these things and you'll notice that they're actually wide in parallel there.
So um, these are just so these are just actually used as um, uh switch inputs to the Uh micro. um I I presume goes into the micro which then tells it well over temperature, shut off the output relays. So by having those uh in parallel like that, obviously, if either one of these heat sinks goes over temperature, it's going to short the contacts and uh, you know, trigger the alarm to shut down. but it doesn't know what uh, which one has gone over temperature either channel one or combined Channel 2 Channel 3.

So it's a bit inflexible in that point of view. like if Channel 2 Channel 3 overheats, it's going to shut down channel one as well cuz it Doesn't Know It's Just going to shut off all the outputs. And here's the Uh Channel 3 uh 6V uh Channel um pass transist and it's the same type 2sd 1047 And even though that's a bit overkill for that sort of Channel it makes sense because well, you've already got four of them in your uh in, in your design, in your bill of materials, so well, why not just uh, reuse it for that channel. So instead of saving a few cents, the designers are just gone.

Oh bugp, we'll just reuse the existing transistor it. It does the job. It's A, it's a no-brainer it's a no risk, so let's just get on with it. So they've done that and as you can see, um, it is uh got a little uh seal type pad in there I'm insulating that from the heat sink of course because Channel 2 and channel Uh three.

This is the past transistor for Channel 3. They share the same heat sink, but they're completely isolated channels so you don't want them Shor in out. So it's very important that that is well insulated from that heat sink because you can see the other Tr transistor down there for uh Channel 2. It doesn't.

It's not um, isolated from the Uh heat sink, it's just directly connected boled uh, straight on there. So they've done well. They've actually, uh, insulated the individual wires there cuz they don't want those uh, shorting out against the heat sink either. Nice work.

And there's our Bridge rectifier down in there for Channel 2. identical one on the other side for channel one. but uh, channel 3 is different. Take a look at that.

So here's the circuitry for Channel 3. uh, wedged down in here and you can also see that they got relays down in here that, uh, associated with Channel 3. So that would be like the Channel 3 output uh, switching relay and stuff like that. There's the filter cap and there's the bridge rectifier.

They've got uh, just four standard diodes there like mounted like uh 10 15 mm off the board there to get some extra heat sinking on the Uh leads so you know they don't. They don't require the same large power Bridge rectifier that they used on the other channels because this is only a 3 amp Channel not a 5 amp channel so they've squeezed. Uh, there's a couple of current shunts down in there. you can see them.
Oh, you can see them down here. Sorry, the light's pretty poor in here, but look at this. I mean um, here's the main uh filter cap and look how close it is to the heat sinks. It's only like you know, 2 3 mm away at best and that's going to get all the radiator uh, heat not only from the heat sink, but the worst of the heat sinks.

Um, this one here which has to handle channels 2 and three. So in theory it's going to get the hottest and so that's poor design. I Mean if you look at the uh, other channels over here, there's a huge gap here. I mean what's that you know? 25 mm Gap or something between the main Fielder cap and the heat sink so that's not bad.

So in terms of uh, you know air flow like it's not directly next to it so the radiant heat is going to be less. But but then the air flow comes through the unit like the vent holes are up here somewhere. It all flows through and these caps are going to keep up reasonably cool. but the Channel 3 one in there don't like its chances.

um sorry I can't see the brand of uh, that cap in there, but it's not. Um, doesn't look to be the same as the ones on the main Channel And of course everyone wants to know what the main caps are, What brand? Well, I'm glad to say they're a Nipon Chemicon brand, one of the top brands on the market. These are the Kmh Uh series 105 C Black ooh Black Version Oh, it's got to be better and these are at 63v, 10,000 mic, and um, these Kmh series are designed um basically for uh, this particular application for linear uh power supplies high Ripple current things like that. So these are rated for uh 2,000 hours continuous operation at 105 C at full rated output Ripple current so not bad assuming they're genuine.

you can see though the silk screen on these I don't know. Can anyone tell me if they've got a genuine Nipon chemicon? if they if if you got to watch this in HD I Saw this on one of my recovered Uh footage videos I Can't see it on the screen very well here, but you can see that the silk screen almost looks not up to par. but hey, I'll give them the benefit of the doubt they're genuine. You pay 450 bucks for the supply you want.

decent caps. Once again, look at all the residue left on the board here. Not impressed at all. But once again the connector glued down which is good and we got some small heat sync Regulators here they're 7815 and I believe 7915.

so positive and negative. uh 15 volt rails there presumably for the opamps that are used on the uh daak and um ADC uh board in here which will, uh, take a look at in a minute. So once again, they've got their own uh Transformer tap here coming in. Not a problem, they got their own Bridge rectifier their own filtering and there's an onboard regulator somewhere on this board down in there which we'll see as well.
and the output switching reays. what brand are they hey? Shan I don't know, may as well be one hung low, eh? Wouldn't trust them, but I don't know. they might be reasonable I've just never heard of them now. I'm not really a big fan of having the Uh extra Earth leads connected to the Transformer lug over here because well, they haven't scraped the paint off and you know it's not the most reliable method it's got to go through I You know, in like it in practice it works, but it's not best practice if you know what I'm talking about.

So that's got one Earth wire going off there to the Uh front panel board. These don't go directly to the binding post terminals as we'll see The Binding posts are actually connected to their own Uh board in there, so that goes off to the Ground Terminal on the front panel and the other one goes off to another lug on the main Uh power supply board and we've got ourselves a genuine Ftdi Ft232. No surprises there that is on our serial SL USB input board and no surprises there. A Max 202 uh driver for the direct Rs232 interface and that board.

You can see there that board goes off to the uh uh D9 serial connector. So um, yeah, whether or not you use the Uh D9 serial input or you use the Uh USB interface, um, if makes no difference. It works exactly the same. It's just a uh serial interface to the machine.

Not entirely sure why they have a relay on there. um I'm not going to bother investigating, but uh yeah, don't know what that's doing now. Um, one thing you won't notice on this board is any uh opto isolation at all because well, you might think, hey, that's a bad thing, right? It's not isolated. Well, you don't actually, uh, need it because um, it goes off to the logic board, goes through this wiring all the way over to the logic board and the logic board.

Um on the front panel has its own uh isolated uh Transformer tap. So all of the uh digital logic um interface here is uh Main's ground uh referenced because it's going to go back. Be Well, it should be because when you plug in a USB cable, it's also going to have the ground Earth uh reference on it. So therefore, you have to run the logic circuit at that ground Earth reference.

Otherwise, you got to use an optoisolated interface and you don't really need to do that, so no surprises that it's not there. and I'll show you that that's the case. Let's go between no pun intended. Hey, let's go between the Uh case ground there and the USB input.

There we go. that's the Um negative terminal of the USB input. So they're uh, ground referenced and you'll find that the logic board is also ground referenced and you can see that there. we'll go between Earth and we'll go between the regulator tab.

There, There we go. It's the logic board on the front is Main's Earth reference, but nothing else is. Of course, the front panel terminals aren't the Um actual Uh voltage output terminals are completely isolated with their own Taps and with their own logic. And we'll find the opto isolators are actually on the logic board.
and there they are as you can see. And that folks, is a lot of Opto isolat as they go all the way across the bottom of the logic board. because you need to have multiple serial interfaces to, uh, all three of the isolated Uh channels. So that's why we've got three separate connectors there with the data going out, so there's a ton of isolation there.

And of course, that logic board uh on the front panel has its own uh, onboard, uh regulation. There it's got its own uh Transformer tap coming off. Yep, there it is. Own transformer tap down there and uh, which? Powers all the logic.

We'll take a look at the processor in there and there it is. It's an STC Micro 89c 58 and STC micro are a Chinese company nonmainstream it I believe it's an 80 51 uh compatible processor. So yeah, not your usual uh you know microchip or rill or you know, St or any of the regular ones. this is STC and other Logic on the board.

just lots of uh 74, HC uh 74 down in there. really? um, you know TTL stuff. Nothing special. There's the crystal for the processor there and uh sorry I can't get uh these Boards out to get uh, these Boards out I really have to take out the Transformer and then essentially strip the entire power supply back to nothing basically and I don't want to do that.

You can see that ribbon cable down in there going off to the Uh display module on the other side. Nothing exciting. It's just a chip on board off the-shelf you know, graphic display module. Nothing happening there at all.

so really no need to take that board out and air flow wise. By the way. nothing special happening. They got no air flow guards or anything like that inside so air is are suck through the Uh side vents at the front of the unit here so it's goes over the Transformer a little bit and then gets uh sucked through the heat sinks and out the fan and the fan pushes out the back and you get see.

tucked away right down in there is an M205 uh fuse presumably for the output of the main uh Transformer tap for each uh channel uh coming in there so you know it's sort of a safeguard. I mean you wouldn't normally expect that to uh blow because uh in, you know General use the current uh, limiting of the channels is going to uh, stop it but it's good to know they've got some extra protection in there and you can change it if it's a little, uh, hard to access. Got a couple of power resistors down in here and they've got little heat shrink uh little uh sleeves over the Um wires there just to stop them shorting out. Now you can't see it in this video because I've already straightened it up.

Um, it was in the previous video that got uh uh, mangled by this pain in the ass camera. um but these two um, current sense shunts they, um, were almost short in. they almost pressed together at this end down in here. So yeah, I wasn't very impressed with that.
so I just nudged them back apart a bit and I just confirmed that both of those power transistors in there for uh, each channel are actually uh, paralleled up. So they're um, you know they're using two devices as the main pass transistor rather than just Uh one and I'm not sure why, because one is, uh, perfectly capable of doing that. Maybe they just wanted the uh, uh, you know, the extra dissipation and uh to get it into the heat sink a bit more efficiently than a single device. So of course, this is a circuit topology where the one main pass transistor or in two in parallel in this case.

but they work just like one. Um is effectively doing both the Uh voltage and the constant current uh limiting as well. So it's not like they're using a separate device for the current limiting. So with two devices, you know, there's multiple reasons why you might want to use two instead of one.

I Mean these are like a, you know, 12 an capable devices more than capable on their own of the actual current handling capability, but you might want to, uh, spread your heat more efficiently across the two devices and then the um, uh Junction between the device and the heat sink is going to be a bit more efficient if you use two of them. It's just uh, better that way. uh, spread your heat around a bit and uh, and also, um, hopefully they will run well. They should run cooler as well, longer life effectively, and also, possibly they just, uh, didn't have the Um gain required in that one.

so they spread it across uh two devices so that could have Um implications in terms of the loop, stability of the power supply and stuff like that. Maybe they found that one device just, you know, it wasn't stable enough. So eh, we're going to spread it across too. But of course they haven't made that same decision with the Uh Channel 3 output.

just the one device. Now let's take a look at this vertical Riser board here, which is the Uh DAC ADC board. I First assumed in the first video, it was just a deboard. but I uh Now find it's both.

so check it out. We got some TL 71s. Nothing, uh Precision happening there at all. And one thing you'll notice is look at the shoddy soldering on these.

look. The parts are at all odd angles. Look at them and it looks like they've like been hand soldered or something. Absolutely shocking.

Soldering on this. Absolutely atrocious. Like some are Okay, those um S So8 packages aren't too bad, but you would swear some of those are done by hand. Absolutely ridiculous.

And further down, another T71 and a T74 quad version. Woohoo! Look at the alignment of those resistors there. Dodgy Brothers. And here's our deck There it is little eight pin So package and uh yeah, no surprises.
It's a reasonable deck. It's a Texas Instruments uh Tlv 5638 and it's a 12-bit uh voltage output deck single one. It's got a programmable internal reference I think it's about. 25% or something like that.

Couldn't quite get the temp C of the whole thing, but it's in the order of like 10 PPM per degree C for the gain, differential, nonlinearity less than half least significant bit and uh, it's monotonic over the whole range, so you know it's a reasonable deck. Does the business serial interface, and of course, um, the absolute reference in here is, uh, not good enough at .25% to match the 0.05% absolute performance of this uh power supply. But as I've said many times before, it doesn't matter cuz you can calibrate that out. This thing could be, you know, have a 5% reference.

Makes no difference whatsoever. It's all about the stability and the drift with temperature. So as long as it's uh, stable, not a problem. Calibrate it out in software.

and this software actually comes with calibration software so you can, uh, tweak it. Or at least you can get the Uh calibration software. You enter the password and you can actually uh uh tweak the thing. and they obviously do that in the factory before they send them.

But one of the biggest surprises I Uh found on this board was this chip here this little Uh 14 pin package. It's actually an MSP 430 processor. It's an MSP 430 F2013 and it's a mixed signal uh, low power micro controller. not very powerful.

It's only got uh 2K of flash memory, 128 bytes of ram, but it's got a 16bit Sigma Delta ADC with an internal reference and you can bet your bottom dollar there using that because there are also um, alternative Parts in this series The 20112 and others that are and the 20111 that are identical. but they don't have the Uh 16bit Uh Sigma Delta ADC in them, so they're obviously using that. More than good enough for the purpose. Classic case of one of these.

uh, little basic, simple, low power mix signal microcontrollers. Perfect application for it just controlling a deck doing the loop stuff so you know you can bet your bottom dollar. it's probably uh, independently monitoring. um, you know, the current on this channel.

Shut it down. The over voltage protection. the Other: the current protection stuff. like that each individual channel has its own little 16bit mix signal processor not doing much on 2K a ram.

they probably programmed it in Assembly Language you know, and it's probably talking to the deck as well controlling that. So good choice. I Like it and you can see some uh, star grounding happening there. There's the star point.

you can see them branching out like that. All individual points. Yeah, they there's four running out the bottom there. They know what they're doing.
Oh, and by the way, that Sigma Delta Converter 16bit one also has a programmable gain amplifier build in as well. Perfect little device or something like this. Not much else happening on the board, but we have a classic triple 17 Jelly Bean It's the you know it is the standard low Dropout voltage regulator. You see them in all the 10 Downs you know they're are dime a dozen you know, used everywhere and it's got 3.3 so that's a 3.3 volt version.

Uh, local regulation for the digital part of this and it looks like we've got ourselves a little serial programming interface as well for the micro. Just look at the quality of that soldering or lack thereof of quality. You got to be kidding me. That definitely looks like it's hand soldered and look at the ugly residue left over.

this board could do with a clean before it goes back in. So yeah, it's hard to tell what they've actually done here, like it looks like they have. you know, possibly reflowed some things. but that's just man.

that just looks for all the world like it's hand solded and they got one little Opdo coupler there. Not entirely sure what that one's doing now. I Don't know about you, but I Can't help myself wondering what happens if I pair it up without one of these. um daak ADC control boards on one of the channels channel two I've got disconnected there I Follow the wire wiring it is Channel 2.

Let's power it up. now. if they design this thing, uh, properly, well, you'd expect you know it. It would would do some error checking.

if it doesn't talk to it, it, switch it. down, it, show an error message on the display. You know if they took our pride in their firmware, have they booing? Come on, let's go. You'd expect a power up error message.

No folks. Look, the fan has gone full ball. The fan normally doesn't go full ball like that. But Channel 2 is still showing 24 and five.

Can we even, uh, adjust this 24? Let's say go. let's go something stupid like 50. No, we can't enter that n it's not going to let us do that. Five It allows us to set everything, but of course we're not going to get anything out.

W I Just switched the output on folks. Check it out. 38.6 6.6 amps. Are we actually getting that on the output? No.

we're getting zero volts. Nothing. So really, that's you know, poor design. I Mean you know you got firmware in there? you've got you know you're talking to.

From one microcontroller on the just doing the front panel to to the other micro over here, you think it'd have some power on checks and if it didn't, if there was a problem or something, you think it would pop up with an error message on the screen. That's just poor form. Now we've got a third. uh ADC Dack uh.

Control board over here and it's almost identical layout. um to the other ones. It's um, the other one's longer. Of course it actually extends down there because it needs to attach to the top of the Um railing up there.
but it looks to be an absolutely identical layout. except there's a whole bunch of transistors. or yeah, are they Q Yes, they're Q the designator there. so the transistors are not diodes Q4 there.

so they're all unpopulated. so I'm not sure what's going on there, but everything else looks identical. apart from two bodges. There is a bodge up there.

you can see the uh, they've deliberately shorted two pins there. they've done that on, uh, purpose for some reason. and we have another little bodge near the processor. Don't know what that is.

That's terrible Murial Now, while Channel One and Channel Two the main channels have, um, basically everything on this uh main board here, including some current shuns on the other side and that board I've plugged it back in so you can't see it. Um, that's not the Uh case for uh, well, channel Uh three Control board actually has some stuff down in the middle as you saw before there. It has the Uh Pass transistor mounted on the heat sink, so half of it there and the other half is actually um, well, here's the Uh ADC processor board for it. But um, here is, uh, the secondary board.

the main sort of the main board containing, uh, lots of filter caps and if you can see in there, a couple of current shunts as well, so that's all mounted on that main board because and then the wiring see that blue and the red one jumps over from Uh the Channel 3 board in there over to The Binding post board on the front panel and that Uh front panel board has all of The Binding posts connected to it and a couple of Um Earth coupling caps and the odd relay and that's it. And if you can see The Binding post board down in here, you can see the Earth cable coming in here and then you can see the suppression caps. There's actually four of them. there's another one down in there.

They would be fully rated Uh caps of course, but the Uh markings are on the other side can't quite see them. You might be able to see some relays down in there on the front panel binding post board there for uh, your parallel and Ser series mode for switching your Uh Channel One and Channel two Uh terminals either in series, parallel, or independent and you should just be able to see The Binding post down in there attached to the main board. Sorry, very difficult to see it be impossible to take this thing apart, it take me hours and never get the bloody thing back together. but uh anyway, they do have yes they are Shake proof washers.

You can see on under the nuts on those binding posts so they shouldn't give too much of a problem. Okay, What? I'm going to do now is just, uh, measure the temperature of the heat sink and uh, see what it gets up to. So I'm loading Channel 2 and Channel 3 here because they actually share the same heat sink. so that's a worst case.
So we'll only measure uh, the one heat sink. the other one's going to be identical. Uh, performance minus channel 3. So um, I've got 24 volts uh at 5 uh amps.

so that's 120 wat and I've got 6 volts at 3 amps via a resistive uh, dummy load. Um, here on Channel 3. so that's another 18 wat. So like 138 Watts total.

So I'll leave it running for a little bit. and uh, we'll measure some temperatures. and of course I chose 24 Vols there because that is, um, the lowest, uh, almost the lowest voltage of the highest Transformer tap. So we should be getting on the highest Transformer tap.

We'll be getting the most dissipation in our heat sink, the most voltage drop across our pass transistor, therefore the most dissipation in our heat sink, and uh, we should. um, after we measure that temperature, then we'll drop it down a uh, drop it down to the next tap, which will be about 23 volts and then there won't be much margin or voltage drop at all across the past transistor. So the temperature should then start to drop. so let's feel that.

Oh yeah, yeah yeah yeah. Hty hty Burnie Burnie Wo So let's try and use our uh little IR thermometer here. You know the emissivity may not be that great. It's showing that transistor.

uh, sorry, that dire Bridge down in there 8790 showing the heat sink or the edge of that getting up to 90. But I don't necessarily believe it. Um, the 5V Uh pass transistor there 72 The Uh laser is offset by the way when you're this close 70 odd deg h I don't particularly like that at all. I Mean you know if you put it just on the heat sink? 37 it's hotter than that.

So let's get a real temperature probe. All right, let's try our fluke: Cnx. T3000 Pretty Dar Uh. Accurate device designed for the purpose.

We got a K type thermocouple probe. Let's get in there and let's get the Uh temperature on the Uh directly on the little stud for uh, the screw for that. 5V Uh, sorry the 6V Channel 3 pass transistor there 62 64 There we go. We're almost a 70 on that uh 5V pass transistor directly on the screw of course.

How about the the Diode Bridge down here, it's about 77 or thereabouts on the Diode. Bridge Let's do one of the uh, past transistors for Channel 2. Yeah, we're getting up near 80. Yeah, well, over 80 there on the screw.

Let's see what our Heat sink's doing. Now that's directly on the Uh screw for the Uh thermostat up there so which is screwed directly into the uh Top Fin of the heat sink up there. so that's you know. 61 I think I saw it get up to about 62 heat sink.

That's kind of what it felt like. Yeah, so I've had this going for maybe 10 minutes and that, um, that heat sink is, you know it, 65 something like that I mean sure, we've got the case off, but I wouldn't expect it to get too much, uh, warmer. I mean that Fan's going uh uh. full pilt there.

So really, you know it'll be sucking a fair amount of uh air through keeping it really, relatively, uh, cool. but yeah, um, as uh, we uh said before like I think those uh, thermostat cutouts are right. Well, they've got 80 marked on them. so I presume that they 80 C cutout and that would sort of make uh, sense from a design point of view.
Probably wouldn't want to get them much hotter than that. But there you go. We've got Channel 2 and Channel 3 on, um, practically full load there and well, we're 65 odd de. So what I'll do is I'll just change this.

uh down on Channel 2. let's go in here and vset and we'll just change that down to we hear the relay click there we go. 23 volts. so we're now one tap down.

but we're on the high side of that one tap so we should be dissipating. uh, less power in the past transistor. So our temperature should drop in theory. so let's see if it does, Let's keep our probe I'm going to put it back on.

There needs to get up to Temp I need to keep a consistent uh pressure on that, but we should actually see it. Uh, eventually come down. it's just getting up to temperature the thermocouple. but once it does, it should start to climb down.

There we go. We might be seeing that now. And yeah, that's really starting to drop now. So before what we saw with 65 C was the worst case.

uh design scenario. worst case load scenario: Uh, really. pretty much so. although we could check all the Taps and the various margins on all the Taps, but jez don't want to Guild the Lily just want to get a ballpark.

So yeah, um, that seems to be operating at quite a safe temperature. so pretty happy with that. Uh, pretty well designed. They do know what they're doing with the multiple Taps I'm sure they've done their calculations and their um, thermal uh, tests and measurements and all that sort of stuff.

And for good measure, they've put in the uh, thermal cutout as well. for any you know, thermal failures with thermal runaway and things like that. So nice work. What's the brand of the stupid fan in this snow brand? H 24 volts, 0.2 amps.

What a pain in the ass. that thing's coming straight out. So there you go. I had to go through and shoot all that video again.

Oh well, maybe I missed a few things from what I had in the previous one. Uh, it doesn't matter, it's like uh, 7:00 on a Sunday night H I'm going to be in trouble so I got to head home. Oh boy. So I hope you like that tear down that was the atten PPS 3205 T3s and what can you say? Well, from a tear down point of view I you know it was basically what I was expecting I was a bit disappointed.

um, with the quality of the soldering. but uh, it's just a cleanliness and sloppiness in in uh, terms of assembly. So they've clearly, uh, shot this thing out to the cheapest assembly house in China I mean and it shows, but it seems to be quite well engineered. I'm impressed, you know, nipon K chemicon uh, caps and they've got thermal cutout and it seems to be, you know, recently designed, big, nice, beefy rated Transformer in there.
nice safe operating temperature seems to be, so you know I'm reasonably happy with it. I mean you know it's not a it's not a high-end unit, it's built down to a price and well, the tear down kind of reflected that because I didn't really find anything inherently, uh, wrong with it. You know, a bit of sloppiness here and there, but overall, not a bad design. So there you go.

If you like the tear down, please give it a big thumbs up and give me some sympathy as well for losing those bloody files. Pain in the ass, stupid dumbass camera A God. it's just like Canon's Pruma camera, you would you know? Top of the line: Pro Sua Camera You think it' be? you know, Usable Bloody hell. Losing fire was inexcusable.

Unbelievable. Anyway, if you want to discuss it, jump on over to the EV blog. Forum Catch you next time.

Avatar photo

By YTB

23 thoughts on “Eevblog #440 – atten pps3205t-3s triple output power supply teardown”
  1. Avataaar/Circle Created with python_avatars Ted says:

    Its good to know I am not the only person who has to do complex operations twice. Thanks for the vent…makes me feel better.

  2. Avataaar/Circle Created with python_avatars John Cliff says:

    Good day there Pal. Git your arse down to the Boozer and have a few jars of your favourite beer. You sound like you need it after loosing all that work you'v done. At least you won't have a dull grey damp day like we have here in the U.K. Love your videos. Keep up the good work. 73 de John – G0WXU ,Ex. 9M2RH , 28 COMWELL BRIGADE. Mallaca. Malaysia.

  3. Avataaar/Circle Created with python_avatars Jeff M says:

    Isn't Atten just a lower grade Siglent? I've seen so many products that look identical.

  4. Avataaar/Circle Created with python_avatars Слави Стоянов LZ1SSA says:

    Въпреки вентилатора,охладителите са малки.

  5. Avataaar/Circle Created with python_avatars N Tag says:

    At least the boards are glass epoxy. Frame, cover, front panel and other mechanics are just fair.

    Design is heavily dependent on fan. A mixed blessing.

  6. Avataaar/Circle Created with python_avatars Randy Carter says:

    The reason for the 2 paralleled outputs. It's not just the current, it's the watts. When you multiply the rated voltage with the rated current often exceeds the power dissipation in watts. By paralleling 2 transistors you double the watts of dissipation.

  7. Avataaar/Circle Created with python_avatars Crocellian says:

    Dave, you finally did it. You failed to understand a heat sink. They take hot spots and dissipate them.

    Again, you stop at nothing to bash the Chinese.

    Guess what, without them, your industry would be dead.

    Chew on that.

  8. Avataaar/Circle Created with python_avatars Doug Gale says:

    Plastic front is a bad thing? Isn't it safer than a conductive front panel?

  9. Avataaar/Circle Created with python_avatars Andy P says:

    I think the vital parts this power supply is missing is it hasn't got a current shunting valve entering a resistive thermionic flange 🙂

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

    it is a dvaid thing

  11. Avataaar/Circle Created with python_avatars Dawn Minilla says:

    A single 100 watt transistor most certainly can not handle this supply. With a top tap of 34v that's 48v dc, minus fwb 1.5v, minus wire resistance of maybe a volt or so, transformer droop at full current maybe a couple volts, second to third tap transfer of 22v, that leaves you with roughly 20 volts and at 5 amps is 100 watts. Transistor derating curve for this device at 80c is down to 55 watts and at 100c is only 40 watts, so you most certainly need two transistors and is actually a close shave. Not sure why he thinks one is enough, did I miss something. And the 2sd/2sb complementary transistors were designed for audio use, not power supplies. Don't know where he came up with that.

  12. Avataaar/Circle Created with python_avatars Jože Mikec says:

    Haha, thumbs up ! I'm laughing at your reaction on video clip failure 🙂 Love EEVblog, realy nice job, i'm learning tone of stuff, thanks for your uploads..

  13. Avataaar/Circle Created with python_avatars Samsung Handy says:

    Thumbs up!

  14. Avataaar/Circle Created with python_avatars Fingloid Fing says:

    on my unit, one of the D1047 power transistor failed because of poor placement on the heatsink. i notice that the main power board was designed to hold four D1047 power transistor. so I added the twoo missing D1054 and the one that failed.
    my question is, is it right to add the additional 2 Power transistor? i cannot see a power supply on the internet that has 4 D1047. what do you think? im talking on a power supply only one channel

  15. Avataaar/Circle Created with python_avatars Amy Marie says:

    just looking at this it screams CHEAP!!!!!!!!!!!!!!

  16. Avataaar/Circle Created with python_avatars Julius Ski says:

    Nice review, just as a tip: you could try to short thermostat contacts to see what happens once heatsink(s) reach 80C point.

  17. Avataaar/Circle Created with python_avatars hereiam2005 says:

    I think some Canon cameras has twin SD card slots or SD/CF combo so this kind of shit does not ruin your day.
    I shoot photos as a hobby and always use the camera's second SD slot as backup – sloppy card writing ruined so many of my photos before it wasn't funny.
    Short of that I think Dave could do with a dual camera solution.

  18. Avataaar/Circle Created with python_avatars Harry Haefner says:

    Nothing like overpaid IT people, not thinking (related to your camera record function)!

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

    I used to be  a lead technician on a electronic manufacturing line for over 10 years and I've seen soldering messes like what you show in your videos. One thing that made this seem to me to be more of an isolated case is that you had other boards in the PS which showed decent soldering. I've seen every cause of soldering problem issues from dunking boards in the WAVE solder to just a little chatter on the guides shaking components out of the pre-paste. All those defective cards get put aside and eventually wind up in a re-work and then it can be" solder sucker to the rescue" scenarios. I've seen the "tale-tale" signs of these defective cards in your videos, they had whisps and slivers of hair fine solder on the boards and also solder balls. So I think these cards were re-worked and were just not cleaned properly after the re-work. That left-over solder on the surface is just an accident waiting to happen. Over time, heating/cooling/vibration can dislodge those hair-fine solder strings and they can short out between component leads, etc.
    I agree with you, it was a pretty crummy job of quality control on that soldering, I can understand why it is that way, however it is unacceptable. My point is that I feel the soldering is more of an isolated case and not respective of the continued over-all process. But I've seen stuff like this get thru and with all the manufacturing quality control safeguards, it only takes one person out of hundreds not doing their job for something like this to slip thru.

  20. Avataaar/Circle Created with python_avatars GameProgrammer79 says:

    That was a power supply GHOST that went inside your camera and screwed all your videos..  some times ghost dont like tear down  and haunt 😀

  21. Avataaar/Circle Created with python_avatars smbrob says:

    At 28.49 I see dodgy solder connections on the TLC568/ U7 I would swear they are broken.

  22. Avataaar/Circle Created with python_avatars 4D Workshop says:

    I don't think there is an issue with the cap that is almost 2mm away from one of the big heat sinks.  As you mentioned, the air flow is going by the cap, then through the heat sink, towards the back of the unit.  This means that the air flowing by the cap is way cooler than if it were flowing the other way.  Does that make sense?

  23. Avataaar/Circle Created with python_avatars Matheus Moreira says:

    It's interesting the way you criticise others' designs, like if you would have made them any better… ¬¬

Leave a Reply

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