Strange 7603 fault


Morris Odell
 

Hi all,

I have a 7603 that has developed a frustrating fault. It works perfectly from cold for about 5 minutes and then goes dead. The trace blurs and fades out and nothing works at all, even the graticule lights. That makes me think it's a power supply problem, possible temperature sensitive. Once the scope is left alone for half an hour or so it's OK again for another 5 minutes.Has anyone had a similar experience or have any idea where to start looking? Any advice gratefully received!

Thanks,

Morris


Tom Lee
 

The best way is to measure the various supply voltages and see which ones are failing. That, and a little reasoning, will give you a good idea where to look.

--Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 1/23/2022 21:49, Morris Odell wrote:
Hi all,

I have a 7603 that has developed a frustrating fault. It works perfectly from cold for about 5 minutes and then goes dead. The trace blurs and fades out and nothing works at all, even the graticule lights. That makes me think it's a power supply problem, possible temperature sensitive. Once the scope is left alone for half an hour or so it's OK again for another 5 minutes.Has anyone had a similar experience or have any idea where to start looking? Any advice gratefully received!

Thanks,

Morris




 

On Mon, Jan 24, 2022 at 06:49 AM, Morris Odell wrote:


Has anyone had a similar experience or have any idea where to start looking?
A common problem that might produce this kind of behavior is that the soldering of the collector pins of the holders
for the power transistors becomes bad. It is usually clearly visible as a crack in the solder around the pin. The reason
was bad choice of surface treatment of the pins. Later modified collector pins with a different surface treatment
and longer pins became available.

/Håkan


Clark Foley
 

Hakan’s assessment is the best place to start. As I recall, the soldered nutblocks that held those power transistors became excessively hot when using spectrum analyzers or the 7D01 logic analyzer. Prolonged or use in poor ventilation could lead to solder failure as well. The change in plating helped initially to get better soldering at the factory. Many were assembled with insufficient heat the get a good solder joint. Even if used with low stress, heat cycling over several years produced the same failure. Ultimately a fan was added to reduce the stress on the entire system.
Inspect the solder joints for the nutblocks. These are hex standoffs with no. 6 thread at one end and solder pin at the other. If you give the nutblock a light twist using your fingers, you’ll know. A failure will twist easily. We had some the fall out on there own. You will need a very high-powered soldering iron re-solder them. Take care to heat the block directly; apply the iron to the pin, not the board. Flowing solder into the plated hole will not be sufficient to heat the block without damaging the board.
By the way, does your 7603 have a fan?


Jim Ford
 

May have mentioned it before, but I had a similar problem with a Wavetek 166 function generator.  Turned out to be a poor connection on a power supply connector.  In my case, the solution was bending the pin to press harder onto the socket contact.  While you may have a cap or transistor issue, it could be as simple as a connector problem.                       Jim Ford Sent from my T-Mobile 4G LTE Device

-------- Original message --------From: Morris Odell <vilgotch1@gmail.com> Date: 1/23/22 9:49 PM (GMT-08:00) To: TekScopes@groups.io Subject: [TekScopes] Strange 7603 fault Hi all,I have a 7603 that has developed a frustrating fault. It works perfectly from cold for about 5 minutes and then goes dead. The trace blurs and fades out and nothing works at all, even the graticule lights. That makes me think it's a power supply problem, possible temperature sensitive. Once the scope is left alone for half an hour or so it's OK again for another 5 minutes.Has anyone had a similar experience or have any idea where to start looking? Any advice gratefully received!Thanks,Morris


Clark Foley
 

One small error on my part. The nutblock is not hex standoff. It is more of a square with beveled corners. The Tek common parts catalog Terminal, Stud. Part numbers 131-0847-00, 131-0847-03.


Morris Odell
 

My sincere thanks to all who replied on this thread. I'll get stuck into the scope and report on my findings in due course. Since my original post my "spare" 7603 has expired as well with signs of a power supply failure so it's going to be a double effort.

Morris


Morris Odell
 

To my great relief the problem with the "spare" 7603 turned out to be one of the plugins. Scorch marks on the side cover were a dead giveaway :-O It was a 7A26 in which a tantalum cap self destructed in a spectacular and very stinky fashion. Fortunately there was very little collateral damage. I took the opportunity to replace all 4 on the main amplifier board and now all is OK.

The original 7603 problem still awaits!

Morris


Morris Odell
 

Clark Foley wrote:

By the way, does your 7603 have a fan?

No it doesn't.

I've opened it up and sucked out the dust. Some of the solder joints holding the filter caps look dodgy almost as if they had been resoldered in the past. The nutblocks holding the power transistors look OK superficially but I'll re-do them anyway as suggested and give it a try.

You have to admire the mechanical design of that scope. It's very elegant.

Morris


Jim Ford
 

Great!  A good reminder for those of us with multiple plug-ins to check for smokers in ours.     Jim Ford Sent from my T-Mobile 4G LTE Device

-------- Original message --------From: Morris Odell <vilgotch1@gmail.com> Date: 1/25/22 10:33 PM (GMT-08:00) To: TekScopes@groups.io Subject: Re: [TekScopes] Strange 7603 fault To my great relief the problem with the "spare" 7603 turned out to be one of the plugins. Scorch marks on the side cover were a dead giveaway :-O It was a 7A26 in which a tantalum cap self destructed in a spectacular and very stinky fashion. Fortunately there was very little collateral damage. I took the opportunity to replace all 4 on the main amplifier board and now all is OK.The original 7603 problem still awaits!Morris


Morris Odell
 

Well I've made a little progress. I cleaned and re-soldered the low voltage power supply connections as suggested and the voltages are all rock solid. However the problem is still there - it runs normally for a few minutes and fades out. I got out my HV probe and checked the HV test point. When it was running it tested normal, but when the fade out happened the HV died.

So it's a HV problem. The HV test point is connected to the CRT cathode and the supply comes from a half wave diode rectifier CR1232 which is then smoothed by a pi network consisting of C1232 (0.01), R1234 (20K) and C1234 (0.01). Tomorrow I'll open up the HV compartment and look into those components.

Any suggestions on troubleshooting the HVPS are most welcome!

Morris


Mark Vincent
 

Morris,

Replace C1222 with a high temperature low ESR type. Also go ahead and replace the two 1mfd types with a film type. Using a 250V here is fine if you have them. The first one, C1222, is possibly going bad or is already bad. See if these help or fix the problem. It is possible the transformer is acting like many in the 500 series in getting leaky windings. I hope only the electrolytics are the problem instead of a bad transformer.

Another thing that can cause problems are the connectors to the pins on the boards. Cleaning the pins and moving the connectors up and down a couple of times each may solve problems. The same is true for the transistors and ICs. I am not saying you do have dirty pins/sockets. You will have to check yours to see if you have any that need cleaning.

Mark


Morris Odell
 

Thanks Mark,

In the 7603 you need to do quite a complex procedure to remove the entire HV unit before you can work on it. I'll start on that later today. Because it's so complex, I'll replace whatever I can before putting it all back together. I don't think the problem is the transformer as it occurs suddenly and does not creep up gradually as in the 500 series. I suspect a faulty capacitor.

Morris


Mark Vincent
 

Morris,

I know the high voltage supply area is hard to get to. I have a 7603. Doing all the work at once is a good idea. The reason I mentioned the two 1mfd electrolytics is because of the difficulty in that area to work on. Once the board is out enough, it not too hard to work on. Someone recently said to use 35V in this circuit. A good idea since the voltage measured was near 25V. The 47mfd is not critical, if you have a 100mfd, that will work fine. If you have an ESR meter, that would be a good test to see if the ones you pull out are high to open. The reason I said to use film for the 1mfd is that they are not going to dry out and the ESR is almost zero. While you are in there, check other parts to see if any need replacing. Other models are had to get to in some areas so doing all at once then testing is best. I was thinking your high voltage failure was like or similar to the other type of transformers in that it would drop down instead of suddenly. You cleared that up for me.

Mark


tinkera123
 

Hi Morris,
I have nil experience working with the HV supply (no Probe and a general nervousness around high V's ... :) ) and these Scopes in general .... however, unless you are certain it is a HV problem I would be having a look at the Current Sensing resistors on the LV supplies. The HV is run off the +15V supply and dependent on other LV supplies ... so if some other fault is turning off one of these LV supplies, they generally all shutdown, hence no HV.
--
Cheers,
Ian


C G
 

Hi,

If you reconnect the HV +15V supply harmonica, beware of the PCB error in some series showing the pin 1 triangle at the wrong place, resulting in F814 fuse on the rectifier board to blow up. In the service manual, the photograph Fig.8-10 shows its right place, near the R1193 adjustable resistor.

A and B connections between HV and Z axis are crossed. This not clear and sometimes messed up: check the schematic.

If the HV switchmode transistors don't oscillate :
- check their drivers before suspecting that the TO-3 (Q1216 & 1218) don't have insufficient gain as sometimes evoked on the net (they were hometaxial geometry and low gain was normal compared to epitaxial); if they're dead, it is possible to swap them with epitaxial ones, providing you adapt their polarization (use higher voltage for Vce, as 2N3055 that may be borderline)
- check their DC base polarisation at startup with respect to local ground (emitter - triangle ground)
- check the ESR of C1222
- you may also desolder both the cathode of the HV diode CR1232 and pin 11 of T1225 off the HV multiplier block U1230 to eliminate the failure of the latter.

Best,
Christian

Le 28 janv. 2022 à 00:39, tinkera123 <iacbell@outlook.com> a écrit :

Hi Morris,
I have nil experience working with the HV supply (no Probe and a general nervousness around high V's ... :) ) and these Scopes in general .... however, unless you are certain it is a HV problem I would be having a look at the Current Sensing resistors on the LV supplies. The HV is run off the +15V supply and dependent on other LV supplies ... so if some other fault is turning off one of these LV supplies, they generally all shutdown, hence no HV.
--
Cheers,
Ian





C G
 

Hi,

If you reconnect the HV +15V supply harmonica, beware of the PCB error in some series showing the pin 1 triangle at the wrong place, resulting in F814 fuse on the rectifier board to blow up. In the service manual, the photograph Fig.8-10 shows its right place, near the R1193 adjustable resistor.

A and B connections between HV and Z axis are crossed. This not clear and sometimes messed up: check the schematic.

If the HV switchmode transistors don't oscillate :
- check their drivers before suspecting that the TO-3 (Q1216 & 1218) don't have insufficient gain as sometimes evoked on the net (they were hometaxial geometry and low gain was normal compared to epitaxial); if they're dead, it is possible to swap them with epitaxial ones, providing you adapt their polarization (use higher voltage for Vce, as 2N3055 that may be borderline)
- check their DC base polarisation at startup with respect to local ground (emitter - triangle ground)
- check the ESR of C1222
- you may also desolder both the cathode of the HV diode CR1232 and pin 11 of T1225 off the HV multiplier block U1230 to eliminate the failure of the latter.

Best,
Christian


Morris Odell
 

Thanks to all for the suggestions. The situation so far is:

1. The HV supply runs from the unregulated +15 and the LV supplies are all in spec during the HV failure so I'm sure the LV supplies are not the problem. The fuse to the HV supply is OK too.
2. All the small transistors test OK and remain so even when running warm in the curve tracer. I haven't tested the two power transistors yet.
3. The HV board is the later model with two little DC restorer cards piggy backed on it. Everything looks OK
4. In this board one of the 1 uF caps has been replaced with a 2.2 uF. The circuit is a little different and those two caps can have as much as 130 volts across them.
5. All the electrolytic caps have normal ESR except the 1 uF which is high. I don't think that would cause the supply to fail though.
6. I'm going to replace all the electrolytics today and see what happens.
7. When I disassembled it and discharged the tripler lead there was no spark - could this be a sign of tripler failure?

More news as it breaks!

Morris


Morris Odell
 

Well no success. I went through the complex disassembly procedure to get to the HV board. Replaced all the electrolytics and tested all the transistors on a 575 curve tracer. They were all OK even after warming up for a while. Put it all back together and the fault is still there. It runs perfectly for a few minutes and then the HV dies. After a cool-off period all is well for another few minutes. The HV board looks perfect.

I'm not sure what to do now. I could replace the HV rectifier diode but the original is so nicely installed with an anti-corona coating that it would be a shame to disturb it. The tripler also looks normal but if it's prone to failure I don't know where I could find a replacement.

I do have another 7603 to use but I really would like to fix this one rather than keep it as a parts unit.

Morris


 

Morris,

Switching to the other 7603 and keeping this one for parts is a good intermediate solution, since it leaves the faulty instrument unmolested (and gives the replacement instrument some exercise). It also gives you time to think about how the faulty instrument might be repaired. Can you use the 575 to test the HV rectifier and the tripler diodes? You've checked everything else, which means that extra suspicion should fall on those two components.

-- Jeff Dutky