Hi Just a very quick follow-up video to the Ry Gold Dp832 uh bad design investigation video that I did some time back. If you haven't seen it, this won't make much sense. So I'll link it in down below. so check that out.

and yes, I've uh thanks to John South Addim Mona Um here in Sydney they're the Australian uh representatives for Ryal. They have upgraded my uh the PCB to the new version to fix this exact uh thermal problem which I discovered in the Uh Dp832 power supply and has it made a difference before and after. Let's do some very quick measurements to find out. Yes, I'm uh getting about 247 uh volts Mains voltage here in the lab.

there's no load connected at all I've had this turned on for a couple of hours now I'm actually monitoring the heat sink through the side of the case as we did in the previous video and last time we'll get in I think from memory well over 90 just through the side of the uh case like that and now we're getting a much more reasonable 57.6 de uh C 50. You know, let's round it up to say 58 so that is certainly much much better. Now we're going to uh, crack the thing open and uh check, see, look at the design differences and compare it with the old Board of still got the old one and this is the old crappy little uh heat sink as you saw in the previous video that uh went got to over you know uh well 90 with the case on a well over, 110 or more with the case off with no airflow now I Think they may have actually increased the fan speed slightly. but I could be uh, wrong about that because I can't just whack the old board back in and uh, try it.

It has to do with the uh firmware which isn't uh on this board. this isn't the main control board for that. So um I will just switch it off and I'll up the gain here on my microphone see if we can hear a difference. It does not go at full speed though.

So uh, that rumor um has been put to bed. No, they haven't just set firway, set the fan to flat out. It does actually slow down. So here we go.

So there you go. Hopefully you heard that you heard it initially. speed up at power on and then, uh, slow down once it was actually on. so it does not operate at full speed.

They haven't tried to fix the problem that way. they've increased the size of the heat sink and of course the fan still does speed up under uh software control if we load down the channel. So I'll do that now. So how do you check if you've got a new unit? uh with the modified upgraded Uh heat sink? Well, I'm not aware of any uh serial number uh, you know, ranges after which uh, the Uh mod has happened.

Uh, not exactly aware of that, but all units currently sold should actually through any reputable dealer. um, authorized dealer should have the mod from, uh, certainly the time of this video. it's actually um, they actually have been on sale for a bit of time, but you can actually see, uh, confirm it directly through the right hand side upper vent. Grill here right at the back.

if you can have a look through there and if you can see that large, you can see it down there. Large silver heat sink like that that goes all the way down like that it it should be a silver. uh heat sink. Well, silver aluminium.

uh heat sink. Of course. just bare aluminium, not black anodized. If it is small or black anodized, it means it is one of the two previous versions of the board.

And of course here is my original board. The first version as I showed that is really, really bad. Really, really horrid. But there is another version of the board in between this one and the new one.

I've got inside here which has a rearranged heat SN The capacitor moved over the heat sinks over here and it is a smaller. well, it is a larger black anodise, but not nearly as large as the unanodized one that we're going to see inside here. and I've taken the case off and you can see the large heat sink right down in there, so you know it is pretty easy to, uh, tell you've got the right unit through the vent holes. And here's my unit open with the upgraded heat sink as you can see, um, much much larger heat sink than the little tiny pissant one that we had on here.

It's uh, not black anodized. Not a huge deal really. It is a thin tab. um LM 317 one of those thin ones.

but once again, that's not a huge deal. and um, yeah it is. It is pretty darn large. It's as large.

um, it's about the same height as the uh, largest heat sink on here. I Won't bother really getting good angles on that, but there you go. it is just as big. um it it you know and it brings it down to an adequate temperature.

They now don't have the Caps uh right next to it at least. Um, so you know it's not too bad at all. The airf flow of course comes through the vents at the side like this and is sucked out the back like this. so you know there is some air flow over that heat sink.

So what I'm going to do is I do the same thing as last time. I'll power this thing on and I'll measure the temperature outside of the case just for uh, the reason that we can do it and to verify that it doesn't get too hot with no air flow. But of course, unless the fan fails in your unit, no airf flow is you know, not uh, part of of the real uh, you know operational aspect of this unit. We're just doing this for kicks and the version of this board is 2.10 the Uh 26th of March uh 2013.

So that was actually done before my Uh design investigation. this changed. but I definitely know. As I said, they have upgraded this heat sink to a different size one.

They just haven't changed the layout since then and my old board is there. It is. Uh version 2.0 the 14th of January 2013. Now as far as the differences between these two boards, well I can't see any difference at all apart from the Uh culprit LM 3175 volt Supply around here which they've um obviously changed.

So they have made two upgrades to this board. The first was to you know they realized this was too small. Oops, they rearranged it. uh, moved the capacitors back, added the footprint for the larger Uh heat sink over here as that getting hot.

oh oh yeah I can feel that oh o touchy Burnie Burnie But then they originally had a smaller heat sink than this anodized they had a smaller black andise one. Now they've gone for a larger one which I think matched the same footprint so they didn't have to upgrade the board. But apart from that I can't see any changes anywhere else on this board at all. really unless I'm missing it.

but you know I'll post some photos up. but really I think that's the only thing they've changed. and if I get down there right on the tab that little thin piss an tab they've got on that Lm317. Uh yeah, I'm getting over 70.

I could sort of leave it there for a bit longer, but uh, certainly you know, well within the operational uh, limit of that uh uh case. Anyway, and in turn, the Uh Junction itself inside the Lm317. so not a problem at all even outside of the case. But as I said, this is not the real temperature cuz the real temperature.

you would have some air flow over this thing at the moment. we don't this is just free air and I've left that for a bit probing the uh top of the the inside Channel there of the heat sink. Oh, been there so long my fluke Switched Off There we go and we're getting over 81. See once again still with inside you know operational margins.

even with even with no air flow, Much better than what it was before of like well over 110 I think someone even reported like 120 or 130 or something. Ridiculous. It was off the charts certainly well over the Uh Max operational uh, die temperature of an Lm317. and no, they haven't uh, upgraded the Uh Transformer or the Transformer tap.

It's certainly not part of the upgrade, whether or not they'll do it in the future to get a slightly lower voltage into the Lm317 after the full wave Bridge rectification there I don't know I haven't uh heard anything but presumably uh, they won't change that tap at all. So as you saw right at the start of the video I am on the very quite the high side. almost 250 volts here in the lab. Uh, sometimes you know 248 common, certainly over 245 nominal most of the time.

so um, unfortunately. um yeah, we do get a quite a high Uh voltage into our Lm317 there. nothing's changed in that regard. So that's uh, basically problem resolved here.

I mean no issue whatsoever. If you've got an older Uh unit affected unit, Ryol will, uh, replace The Bard free of charge for you through your Lo local dealer. Not a problem at all. Just contact your local dealer and I'm sure they'll do that.

No issues whatsoever and uh, I'm you know pretty much worst case here, the 240 or 250 volt uh scenario. things like that and uh yeah, it doesn't seem to be an issue anymore I Still would have preferred like a lower voltage tap I Still think you know there's too much voltage coming into there the Lm317 there. but eh, you know it, it doesn't matter, it's going to um, work just fine now. Very happy with it.

And I as I said I don't know if that fan is a tinge louder than last time. Maybe it's just my imagination I Don't know whether or not it's just like a two-speed uh fan or whether or not they have actually upped uh, the airflow just a bit as well. Not entirely sure. Anyway, there was one other issue or two other issues.

Actually, the first one was a uh, software issue. Uh, and they have fixed that one. It's where on Channel One only here. Uh, if you set your current limit to 10 milliamps down here, anything below 10 milliamps.

So if we go across anything below, you know, let's zoom in so you can see that I've got. uh, here we go, anything below 10 milliamps. it would show like uh, 40 milliamps on there or something. if you drop below that 10 milliamp.

Uh Current limit. Even though you had no load on there at all, that was fixed. uh, quite some time ago and not a problem. And here's something I haven't noticed before.

it's not I don't know. It's not really an issue I guess, but it just shows the topology that they've got inside this thing. Let's say set this right down to let's say 5 milliamps Uh Current limit. Let's switch off Channel one here okay and then switch it back on and watch the voltage ramp up.

I Got it set to 30 volts. Okay, so pretty much maximum and you'll see it won't instantly jump to 30 volts. It ramps up. It takes time to charge up, charging the output capacitor on the Uh near the terminal front panel terminals down here and if we you know even set that much lower like down to 1 milliamp there, let's do that again.

You'll see It'll ramp up even slower in proportion to the capacity. You can even work out the capacitor based on the time and the uh uh current there I'm sure. So look at that. It just takes some time to ramp up.

so that's the output topology used in this thing and you can see that that is actually real with the meter on the output there. and if I up that I can just uh up the current there I can make it go real quick and of course if you set it to you know, 50 milliamps or something like that, it's pretty much going to instantly sort of ramp up to your 30 volts there, especially at like 100 milliamps. Not a problem whatsoever. So that's an issue with uh uh, some of the topologies used in these: uh Power Supplies the output configuration of the sense resistor for to do the current um sensing and hence uh, do the constant current and having a large output capacitance.

and as I've mentioned in like My Power Supply Design uh, videos and stuff you want to for a constant current power supply like this, you want to minimize your output capacitance for that charging up reason and also for the reason that uh, when your current limit kicks in, you don't want that capacitor which can store a lot of charge in it to actually dump out excess current over and above your Set current limit to your load. Now there was another big issue that somebody on the Forum are notice because these Uh channels here aren't floating. These two common terminals are actually. these two negative terminals are actually common and you can see that they're shorted out internally here.

The the channel one is an isolated output, but Channel 2 and Channel 3 are share a common terminal and if you hook up the load incorrectly on these, then it can cause an issue. I'll show you okay. we can easily demonstrate this. What I've got is Uh channels 2 and three Here Channel 2 is just connected up to the volt meter here and where As you can see, we're pretty much bang on there almost to the last, uh, least significant digit.

Uh, Channel Three here. I've got I'm just basically shorting it uh out through this current meter so that we can just ensure that the current's exactly the same. and there it is. you know, 1 amp.

N98 and everything's hunky dory when you keep these. Uh, when you separate the channels like this. But as I just show Cod These two common terminals are connected together internally inside here, so it shouldn't matter Theoretically where we return this current Like the current coming out of this channel 3 positive terminal. Here, we should be able to return it to either this terminal or to that terminal.

Aha, let's try that. So what I'll do is I'll take I'll disconnect it. So now we got no current flowing through here and instead of having it come back to there I'm going to have it come back to this terminal over here and watch what happens. We still get our one amp.

Everything's just fine and dandy on that channel. But look what's happened on this channel here 998. It's dropped by 20 odd molts there and for it's still SP Um, probably more disturbingly is that it's still showing 10.5 here. So the actual output voltage we're getting an error on.

and for this, which is a Precision Power Supply in the order, you know? Uh, 0.05% basic accuracy? that kind of thing. This is a really big deal, so there's something going on there with the internal way that that ground comes back. These aren't just shorted together internally and of course that area gets worse. The higher current you go I'm now at 3 amps there and uh, well, it's just look at that, you know that's not good at all.

but you may think oh well, what's the big big deal? Don't you always return current back to the terminal? Well, no, not always. Uh, in terms of these uh supplies, you may just want to run to your product out here. one common wire and the two voltages. You don't necessarily always run those two separate grounds, so that's very common to do, no pun intended.

And uh, you know which terminal do you choose? Either way, you're going to screw it up because of the internal sense resistors there are in the negative lines. So as it turns out, Ryo have actually uh, taken this issue on board and they've thoroughly investigated and they've written this uh, whole document which I'll link down below which uh explains what's actually going on here and well, how to avoid it. Um, they haven't actually fixed the issue because it is essentially a design issue inside this power supply and a limitation of the topology they're using. and uh, they they show uh on this page here this is the proper connection as we saw.

This is say Channel 2 and this is Channel 3 down here and this, uh, red here shows the current path comes out of the positive terminal back through ground and through the current shunt resistor down in there. and then that's our current Uh current. um, error amplifier there and that's all hunky dory. When you have those two separate paths, there's the yellow one.

Here's the positive terminal goes out. Here's the negative terminal back here. but there's a common sense wire which comes back to well, there's a sense wire from each terminal cuz they're doing 4 terminal measurement here. So there's a common sense wire coming back to a single ground point on the board here and that is causing an issue which they they show on this one which is the improper connection which we just demonstrated here.

Uh, channel, uh two. For example, when you run the current out through the positive terminal back down to there, it's fine. But on Channel three here, when you run the current instead of looping it back to its own ground ter Al if you put it to the other Ground Terminal here then that current as you can see on the yellow line actually flows. That current flows through those sense wires, hence causing a an error term there.

which then is going to be different because they're tapping the voltage off at a different point. That's why the voltage. oh here. Sorry.

That's why the voltage. um doesn't actually change. The displayed voltage doesn't change, but the output voltage does actually drop. So that's a limitation of having these lowside current sense resistors and this Uh Sens in Terminal Arrangement they've got in here.

They just, you know, it's just not suitable for this kind of thing. So Ryol actually don't have a fix for this. They're not going to fix the power supply, or at least uh, you know at the moment they don't intend to fix this in the power supply. they're just telling you how to connect it properly.

and when you're powering your product under test, use the correct return ground and well, you know, yeah, that's a solution. Another kind of a Uh solution is to put a big uh grounding strap between the two like that and that can minimize it. Doesn't get rid of it entirely, but it can minimize it. We're one amp here and as you can see there, you know it hasn't dropped by much.

We're still within Uh spec there, and if we go channel three and we increase that to 3 amps, for example, there we go. it's still dropping down a bit, but you'll probably find that well, still well within spec, so that's a way to get around the issue. Uh, just strap those two right at the front terminal like that. So there you go.

I would have preferred that Ryo Uh, fix that thing. but you know, hey, it's not an absolute uh show stoer. You just need to be aware aware of it as a user of this uh power supply. and I'll link in that document uh, down below and Ryo probably update the manual in uh du in due course.

Uh, pointing that out I'm sure they'll get around to that. So yeah. I'm I'm fairly happy with this. Upgrade it.

uh it. You know it did the job. it's fixed it. I'm going to put this back on my bench and use it as my main bench power supply so it's not a bad.

Supply Now they're starting to iron out all the problems and Ry Gold did take it seriously and they fixed it. So um, if you've got one of these old units, go back to your dealer and they will fix it up and replace that board for you for free. And pretty much, um, you know old stock is just going to get sold out. Make sure you buy from a reputable dealer, otherwise you may get stuck with an old stock unit.

Maybe I don't know. It depends on when you watch this video on when you buy this thing, but this is a you know it's a half decent power supply now, especially for the money. Um I think it's you know, a pretty much a killer. Uh Precision Power Supply just that trap for young players there with that Common Ground Catch you next time.