TSP #108 - Teardown & Repair of an Agilent E3632A DC Power Supply

hi welcome to the single path in this episode I got a repair with you for you guys I'm going to take a look at this HP e3 632 a this is a 15 volt 7 amp or 30 volt 4 M power supply I've done a repair of one of these a slightly different model that goes up to 50 volts but this one has a completely different issue it doesn't power on at all so no matter what it's plugged in it will do nothing and this is a fairly old unit probably has thousands of hours of usage on it so I'm curious to see what's the status of the components inside and if we can figure out how to fix it so I started disassembling it and I thought let's check the fuse also to see if the fuse is an issue and we can go ahead and try and remove this fuse there we go let's see what do we have here if I can get it out there it is and yep the fuse is definitely dead now you may be tempted to simply replace the fuse and that's something we're going to do but we have to find out why the fuse died and and this if this is a surge issue or if this is a problem internally and you can see how early it is around the fan it's been on for a very long time so we'll go ahead and replace the fuse but before plugging it into 110 volts we're connected to the isolation transformer and bring it up slowly and then we can monitor the current and low voltages but we'll still be operating in AC so the transformer impedance would be correctly taken account into account at 60 Hertz and we can figure out if the problem persists or not so that should be a good place to start so let me find it a placement fuse and we'll go from there and here as a replacement fuse exactly the same spec gonna plug it in and hook it up to the titration transformer alright here we go let's give it a try so I've already turned this down all the way to the minimum and turn this on so it's at zero volts that I'm gonna enable the output so it's now isolation transformer protected and now let's increase the voltage slowly and see what happens remember that the fuse itself is a forum fuse so we don't expect a lot of current especially at low voltages so let's go ahead and see what happens so I'm going to slowly increase this and then so we're at 10 volts 12 so you know what would have been helpful actually turning the power supply on here we go turn the power supply on you know try again I'm already drawing some current yeah see I met so I'm here sitting at ten volts and I'm already drawing one amp and that would be ten watts so no that is no good if this power supply is drawing one amp at 11 volt AC then it's going to be destroyed once by the time we go to 115 volts and what will happen is the fuse just going to simply die again and there's another advantage of having an isolation transformer now you may be tempted to think that the voltage is going to drop or the currents going to drop when we go up but remember this is not a switching power supply this is a linear power supply it's a laboratory power supply so it isn't it going to be one-to-one for most of the range it's gonna have a transformer in it and a rectifier and capacitors and so on just like in any regular linear power supply would have no DC DC convertors also have to have that the front stage but in some cases because the DC DC converter has no activity at lower voltages you will not see anything but if you see this right away at the very least we have a problem either in a transformer or we have a problem directly on the rectifier section so let's go ahead and take it apart and see what we can find in there there's definitely an issue there's a couple of things we can do to try and diagnose it one step at a time alright and here's a look inside of the unit now my girlfriend's helping me doing this repairs and she's really impatient even more than I am and she just wants to take this apart so I'm gonna go over this fairly quickly now we find everything we expect to find inside this power supply we have a big heat sink where our regulators and rectifiers and everything is mounted on this is a single channel power supply so he only has one of these sections with a fan blowing directly across it now all the AC voltage is going into the main circuit which then gets rectified and voltage regulated sorry that's my that's my cat's tail in the way and I'll be are gonna come from this transformer so this transformer has only a few points of contact to the main PCB and you can see that coming out of here these big wires here are going to be the main outputs which are coming from the transform which are going to get rectified and likely they're not gonna be more than about 40 volts or so because this only goes to 30 volts and any difference between those two voltage is going to be waste in the regulators anyway and then there's a couple of other wires coming out going into some other section there and I'm going to disconnect all of these from the main PCB and what this will do is that it will make this transformer isolated from the rest of the circuit and we can find that if there's short inside the transformer or not if the transformer is not shorted and the voltage is coming out of here a reasonable and we can crank it up to 115 volts and it should be no problem it should consume no current nor real power then the problem is on the PCB and then we can turn our attention to that it's better if the promise on the PCB than in here because if this transformer is dead let's say thermally has finally failed somewhere inside then it will be I don't have a replacement so we cannot fix that so we would have to find a replacement for it that would be difficult so let's go ahead and do this one step at a time remove these plug it back in crank up the current and see what happens and see if we can figure out something from there okay so here we have it this is the first connection from the transformer to the PCB I've removed that and here are the four cables that were coming out of the transformer also removed from the PCB so now there is no connection from the transformer if the PCB now there is still some connection from the AC line to the PCB which goes into it and for voltage selection and perhaps something else I'm not quite sure yet but either way the transformer being error the circuit should give us at least some indication so let's go ahead and try that we're gonna go and turn this on again and let's see what we get so here we go so zero volts zero amps and go back to ten yeah I can see I'm getting nothing this time I can crank it up oh I hear the fan coming on there we go look at that so here's 115 volts we are burning 18 watts and the fan is on so the fan is not powered from the transform which is interesting so there's some circuitry in there but that circuitry seems to be working because we have hundred fifteen volts the fuses not burning these lines are now disconnected so now what we need to do is to go back to the circuit and take a look and see where these guys go and trace from there to see if we can find a particular sort or a problem on the PCB and that would be our next step so here's the back of the PCB now the four points for the transformer contact the PCB are these four points one two three and four and you can clearly see how wide the traces that are connecting to these are because this is the main part of the circuit where the current is flowing so you want very little resistance on those traces and kind of roughly I traced them out where they go we have two of them here and two of them here and obviously one of them is going to this component here and there was a faint trace on the other side which looks like goes also to the other pin now we have four pins here with a wire jumping on this side and this one two three back here are capacitors and you can see holes underneath them for pressure relief in case something goes wrong there so these four capacitors the filter capacitors after voltage rectifier are sitting on the other side of this PCB there so we have our capacitor we have a rectifier so we can go ahead and start doing some measurements see if we can find an issue between these various terminals I'm going to measure between all of these and see if we have a short or a very low resistance somewhere and then we can find out what component could be potentially responsible for that okay let's do some measurements so we are in short detection there so let's measure between these two terminals of the transformer so from here to here do we have a short nope no short no short let's move up nope still good move up ah there we go look at that I'm gonna put in resistance mode because it's gonna be otherwise be extremely annoying and I'm gonna measure it just to make sure that I'm not measuring a capacitor large capacitor and let's see nope you can see the resistance is not moving so I'm not charging a capacitor there's indeed a short circuit there and looking at this we can see that he goes to this component so let's verify that it is the case so this pin is connected to this pin we can see that zero ohm and the pin next to it I suspect is connected to the pin up here and indeed it is which means there's a short between these two and yep you can see the short between those two so this has four pins and this is most certainly a rectifier given that it's output these are connected to the capacitor so the rectifier is a shortness and that obviously is a catastrophic failure of the rectifier the question is is the rectifier the only issue but we don't know that yet because this could be a cascade failure but I would have to get to the rectifier remove it and even put a temporary rectifier in this place and see if it comes back to I have to completely disassemble this and I will take a couple of minutes I think and we can go from that and take a look and I suspect is very very dirty inside so we're gonna have to do some cleaning but it should be pretty straightforward and here we go we took it apart and it's very very dirty and to make matters worse the component that's damaged it's quite inaccessible first of all look at how nasty it's become when the inside are gonna have to completely clean this up so the components on the other side remember it was right here and that is right there sitting behind this heatsink so it's sitting behind these two capacitor so if I want to access that and remove it which we will have to remove it we'll have to take these capacitors off so that I can put a screw in screwdriver in and unscrew it from the heatsink the alternative would be to take the entire heatsink off with all the other components that are on it and I think that's probably more work given that these are elastic together so yeah so but there's not much not much else to say right now just clean this up remove the capacitors get a view of what that component is again likely a rectifier and then remove it and then see if we can find a replacement or something to temporarily replace it wait so we can turn the power supply on and figure it out and everything else is pretty straightforward so let's do some cleaning and some removal all right here is the capacitor remove now we have access to the rectifier and take a close look at it and we don't even need to really examine it many further because it's very clear what has happened if this camera would actually focus there it is you can see there's a big crack in it so it's completely broken most likely a thermal issue or aging eventually has completely failed and you can even see a bit of mark on the heatsink from the damage so this is this is done and the what we need to do is to remove it and see if I either have a replacement for it and if I don't have a replacement for it we'll have to maybe improvise and put something in its place that we can still attached to the heatsink so let me take it out take a closer look and see how it's broken and then we'll see what we can do about it so here it is I took the component out and as you can see it's definitely damaged and I was looking for a replacement because I didn't want to wait for digi-key to deliver one and I happen to have this power supply here and you know what it is is to have exactly the same component in it so that's pretty convenient so I took it out of this one I'll replace this one later so we can go and put this one back in our power supply hook it up to the transformer and see if we see any difference in its behavior and whether it would come back online and if it has any other issue we will be able to find out and here we are reassembled and ready to try again so we're going to put it back into the case because that's really the easiest way to connect all the cables again and then we can see what happens okay everything is back together again and I've done this for safety as well as making sure that these connections all reach to where they're supposed to be so I've connected everything back again and we can go ahead and try it so let's go and turn on the isolation transformer again and slowly crank up the voltage and see what we see so it's already turned on I'm gonna turn on the trench isolation transformer so so good it's zero volt I'm gonna crank it up before it used to go crazy there we go that's pretty good 22 volts it's not drawing any excessive current that's good let's increase and the fan is coming on oh and he just made a beeping noise that's good so it did boot so let's now restart it so it does a proper startup let's try again we go and one more time and only one beep that's not a good thing and yep they're all CD or the vacuum fluorescent display it doesn't work it doesn't come on but the buttons work and that's a good sign so there are two ways this displays fail either they fail because a microprocessor surround is failed or because the vacuum fluorescent display has failed or the driver for the display has failed when the processor fails is tough to fix because you need to have an exact replacement of that part because it's the software is burnt into the rom of that processor and I've done a replacement of that before and somebody sent me a replacement component it's not something you can buy but if the VFD is broken we might be able to find a way to solve it now the fact that this buttons work mean the microprocessor is working because it's responsible for controlling the buttons and also the screen or the VFD screen so let's go and see if any voltage comes out of it because when I turn it on and I rotate the knob I hear clicking noises which means is trying to adjust a voltage that's a good sign so let's go ahead and try that let me turn it back off so we start from a known state so these power supplies when they power on they power on in the off state so we can take advantage of that let's plug that in and let's put that multimeter somewhere here let's say and I'm going to put this one like this I don't want to make sure I don't show anything out let's take it for this one now I mean just a camera here and let's see if you get anything let me turn this on so I'm going to power it on then all we're going to do is going to turn the output on and rotate and then see what happens and hope for that turn it on let's see come on fluke you can do it there we go so let's give it a try it'll hold in my hand just to make sure it doesn't touch anything and here we go so nothing comes out that's normal let's turn it on still nothing and one volt there you go look at that two three should go all the way to 15 there it is all the way fifteen this is his first range and if I switch the 30 volt range I should be able to continue and I can and I should be able to go to over 30 they're there so it works perfectly fine just this screen is not working disable the output enable the output yep looks good to me so now next step is take this part apart and see what's going on inside of it maybe we can fix it up going from here so I buy and I took it out and I tried 100 thing and take a look turn the light off turn this on and it's actually there it's just really really dim which is consistent with how old this unit in how long it's been left on and turned on and by the variation in the intensity of these you can see that you know it's been sitting at some setting that has had those particular digits really aged yeah so there's really not much we can do about that except to replace it because VF T's when they age they get dim like this I don't know of any way of restoring their brightness so now I gotta find out if there is actually a way to replace this and see if someone on eBay or somewhere has a replacement part because it needs to match for this particular model because it has a 15 volt and a 30 volt segment here so it needs those two to be able to be lit up it can't just say any any random number and I'm gonna go a couple of search see if I can find something for it so I was looking around I came across an interesting article where somebody was discussing a potential way of restoring some life into this vacuum fluorescent displays so one of the aging methods is that the filaments that run across the display which are the source of the electrons get oxidization on them and that prevents electrons from being fired once they're hot through the mesh and ultimately hitting the phosphorous screens at the back so some people saying if you drive these for you know of higher voltage and they're normally designed for getting them glowing red that some of the oxidization might break off and a display might get restore a restoration on its brightness to some extent and that's one thing we can try I mean what we have to lose here the only thing is that I don't want to have to remove it so I traced out the pins that are connected to the filament and they are fairly independent of the rest of the circuitry one side of it seems to go only to one of the pins so I'm going to connect the negative voltage the positive voltage there and ramp it up until it grows red and then see we do that a couple of times and bring it back hook it up see if it's bright again if it is that would be absolutely brilliant don't have to do anything so then what's the worst that can happen will this display will get damaged but the one thing that would be concerned is to make sure that there is no other component connected to to those two voltages because you don't want to damage that so that's what I'm going to try it let's see what happens so let's do some measurements on the filament so I've marked the two pins that go to the filament the issue is that we want to make sure that those pins go nowhere else in this record so that when you want to overdrive we don't damage some other component and on the display and I don't think they're connected to anything so let's go ahead and turn this on let's measure the thermal voltage the thermal voltage is going to be an AC voltage I'm going to measure from here which is one of the pins of the filament to the other pin of the filament without touching anything else and there you go we have about 6 volts AC on it and there is indeed no DC voltage on it I can put it on to low impedance mode there can verify this ourselves and there you go I can see six volt AC and no DC voltage that's that's comforting and let's turn that off and go to resistance mode and measure the resistance of the filament and here's our filament resistance we're looking at oh I'm measuring the right thing oh I have to disconnect it what am i doing let me disconnect it from the main supply so we don't interfere with the coils of the transformer and let's measure the filament filament about 10 ohms okay good so now we have some informations a 10 ohm filament six volts roughly across it so that that in Mexico I clear about 600 milli amp is the current AC that is just being pushed into it so what I want to do is I want to do it basically the same thing I want to drive it to the same way and then raise it up in a dark environment where I can see them glowing and will maybe go to 200% I'm not sure it's really a toss there and then make them glow and then try it again and see what happens I'm very curious if it will come back to life okay here is my setup to do this now one thing I mentioned in the previous segment was that I had a resistance and I had the voltage that was coming through it so I estimated the current but remember that's the startup corn just not the current when it's running because when the filaments get hot the resistance goes up and the current collapses so the amount of crime that's going through it is not six hundred milliamps when it's warmed up it's much less than that so you have to be careful not to try and put six hundred milliamps through it you might actually damage it so you know what I've done is I put a 25 amp hour resistor in series with it because my isolation transformer isn't so precise to give me very small steps and I don't want to crank it up all of a sudden you know 220 volts and destroy it so this will at least absorb some of the voltage and I have my multimeter here in AC in series with the filament so we can actually find out if what is the current that's going through it so we can slowly crank it up and see if the filament glows and at the same time measure the current so let me refocus the camera so we can have our eye on the filament and see how bright you want to go and then we'll try it back on the unit okay here we go let's turn the lights off Alexa turn off laboratory lights and here we go let's give it a try and it slowly crank up this voltage and see what happens and there we go and run 30 milliamps going through and look at it I don't know if you can see it or not but they're already glowing you can see let me turn off another light that's in the lab and we should be able to get a better idea of the glowing there it is yep quite nice and bright now I'm gonna keep increasing this until it's well I don't know actually when to stop so let's see what happens here we go it's much brighter than it was before 300 million let's keep it there for a few seconds and the car is not changing that much which is a good sign okay let's see maybe a little bit higher and go back down there we go let's do this one more time up to 300 and down and a third time for luck 330 and keep it there for a bit so yeah this is definitely beyond specification obviously never grow close like this in the system so it should be helping hopefully burning off the oxidization layer and okay let's give it a try I'm excited to see what happens well it didn't work so I went back and I tried it with more current and of course I destroyed it so there goes that the filaments are dead and one of them just broke off apparently you're supposed to get a white heart and I did exactly that and of course I went a little too far and it broke so that is supposed to fix the issue if the filament is oxidized but if the phosphor has aged that would not fix that and I don't know which case it was here but either way it was an utter failure so what I'm going to do is I have to take it off and we have to now find a replacement filament for it of replacement VFD there was really no other way because even what I had before is no longer there so yep that's it giant leap backwards but that's okay that's what happens when you try it hopefully you if you try it with some success let me know but I couldn't get it to work so now I'm gonna look for a replacement well we're gonna had an ordered one replacement screen so I did manage to find one on eBay and it was about $90 which is quite expensive really for what you're getting but this is what happens when parts are rare and difficult to find so now we have to replace it with the other one lots of unsoldering of the pins and replacing I wonder how bright this is looks brand new we'll see what happens and here we replaced the old display with this one and you can clearly see that there is a difference between the digits on the camera you should be able to see that the new displays did use a much brighter and they had they have much wider surface to them so I'm curious if this means that it will perform better I also cleaned up to the unit quite a lot so we can put it back together and see what happens all right it's all put back together let's give it a try and tada looks great look how bright it is and you can see it's actually the brightest vfu display I have in the lab and apparently this was new so maybe $90 is not that unreasonable for something that's fairly difficult to find and you can see that it does say 15 volts and 30 volts which are the correct labels for this particular model because you can get this obviously with many different combinations now the question is does it perform and does it turn on and does it sync to write them on a current we can do a quick test on that with my electronic load okay let's turn it on and when I turn it on I see what - 4 - 5 millivolt and that's normal because this is not being calibrated probably never even calibrated so it's going to be a little bit off there's a very simple procedure for calling this out and I think I've done that in a different video so go ahead and so you can find that video so let's go ahead and crank this up to the maximum voltage it can do on the 7 amp range so 15 volts the most it can do we can see we're really 15.1 - it's pretty close but I would definitely recall this if I were to use it for anything precision now 7 amps I've already set this to sync 7 amps as you can see so we can go ahead and turn it on and there we go you can see that it's syncing 7 amps and it's maintaining 15 volts very nicely so it has no problems there whatsoever and we can go to 30 for 30 volts range and as soon as I do that it can no longer do seven so it can only do four amps and the voltage collapse is because the electronic load is requesting too much so let's go ahead and change the current for that to 4 amps and at 4 M so you can see we jump back to 15 volt so it's definitely working and thank you push that's very helpful and we can continue this 230 volts and you should be able to do about 4 amps there and it is indeed doing that and the AB measurement is actually pretty accurate so very good I would say this is a successful repair and quite exciting something that cost about $90 to fix is not too bad for a power supply like this which is quite expensive now again if you liked this video please give a thumbs up and as always these are possible because of your patreon support and with this because of you guys that they can continue to doing is interesting repairs I hope you enjoyed this nice and simple relaxing one I've got some pretty difficult instruments coming up that I think you're gonna like so I'll see you in the comment section

Channel: The Signal Path

Views: 33,722

Rating: 4.9684153 out of 5

Keywords: Power Supply, Rectifier, VFD Display, Repair, Teardown, Fuse, Agilent, Keysight, Isolation Transformer, Auto Transformer, Linear Power Supply

Id: 9ubOfTHRngc

Length: 25min 40sec (1540 seconds)

Published: Mon Sep 11 2017