Hi In the previous video, we took a look at repairing this Stanford research Sr 6508 pole elliptical filter I scored on eBay real cheap and it turned out to be uh, the Main's filter at the back had a failure in it and once I fixed that, well it powered up and it seemed to at least go through the motions and uh, but I was playing around with this after uh, the video after I shot the video and well, yeah, I noticed a few things weren't quite right I noticed them during uh, the video. things like the uh, you know it didn't sound like the reays were working and uh, you know, stuff like that and the in-out filter wasn't sort of working and well, yeah, a few things that hinted that there might be something more wrong with this and uh, some people also noticed in the video a few strange things which I found in here as well. So there's more to the story here. It's not just the mainsfield, we' got more to fix here.

M Beauty Let's take a look at it now. One thing I noticed and also a few uh, eagle-eyed viewers noticed this as well. Check out this regulator in the back here. it's a 7805 5V regulator and look, it's got a seal pad on it, one of those insulating seal pads, but there's no matching heat sink and also, it looks like it's had the screw in there so somebody has taken that heat sink off at some point.

Why? And I eventually noticed, as did a few eagle viewers that, uh, the display was dimming on this thing and even one person noted at the same time, the fan was actually ramping up. So we may or may not be able to uh, actually see that here. But the this display will eventually dim after a few minutes and well, that ain't right and the fan actually speeds up and gets louder. So why would a lead display like this dim? Well, all of the digital stuff in here I Believe believe is powered from 5 Vols It's pretty obvious that's the case.

probably that 5V regulator that doesn't have the damn heat sink on it cuz I think that's the only 5V regulator in here. Oh, there we go. There, we go. It's dimmed.

It's dimmed. There we go. So the only Re I better turn that off. That's that's not a good thing.

Okay, but the only reason that would dim is if the voltage rail actually dips. I Mean if the voltage rail goes up, the display brightness is going to increase so that 5vt rail must be dipping. Now, the digital board at the back here, as I said, has its own 5V Rail and also all of the displays on the front. Those displays will be multiplexed.

uh too. of course they have to be. There's not enough uh wires coming through to display all the segments, but they're Multiplex but they would be going through dropper resistors and the uh effective brightness is going to depend of on the power value of those dropper resistors and the fixed 5V rail in here. So they that voltage rail must must be dropping.

so let's take a look at it and measure it. first. we have a quick look at the power supply section here. now the Transformers over here secondary of the Transformer going over here and we've got some voltage Regulators obviously not dissipating much.

uh Power at all. they're just freestanding. got a tiny little pissant heat sink on these things? So, but really? cuz they're only driving all the analog circuitry. Now what we got here is because we've got two different analog boards in this thing, all with their own separate Taps separate isolated because everything in this is isolated so the 5vol digital board will have its own tap and its own Bridge rectifier and filter cap.

And likewise, for the two different supplies, we' got plusus: um I I don't know where it's plus - 12 or plus - 15. but anyway, um, split uh Supply and we've got an LM 317 and an LM 337 for one analog board and another matching 317 and 337 for The Other Board And that's our 5volt voltage regulator. As I said, with the heat sink missing, that's a 7805. This one next to it is actually a Uh tip.

uh Power transistor I believe that's driving the fan, so that's pretty much all we have on the regulation side of things. Really quite simple now, of course, power supply issues like this and power supply diping will explain some of the stuff we saw in the previous video like the reays uh, potentially not uh working if they don't have enough energy to operate the relays. So maybe it's not just the 5V rail uh, the relays cuz the relays will be on the analog uh boards here. so maybe the analog rails are out too.

But anyway, let's stick to the 5V rail. I'm measuring the 5V rail. uh with the meter here. let's power it on and see what we get.

Here We go: 4.99 Everything's fine. hello, hello, It's ramping down. It's ramping down and we might actually see this is interesting. There we go is ramping down.

We might start seeing a dip on the display if it goes too low. I Mean you know it's still not an issue at the moment? I Mean you know, still within that nominal 5V spec you'd expect of a 5vt rail. So 4.75 Here we go. Whoa whoa, whoa whoa.

It's dropping. It's dropping. Should see the display dim. Holy.

it's going I mean the digit? The process is still working even though it's way outside, it's under its nominal 4.75 Well, there's something seriously going on there to give the Digital Credit it's still working. The process is still working at 3.9 now. of course. That's pretty much the kind of thing that you'd expect if your regulator is, uh, overheating, and, uh, dropping out.

you'd expect funny business like that. So you know how it started off at the 5 Vols and then it ramped down. Well, that's of of course, when it first powers on, the D is just fine. it's You know it's drawing the same amount of current as later, but it hasn't had a chance to really heat up yet.

So once it does Bingo it starts dropping. so we need to whack as a first order. We need to whack a heat sink back on there now based on the Uh I'm not sure the original size heat sink that was on there, but hey, there's physical limits to how big it could have actually been. Uh, likely to be very similar to one of these.

probably slightly bigger, but it can't be very wide because there's that power transistor next to it. Can't be very tall because of the wiring. can't stick out much cuz we' got a T92 there and other stuff. So really I'll have a rumage around and see what I've got in terms of heat sink.

but why is it missing? Like did it? I Can't believe that it just you know it came loose through vibration or some other thing. It's like somebody's deliberately taken the bloody heat sink off. I Don't really have much uh Choice here. I've got a I've got a couple of uh, these clipon uh type heat sinks, but they're not very good.

You know what are they? you know? 40 SE water or something horrible like that. I've got this, uh, largish one. It's quite. uh, it's quite thin, but that would probably do the business if I can, uh, squeeze that in there I've got a a screw on, uh, one with a threaded hole, but that, uh yeah, could probably maybe screw that on, but you can't get the screw into the back of that thing.

you can't get access. I might have to take the filter out again to get a a screw through there. Oh, it's all looking a bit I don't know I'll try and squeeze this one in see what that does. I'm not sure the specs of that.

that's like, you know, maybe 15 20 C per watt. but hey, it should do it because ultimately, it shouldn't need much. uh, heat sinking because it's just powering the digital circuitry here. uh, Z80 processor for those playing along at home and your ROM and your RAM and your Gpib interface with your National Instruments chipset and really not much else at all to actually pair it.

So not a huge requirement. and you can tell from the design of the thing that a it's freestanding and B there's not much room around it to put a heat sink. and they didn't design a a PCB mounted heat sink in there with the studs in the bottom like that one for example, so they didn't. You know? design that in so it you know it must not be dissipating much power at all, but with absolutely no heat sink.

Clearly, it's not enough. I wouldn't have expected it to fail so quickly overheat so quickly though with no heat. SN I sort of. Maybe you know, expected a bit more margin in there, but oh well, that's what it's doing.

So let's put the heat sink on and see what happens. So there you go. That's the best I could do with the heat sinks to hand. I might be able to find a more Optimum one if I could Salvage it from an old board or something like that, but it just wedged in there right next to the power transistor.

here. it's not shorted uh, out there to the tab of the other transistor. You just have to be aware of that and I just bent over the to220 there. But you know, you can't imagine there being much of a bigger original heat sink in there, so that's got to do the job.

unless of course there's some other something else on the digital circuit actually loading down the regulator. Hey, we don't know yet. Let's just power it back up, see if we get that sort of same, uh, ramping effect, and let's measure the temperature of the heat sink. All right, let's pair it up again.

There we go. 4991 and holding, we expect it to maybe drop a smid in as it starts to warm up, but shouldn't get that uh, runaway effect that we got last time and well, yep, it's holding steady so two your thumbs keep going. We need to get in there and measure the heat sink, but I'll leave that on for like 5 minutes and uh, see how see how warm that gets. It's yeah, it's dropping a tad, but no big deal.

Probably should measure the input voltage too. Now it's curious I did get the speed up of that fan again, but uh, our supply hasn't dropped so there's no correlation there. so it might have a temp sensor in it. that's why it looks like uh I based on one track I saw it looks like that power transistor driving the fan.

so um, yeah, it could. It could have a temp temp sensor and then just uh. Pwm's the fan and controls the spe of it and also the problem I was getting last time of this. AC DC coupling switch here, not driving the rear lay well yeah, you should be able to hear that that now works along with the one over here and we're not getting that overload thing we were getting before.

So clearly that uh 5V rail was just yeah, causing all sorts of issues. But we need to go through of course, systematically measure all the other uh rails. it's probably been on for like 5 minutes now. Haven't measured the temperature yet, but it's holding in there just fine as you'd expect.

So uh yeah, that's the biggest heat sink you could imagine on that uh originally fitted to that device, so it's obviously holding in there fine. There's no other overloads on the rail uh that we're aware of and we've got our relay back. There's still no relay action for the Uh filter here, but that could just be a MX uh to bypass. it may not actually be a relay in there for that.

and we're looking at an actual Uh tab temperature on the T 220 There of you know it's it's getting upwards of 70 or there about. So yeah, it's It's not preferable, but it's it's not that bad at all. So yeah, there's certainly no, uh, gross overload condition there. And the analog LM 317 there is, actually, uh, higher.

look at that 72.4 And if we go over here, we're looking it for the 337, the other rail, You'd expect it to be uh, pretty well matching. and yeah, I mean, jeez, that's not great at all. This is pretty piss poor thermal design in this thing. I mean if you designed this thing properly, it wouldn't even need a damn fan and we'll just measure our 5V rail there.

yeah, looks pretty clean, no worries. and uh, we can go in there and AC couple that if we want and have a look. 5 MTS per division? There we go. Not a drama whatsoever.

So that 5 rail is just running hunky dory and we'll just check the input voltage to the regulator, see if it's welld designed, or see if there's any filtering issues. and well, the filter looks reasonable. You know we got about a 1V Ripple on there and 2 Vols per division. We're looking at 2 468.

Well, it's peing around nine there. Well, that's a little bit High especially when you haven't designed in a proper heat sink in there. and that's why it's running at that temperature. I Would have you know, drop that down to about eight, but then that's not.

uh, counting for aging of the cap and uh uh, stuff like that as it loses, its as it dries out. With age and things like that, you'd expect to get more Ripple on there. but yeah, that's why it's running a bit hot, but it's certainly not. you know, crazy out in the ballpark, like 12 volts or something like that.

so that's not too bad. And we'll measure the Lm317 rails and these are really annoying cuz there's no voltage test points on this board at all. It's really damn annoying. You got to figure it out for yourself and you can't see the traces cuz they're on the bottom and had to buzz a few things out and use a bit of common sense.

and I Finally found the ground point on there and ah, it's just ridiculous. Anyway, that's our output voltage of the Lm317 and that's around about we're 5S per division. It's around about 152 volts or thereabouts. and the input voltage if I can get in there.

And yeah, there we go. There's our input voltage. so you know 5 10, 1520 Bo Jeez, look at that. That's really Overkill And the Ripple doesn't matter a rat.

But yeah, that's pretty gross overkill for a 15 volt rail. No wonder they're getting that thing up to 70. Unbelievable. We're getting exactly the same on the negative rail too.

Check it out. you know, round about that. 22, 24 Vol Something like that? Nuts. So unfortunately.

yeah. EV blog Curse again. It's incredibly simple. Nothing complex is having like a regulator would have been blowing at least something.

but no. Anyway, all it was was a missing heat sink. Why that heat sink was missing I Don't know. Absolute mystery.

Ridiculous. Didn't find it inside the box. I Don't know somebody's been playing around with this thing. It's ridiculous.

Anyway, Uh, we're getting our AC and DC coupling on there properly. We've measured all the voltage rails of plusus 15 volt rails on both analog boards. Uh, we measured the input to make sure that the Ripple is uh, well above the minimum Dropout voltage of the regulators. and uh yeah, pretty piss poor design in there.

Not that happy with that. uh at all the margins in there. maybe? um, they might be slightly different because I've got I'm running over 240 volts here in the lab. uh, like 245 or something nominal I think here in the lab.

So uh, quite a high Supply voltage, so mine might be on the high side of the margin that's for sure, or they certainly will be. But yeah, anyway, not that pleased with the uh, power supply thermal design of this thing. Pretty poor as I said, they could have done away with that fan mounted, designed the uh, uh, power supply properly in there, mounted on some big aluminium blocks, down to the shazzy or something. And really, this thing doesn't consume a huge amount of power, so you could easily get away with uh, putting those things to the shazzy and having this a completely fanless design.

It's just no excuse for it. It's just. ah, it's just poor form. it really is laziness.

So I'm going to have to do some more work. uh, testing this thing. but we can do. you know, the AC and DC coupling Works a treat.

Now it seems to the rear laser haven't actually measured it, but we can go up in the input gain notice the overload lead slightly starts to come. Come on at the maximum gain, we can actually uh make that go away. put a Terminator on the input here cuz we've got Channel a selected and bang it's gone and similar thing on B we can go over and we can test B look at that beautiful and we can go a minus B Here the differential input mode. we're still overloading.

We'll need a second Terminator on there. Ah, look at that. Bob's your uncle. What a Bobby Desler I Love it.

It's a winner anyway. I Hope you enjoyed that little follow-up video Even though sorry there wasn't much happening there at all. What a bummer Anyway, just a missing heat sink me hope you enjoyed it. Catch you next time.