#34 - Tektronix TDS 540A oscilloscope repair

hello today we are going to have a look at this Tektronix scope scopes of this line tds 500 and also 400 and 600 are known to fail because of leaky electrolytic capacitors I have already repaired the few such scopes before I started making videos and one of them I still have I used it in many previous videos it is the same model as this one 540 a another scope I repaired was 620 a which is 2 Giga samples per second but it had only 2 channels so I prefer to keep my 5 foot a which has 4 channels and one more reason was deep memory option which my 5:48 has and this scope also has deep memory option but it also has F of T option so if I am lucky with this repair I will have a chance to upgrade my scope I bought this scope on eBay for $200 it was sold by some sort of an electronics recycling company which is not far from my place so I picked it up myself and didn't pay for shipping now let's try to turn this thing on these scopes are quite noisy and they also take quite a while to boot up it's probably better not to turn it on so that not to make the problem worse but I just want to show you the symptoms so this constant beep is not normal there you go and boot it up so as you can see the symptom of the problem is here multiple self-test failures sometimes such scoob's don't even boot up and I tried repairing one of those but I didn't succeed the corrosion on the board was so bad that some traces were eaten away and I found a few but it still didn't boot up so after quite a bit of struggle I gave up and sold it for parts I'm hoping this one is not going to be so bad now let's have a look inside here is the back of the unit and options one F and one M are marked here one F is floppy drive one M is deep memory and here we have a sticker in version two option two F which is f of T here yeah so we have three balls to deal with here this main processor board on the top of the unit the acquisition board on the bottom of the unit and smaller board which is the front panel in this version we also have tiny board here for the printer port and the serial port it also has IC and a few electrolytic caps so here I see some corrosion around the switch next to this capacitor and I see liquid stuff from this capacitor on the board and here as well and check this out this capacitor is just soaked in this corrosive stuff which must have leaked from this cap here is the bottom and here we see the acquisition board and this smaller front-end board there are no capacitors electrolytic capacitors in this front-end board but a lot of them on the acquisition board and I see a really bad corrosion here can you see corrosion on the pins of this chip here and check this out this looks pretty bad here I removed the mainboard the acquisition board and this small interface board and the way to remove the front panel is this we need to gently pry this top open to release the plastic latches and then gently pull it off so now the top part is off and now we do the same thing on the bottom and gently pry off the latches there you go all right so we took off the frame so now it's really easy there you go and now we need to be careful with this there you go here is the front panel board here is the date on the acquisition board seems like 8 to 93 and on the main board here 920 to 93 here is some corrosion on the pins of these two big chips this is really ugly so we have 96 capacitors to replace yes we need to replace all of them and they are of two types 33 micro farad's 10 volts and 10 microfarads 35 volts here we have a couple of reels which I picked up on sale in a local electronics store for just 10 bucks per reel and this really is 10 microfarad 35 volt capacitors and that reel is 30 3 micro farad 10 volt capacitors so I will start with the acquisition board because the most of the capacitors are here I will remove the 30 3 micro farad ones first then thoroughly wash the board then soldered a new capacitors and then removed the other kind 10 microfarad ones then Warsh and them again and then finally solder 10 micro farad capacitors to remove the capacitors easily I use to filter in irons one is this wonderful soldering station which I always use and another one is this unregulated iron which I almost never use so I do it with both hands heating up capacitors from both sides at the same time and with some practice it's not too bad I'll show you for best results I try to scrape off corrosion from the southern joints and with some flux here we are I removed all 33 micro farad capacitors it seems like I miscounted before I had 47 and 19 here so 46 our sitting here and two are still soldered through this little board not convenient to the solder with this method perhaps I need to disorder this little board first which is really easy to do I have the soldering station with vacuum for desoldering through-hole components I don't remember how did I do this with my 548 scope which I repaired several years ago and I do remember about 620 a scope and I still have a photo of the acquisition board and the acquisition what the air is completely different and there is no such board on it at all and besides these two capacitors I need to clean the pads with this week and wash the board all right I clean the pads and disordered this little board as you can probably see there is some corrosion on it as well now it's time to give this board a really good cleaning the most drastic approach would be to start with a mild cleaner like this for example and then use distilled water to wash and rinse the board and probably use alcohol after that but I don't feel like doing this drastic thing right now I will just use alcohol and toothbrush to scrub it everywhere really well and we can always come back to the cleaner and distilled water approach later if using alcohol doesn't fix the problem the first round of cleaning is done I'm using compressed air to force the alcohol from under the chips so it can carry away all the in dissolved junk and I've done it several times I'm using 91% alcohol I could use pure alcohol with no water at all but I think it's better to have some water to dissolve whatever is not soluble in alcohol right I saw that the 33 micro farad capacitors 46 of them and now I'm going to remove 10 micro farad capacitors alright 10 micro farad capacitors are removed and the pads are cleaned ready to solder new ones the acquisition board is done and this little board is back in place all the boards are done except the small interface board and I put the boards back into the scope let's give it a try no luck so far one of the error messages is still there so it's time for more drastic measures let's check the operation really quick channel one connected to the probe compensation it shows the signal but it doesn't trigger at all let's and go to normal mode and here we are it doesn't trigger at all let's see if I can force yes I can force the trigger channel 2 let's change the trigger source to channel 2 and it doesn't trigger I can force it but it doesn't trigger at all same thing with channel 3 and channel 4 here is the error log I booted the scope several times so there are multiple messages here the latest message is about some parameter which was expected to be very low in this range from two and a half times ten to the minus ninth to 1.2 times 10 to the minus eighth but the actual value was 0 and above there is a message about DAC range test failure there is a DAC on the acquisition board to set certain voltage offsets and perhaps the trigger level as well so I turned the scope upside down here is this deck and I started checking voltages all around the place and I think I found the problem let's have a closer look so here not far from the deck there is a quad op-amp which is classic tl0 7/4 and I checked our rails on it and I think the negative rail is wrong so let's check the negative lead goes to the ground and on this capacitor I see the positive five volt rail which is a tiny bit low but not a problem and on this capacitor I see the negative 5 volt rail and on the op-amp pin four is the positive rail and it's there and pin 11 is the negative rail and it's not there and here is a ceramic capacitor right next to the op-amp and the side closest to the op-amp I see this negative rail and it seems to me that the trace going from this side of the cap under the chip and it's probably corroded they are not reaching pin 11 that's what I suspect so I opened my scope which is identical and checked and sure enough in my scope this side of the cap is connected to pin 11 and in this scope it's not connected you see we should here in the sound but there is no connection with some very careful so during the connection is restored we still have a problem but it started triggering I checked all four channels and the triggering is working so I continued checking voltages across capacitors and power rails on the op amps and I think I found two more problems one of the problems is here on this op amp pin 8 is the positive rail and there is no power here it is connected to this capacitor and I would have hard time figuring this one out if I didn't have a working scope so I checked the working scope and it turns out this should be connected to this spot there is a trace between this capacitor and this resistor and the via and in the working scope this spot is connected to pin 8 and there is no connection here and the second problem is around here pin 8 of this op amp it is connected to the positive 5 volt rail and I checked the powers present here and there is a trace to the positive pin of the capacitor but there is no connection and there is no voltage across the capacitor so it seems to me that the only consequence of this is that the bypass capacitor is not connected so it can lead to some noise on the 5 volt rail around here more careful so during this metal wire here and this one here check this out it passed the self test and it passed the signal path compensation test as well it seems like I got myself a working scope and by the way just out of curiosity let's check how bad these capacitors are we are at 1 kilohertz and let's set the second parameter - yes sir and let's start with the new 33 micro farad capacitor just for reference there you go about 30 micro farad's and 1.2 ohm zs sir and the capacitance is a little lower because of 1 kilohertz it's gonna be closer at lower frequency now let's try it 100 Hertz there you go 33.7 but for us our measurement it's better to have higher frequency so let's go back to 1 kilohertz and let's bring back the user and now for reference 10 micro farad capacitor also a new one there we go 8.2 micro farad's 5.7 obvious sir so now let's check a few random 30 3 micro farad capacitors which I removed from the acquisition board and there you go 1.6 nano farad and the sorry is out of scale completely dead let's try another one same thing a few nano farad and noise are completely dead let's try one more same thing and one more same thing and one more the same thing and let's try a few 10 microfarad capacitors exactly the same thing completely dead completely dead same thing I've seen much better one of the scoops I recapped had half of the capacitors almost fine maybe with slightly high ESR but these guys which he randomly checked so for her completely dead this little interface board is recapped as well these four chips on the acquisition board have running quite hot check this out the first one is running at about 95 or so degrees C and the next one I see ninety seven point nine and the next one I see hundred degrees C and the next one is a bit cooler for some reason about ninety degrees C and I suppose this is normal because it's exactly the same in the other scope which I had for several years and I hope it's gonna be a bit better in the closed case because now I disrupted the airflow I found another problem here we have 15 megahertz sine wave and the sampling rate now is 1 Giga samples per second and the time base is 50 nanoseconds per division it's a bit jittery here and the screen which is not normal but look at this if we change the time base to 20 nanoseconds per division and the sampling rate goes higher than the real time so now it is equivalent time sampling look what happens to the wave let's go even higher so now there's a parent that there are some gaps in sampling it seems to me that there is some problem with data pipes perhaps another broken trace somewhere and this is happening on all four channels a few weeks later I didn't have much time lately so I worked on this scope little by little during many evenings sorry I didn't document each step first of all I tried putting this CPU board into the working scope and I saw those gaps in the waveforms so it must be responsible for those gaps and second I tried putting this acquisition board into the working scope and with good CPU board and I also saw some problems first of all that jitter when this waveform was not quite stable we've seen that before and I saw some gaps in waveforms as well sometimes and sometimes self tests failed as well so I started looking into the acquisition board checking the traces and vias around the capacitors and I couldn't find anything for quite a while and then a miracle happened I found full schematic for this scope before I used this manual for TDs 520 B it has some circuit description and some schematics which was much better than nothing it gives some idea what's going on but these comedic SAR very different from 540 and then I found this topic regarding tds 644 a on the eevblog forum and here somebody posted this link to this site and among other things look at this full schematic for tds five forty five forty four and six twenty six forty six forty four wonderful so having the schematics I started checking the connections and voltages everywhere and I found a problem here pin three of this op amp must be connected to the ground but it was not so I fixed it by soldiering this little wire but it didn't fix the problem so I checked more and found one no problem here there is a via under this op amp and it was broken so I restored the connection by running a little wire from this pin seven over the edge of the board to the via on the bottom here it is after fixing these two problems I never saw self test failures and waveforms were stable so I started checking the CPU board I spent quite some time checking the CPU board testing the traces and vias around the capacitors randomly checking different data bases I used the work in a Scylla scope to look at the waveforms on the memory chips and such I couldn't find anything wrong and then it occurred to me to run the signal path compensation and the gaps in waveforms disappeared so it seems to me that those gaps in waveforms were the result of running the signal path compensation before when the triggering was not working properly so the lesson from this is if you fix something on the acquisition board run the signal path compensation before doing anything else so now everything looks nice and stable no jitter no gaps and I am so glad I managed to repair the scope having full schematics certainly helped a lot one more thing I would like to do these to Dallas chips have batteries built into them and they are certainly pushing limits of the runtime by now and as I understand this chip has a real-time clock and no critical parameters but this chip certainly has some critical configuration parameters options and calibration and such so if battery fails in that chip the scope probably won't even boot up but I don't really want to replace these chips these are expensive and who knows probably will work for many more years so what I've done with my other scope i disordered this chip made the backup and soldered the socket in its place and put the chip into the circuit so I would like to do the same thing here there you go quick and easy having a good disordering station the circuit is in place I am using this programmer which is mini Pro or TL 866 CS and I need to use this old laptop with windows on it to run the software the chip is GS 1650 Y and the programmer does not support 1650 but as I understand the scope uses this chip as 1250 and the difference as I understand it is that 1650 has several partitions or pages which are not used in this scope so we can read it and me back up as 1250 I read the chip and then run verification just to make sure that reading twice results in the same thing the chip is back in the circuit and I am thinking about putting the backup file on some sort of an old small and useless for other purposes memory card so I can attach it here with the tape or something so even if I decided to sell the scope and the battery failed it would be a relatively simple fix the card is attached with a label printed on a label printer very handy tool highly recommended the scope is back together and check out the F of T here is a service manual for this model and there are no schematics in this one but there are performance tests here they are and a lot of equipment is needed here most of which I don't have so I had to limit my tests to whatever is possible will be 15 megahertz generator plus DC offsets and such so it's hard to be sure that every obscure feature works fine but whatever I checked so far looked okay I hope you enjoyed this repair video if you did give it a thumbs up and consider subscribing to the channel I have a pile of projects waiting thanks for watching bye

Channel: FeedbackLoop

Views: 41,963

Rating: 4.9658117 out of 5

Keywords: electronics repair, diy electronics

Id: 7V0LCL4mL-8

Length: 38min 39sec (2319 seconds)

Published: Tue Dec 29 2015