Solartron 7150 check and a small repair

[h.bashar]

Hi, earlier this week I scored a crusty old Solartron 7150 multi-meter, this is my first 6.5 digit multi-meter so naturally I was very excited about it, when it got delivered the meter case has a missing handle and part of the case is broken, quickly plugged it in and it booted successfully. As many of you know the LCD display back-light in these meters had four yellow LEDs that are dead after many years of service, so those got replaced with bright blue ones.

This was my initial diagnosis of the problem which I will keep below just for future reference but it was wrong and I will explain further down:

I had a problem with testing the meter because I didn't have any accurate voltage source, so I hooked up a 9v battery to the front test leads and while I think it measured it to 9v, the LS digits where not stable and had a negative trend, I tried the resistor range with 0.1% 10K Vishay resistor ( I think it was 25ppm) and the same problem occurred, so I opened the meter and checked the voltages on the power supply regulators and they all checked  out as per the service manual, the only  problem I encountered was when I tried to measure the output voltage across C311 (the +15V rail) which is a 1uF tantalum capacitor, while I was placing the positive lead of the meter to the positive marking on the capacitor and the negative lead to capacitor negative I measured -15V, so I was confused so I re-checked the test meter leads and they were hooked the right way so something fishy was going on, I consulted the service manual for diagrams and also the PCB marking under this capacitor and it all pointed out that this capacitor was soldered the wrong way around!
Imagine in all these years someone was using this meter not knowing about this fault, I am surprised it passed any calibration, even the 15v regulator IC was running hot, considering it’s a one amp regulator supplying  opamps. So I disconnected one capacitor leg and checked its value with my DE-5000 LCR meter and lucky enough it was still in working order. I disordered the other capacitor leg and put it back the right way around this time and tested the meter again and this problem has disappeared. I don’t think this meter has been repaired before but it’s surprising that such an error passed unnoticed all these years.
There is still some work to be done for the meter as the mains input filter needs replacing, many dry joints on the PCBs and a calibration. I hope this post will be helpful for someone.


There was nothing wrong with C311 except that the marking on the capacitor was printed in reverse showing the wrong polarity , which I didn't catch and as a result I put it back the wrong way around based on the reversed silkscreen print , now I am pretty certain that I fried it and it will need replacement which will be done in due time, I will be changing this cap and check the rest as well for miss- labelling and report my findings.
For all people who helped me with this thank you so much 


Updates:
++de-soldered C311 and tested with LCR meter and its still functional see picture WP_20160820_001 (Custom)
++ went over with LCR on the remaining polarized CAPS on the board two were toast C305, C306 (100uF) on the 25V rail (pictures WP_20160820_015 (Custom) and WP_20160820_016 (Custom))
C302(220uF) is fishy (WP_20160820_017 (Custom))
++Rest of CAPS were fine but at this point I lost all faith so all have been de-soldered and pads cleaned waiting for new replacements.

New Update:
CAPS were all replaced with Vishay ASM series and new mains entry module was installed  , everything looks in working order now 


Best wishes[/size][/color]
Hasanain

[Prehistoricman]

Did the cap appear to be soldered in that polarity from factory, or did it look as if somebody tampered with it?

[h.bashar]

As far as I can tell yes , evident by the dried solder joints, what surprised me further that it didn't leak, open or short in all those years, trust me I didn't expect to find that thing in such a high quality multimeter.

[h.bashar]

The 100uF tantalum are similar to this from RS:
http://uk.rs-online.com/web/p/tantalum-capacitors/5381714/
i am not sure about their behaviour when the polarity is reversed other what they taught us in school (dooms day and all), the data sheet for this one says it can take 0.35V reverse voltage at 125degC so this one was taking 15V reverse for God know how many years of operation, and it still checked Okay by the LCR meter! wonderful!

[Prehistoricman]

It's a miracle! Would this unit be hand-soldered? Maybe it's human error.

I noticed something interesting. I put up a post today about my switch-mode bench supply and uploaded pictures. Those pictures were taken from a Windows Phone, and cropped with a tool that adds (Custom) after the name. I look at your file names and... wow!

[h.bashar]

I know right  , some of the parts must be hand soldered, so probably a human error. yes its an old Lumia920 WP and yes it was a free tool from the net , how is it going with power supply repair? opto-couplers tend to fail trust me I know because I repaired some equipment before that had an issue with TL113, not a SMPS but it failed because of age. Now put that in a closed power control loop and the magic white smoke will escape.

[Prehistoricman]

Opto-couplers? But they're just a light and receiver... wouldn't have ever thought to check them. Now I love magic smoke... but only in shit I don't want!   

I'm realising I have a fairly large lack of tools as far as debugging/repairing goes. I don't have a scope or freq. generator, only a clamp meter.

[h.bashar]

You are right, but the problem with any LED it fails after some time, if you look for example LED data-sheet it will specify life time in hours, now I wont say I know much about SMPS failures because I tend to steer away from non-isolated mains in general, of-course it will be much easier if you had an oscilloscope to see the change in PWM frequency as you load the power supply. If you are going  to keep playing with this thing (which can easily kill you) I suggest an isolation transformer and heavy precautions specially around the mains parts of the supply and make sure the voltage on the input filter CAP is bled by the bleeding resistor each time you approach that part. As for a frequency generator it all depends on what sort off frequency you are after, an old 555 timers can work wonders so try that if you want. Finally I will stay away from using inductive loads to test a SMPS, unless they are designed for it, try to approach the problem with a resistive load first and see if you can approach the stated current of the device without problems.

[Jwalling]

Tantalum caps that are installed backwards are like time bombs with an unknown fuse length. I've seen them explode almost instantly, and I know of situations where it took months, setting fire to circuit boards that had been in the field for quite awhile. In the latter case, a company I worked for had a memory board (ISA bus 4MB DRAM memory card) that had the silk screen polarity marked incorrectly for one of them. We had calls for many months after the ship date to the support team about systems that caught fire.

Nasty things, I don't know why they're still in use...

Jay

[h.bashar]

Hi Jay, I am just surprised it took about 30 years until I discovered it, and it still working!

[Prehistoricman]

@h.bashar

So far, I've only operated the supply with its earthed and insulated case on. Both the mains caps have their own bleeding resistors, thankfully. If not, I'd install some.
I've just ordered a really cheap kit for a component tester and frequency generator.
The inductor was the first component on my desk that I'd used many many times before as a load, as a spark generator, and a brick that absorbs heat. I've got some resistance wire that I'll use in the future though.

@Jwalling
Do tantalums have a certain niche? I can't see them being so dangerous and used without reason.

[h.bashar]

Very interesting read on Tantalums  in https://en.wikipedia.org/wiki/Tantalum_capacitor, basically high capacitance per unit volume so physically smaller than aluminium caps. Also there is a section about reverse voltage that reads:

"Tantalum electrolytic are polarized and generally require anode electrode voltage to be positive relative to the cathode voltage.

With a reverse voltage applied, a reverse leakage current flows in very small areas of microcracks or other defects across the dielectric layer to the anode of the electrolytic capacitor. Although the current may only be a few microamps, it represents a very high localized current density which can cause a tiny hot-spot. This can cause some conversion of amorphous tantalum pentoxide to the more conductive crystalline form. When a high current is available, this effect can avalanche and the capacitor may become a total short.

Nevertheless, tantalum electrolytic capacitors can withstand for short instants a reverse voltage for a limited number of cycles. The most common guidelines for tantalum reverse voltage are:

10% of rated voltage to a maximum of 1 V at 25 °C,
3% of rated voltage to a maximum of 0.5 V at 85 °C,
1% of rated voltage to a maximum of 0.1 V at 125 °C.
These guidelines apply for short excursion and should never be used to determine the maximum reverse voltage under which a capacitor can be used permanently"

So what the hell is wrong with my capacitor still working after all this abuse! Go figure!

[picburner]

This thing intrigued me so I opened my 7150 (mine is newer than yours, Solartron changed designator's names, and I could verify that C161 (your C311)
is properly placed and the tensions across it is right.
From the photos you've done seems that the manufacturer has printed the label on your capacitor C311 incorrectly rather.
In fact, C308, C309 and C310 have the right label (+sign) while C311 not.
At this point I think you've put C311 now in the wrong way....

[Jwalling]

@h.bashar

So far, I've only operated the supply with its earthed and insulated case on. Both the mains caps have their own bleeding resistors, thankfully. If not, I'd install some.
I've just ordered a really cheap kit for a component tester and frequency generator.
The inductor was the first component on my desk that I'd used many many times before as a load, as a spark generator, and a brick that absorbs heat. I've got some resistance wire that I'll use in the future though.

@Jwalling
Do tantalums have a certain niche? I can't see them being so dangerous and used without reason.

There may have been good reasons in the past, but with all the advances in cap technology (high value ceramics and solid polymer types for example) since then, I don't see a good reason to use them in modern equipment.  Maybe someone else would know better?

Jay

[Jwalling]

Hi Jay, I am just surprised it took about 30 years until I discovered it, and it still working!

Me too. Must be a world record. Given that the particular rail is supplied a low current three terminal linear regulator, maybe it got used to being abused? 

I wonder if it would blow up if correctly connected to a high current source?
Would make an interesting experiment.

Jay

[h.bashar]

@picburner, the silkscreen is the correct way and the current placement of the capacitor as well, how do I know? using continuity test the CAP+ is now connected to output of the +15V regulator (IC301) and the CAP- is connected to ROME (0V GND) and to + of CAPS C308 and (C309 which you can see in the picture on the top layer photo of my board) and finally it measures the right polarity voltage across C311 when testing with the multimeter so I am pretty sure it was the wrong way around. In fact the orientation of capacitors in both of our pictures are the same 

[kutte]

This thing intrigued me so I opened my 7150 (mine is newer than yours, Solartron changed designator's names, and I could verify that C161 (your C311)
is properly placed and the tensions across it is right.
From the photos you've done seems that the manufacturer has printed the label on your capacitor C311 incorrectly rather.
In fact, C308, C309 and C310 have the right label (+sign) while C311 not.
At this point I think you've put C311 now in the wrong way....

just opened my 7150 to see it looks like picburner's device. So I agree with him: your c311 has been mislabeled and is is now put in the wrong way.
Kutte

[Jwalling]

This thing intrigued me so I opened my 7150 (mine is newer than yours, Solartron changed designator's names, and I could verify that C161 (your C311)
is properly placed and the tensions across it is right.
From the photos you've done seems that the manufacturer has printed the label on your capacitor C311 incorrectly rather.
In fact, C308, C309 and C310 have the right label (+sign) while C311 not.
At this point I think you've put C311 now in the wrong way....

Good catch! 
In any case, I would discard that cap if it's been powered up while in backwards.

Jay

[Kleinstein]

If in doubt with the polarity one could do a test with a non polarized capacitor (e.g. MKS should be similar size) and than measure the actual voltage at the capacitor. There might be a voltage higher than the normal positve supply, e.g. if the cap is part of a charge pump.

[h.bashar]

@Jay, Kutte, my mod picture look exactly the same as picburner picture, the capacitor was the wrong way around, nothing wrong with the silk screen, it looks the same on both photos!, it might be that you are pointing to the one with red arrow, that one is how I found it and the last picture is what you should be looking at.

[Jwalling]

@Jay, Kutte, my mod picture look exactly the same as picburner picture, the capacitor was the wrong way around, nothing wrong with the silk screen, it looks the same on both photos!, it might be that you are pointing to the one with red arrow, that one is how I found it and the last picture is what you should be looking at.

Well, all I know is that the the picture titled WP_20160818_016 (Custom).jpg the C311 cap itself  is marked incorrectly - picburner is right.

Jay

[h.bashar]

Dear Jay, the picture you just referenced is how the capacitor was originally found, i.e in the wrong position, the correct orientation is in the last picture of the first post. The silk screen in both cases is the same my picture and picburner picture unless I am missing something.

[Gertjan]

Hi h.bashar,

I think you misread the posts of Jwalling and kutte.....
They are not referring to the print on the silk-screen, but the print on the capacitor itself.

Just compare your C301 with the same capacitors next to it. Look at the shape of the housing, and where the plus mark is printed....

regards, Gertjan.

[Jwalling]

Hi h.bashar,

I think you misread the posts of Jwalling and kutte.....
They are not referring to the print on the silk-screen, but the print on the capacitor itself.

Just compare your C301 with the same capacitors next to it. Look at the shape of the housing, and where the plus mark is printed....

regards, Gertjan.

Correct!

Jay

[h.bashar]

I can't believe I didn't notice this , you guys are absolutely correct, I am sorry for dragging you in a wild goose chase  , I will be editing the first post to reflect this finding and thanks again.