Repairing a BWD 521 cathode ray Oscilloscope - no traces, hissing noise

[joshAU]

Hi all.
I recently purchased a (obviously) second hand CRO oscilloscope.
It worked fine for a couple of days, I was just getting used to it.

Then one day, I left it on for 10 mins or so with both traces active on screen.
A while later, I heard this high frequency hiss, so I switched it off.
The traces were still active when I switched it off.

Now, when i switch it on, I get no traces whatsoever, although the power switch turns on, power light turns on.
After a minute or so, which is probably the standard warm up time before you'd normally see the traces, I can hear the hiss again.

I removed the side cover, and the noise seems to be coming from the EHT supply board.
Checking further, there is a 2N3055 NPN power transistor that appears to be the source of the noise.
Furthermore, it quickly gets too hot to touch!

I've checked all the resistors on the EHT supply board, and all seem OK.
I've checked the voltages on the board, and many seem lower than expected.
I didn't have any equipment to test the 3000v, -1080 or -1000 lines, but here are the other voltages I measured.

reference | actual
-11.5 | -7.7
-50 | -44.5

18 | 16.5
50 | 47
165 | 85

Now, I realise many people may suggest just replacing it with another scope - and probably a digitial one, but for what I want to do with it, it has to be analog.
I can either try to repair this unit, or take another punt on buying another second hand CRO.

So, I'd like to try to fix it.

I suspect the 2N3055 is the culprit, but I am concerned its behaviour could be caused by a fault or short in the flyback transformer in the EHT supply, or possibly even the tube itself.

I haven't yet discharged the tube and cleaned the connection from the flyback transformer cable - I have read that sometimes in some CRT devices this fixes the hiss, but given the heat generated in the 2N3055 I'm thinking more likely a short in the transformer...

Does anyone have any suggestions of where to start, or any ideas that may help?
I did happen to get a brand new (30+ years old) tube as a spare with this unit, so if it is the tube, that's not such a problem.

The service manual for the unit is here:
https://elektrotanya.com/bwd_electronics_bwd-521_oscilloscope.pdf/download.html

The relevant part of the schematic from the manual I've attached.

If you require any further information, please let me know.

Thank you.

 

[spec]

HI josh,
Welcome to ETO

Your BWD 520 scope symptoms are classic. The hiss you hear is due to a high voltage high frequency ark-over. This will place an extra load on the 2N3055 inverter transistor which will cause it to get hotter than normal. Your scope has a relatively low EHT at 3KV, which means that fault finding/repair should be easier.

The thing to do is to expose as much of the EHT (Extra High Tension) circuitry as you can. Then turn the scope on in a completely dark room and, hopefully, you will see a blue corona somewhere. This will give you an indication of the area of the problem.

EHT circuits are badly designed on some scopes- I had an expensive Gould scope that would ark over from time to time.

As you imply, dirt can cause arking: an unpleasant greasy black gunge builds up over the years

In any event, once you have done the test in the dark, clean all areas involved in the high voltage and inverter circuits (this is a wise precaution with any used scope). Simple cleaning will often cure the fault. Pay particular attention to the EHT lead and the rubber capped plug that normally connects to the tube side near to the face. The plug can just be pulled out, but be gentle. The area around the hole in the tube where the EHT connector fits should also be cleaned.

It is unlikely that the 2N3055 is faulty because it controls the inverter which generates the EHT and, as you have a hiss, the EHT must be there. Thus the 2N3055 must be operating. It is unlikely that the tube is faulty either.

The most likely faults are:
(1) Damp
(2) Dirt
(3) Dry/corroded joint
(4) Insulating screen dislodged/ corroded
(5) Component dislodged and is too near to the chassis or another component.
(4) Cracked printed circuit board
(5) Faulty EHT cable/tube connector
(6) Faulty inverter transformer
(7) Faulty capacitor, especially the high voltage capacitors- this is a common fault. The capacitors tend to crack.

Good luck with your fault finding and don't even think of discarding your scope- a scope is for life

spec

 

[atferrari]

No reccomendation about dealing with high voltages? Any residually something still charged?

Not that I know anything about that. But fearing a lot....

 

[spec]

No reccomendation about dealing with high voltages? Any residually something still charged?

Not that I know anything about that. But fearing a lot....

Yes, well noted atferrari.

josh,when the scope in turned on never touch the scope with the covers off as some lethal voltages will be exposed.
Be aware that the tube and the capacitors can hold their charge even after the mains supply has been turned off.

spec

 

[joshAU]

Hi Spec, and thanks for the welcome, and for your detailed advice.
Thanks also for the warning atferrari and spec re high voltages.
I was aware of the dangers, but you never know, so thanks again.

After discharging the tube, and avoiding going near any caps, I removed the EHT board from its enclosure.
I found a broken/loose solder joint from that board to a ground connector, which I resoldered.
I also cleaned the EHT connector on the tube.

Now, when I power it up, I get absolutely no hiss.
I still don't get anything on screen either.

I decided to reseat the EHT connector on the tube again, but I noticed that the tube didn't seem to have ANY charge in it, despite having had the unit turned on immediately prior.
ALSO, I didn't get any spark from it the first time I tried to short it to the chassis.

I remeasured the voltages on the points I did before, and noticed that the voltages all seem to have increased, and now seem normal, except for the +165V which reads 89.3 volts - still way too low, but a bit better than 85v.
eg, -11.5 was reading -7.7 now reads -11.7
-50 was reading -44.7 now reads -49.2
18 was reading 16.5 now reads 18.1
50 was reading 47 now reads 49.6

I'm guessing the arcing was causing the voltage drops.
The +165 volt that is at 89.3 is obviously too low, and I think it is the only voltage I can safely measure on that side of the transformer.
For some reason, despite being clearly labelled on the pcb board as +165, the circuit diagram doesn't mention a + 165v point?
It does however mention a +150v point in around about the right place.
Its a bit confusing however, as I think there is a typo in the manual.
The point +165, as printed on the board, connects directly to the diode labelled D15 - but there is no D15 on the diagram, so I think it must actually be D19 on the diagram, as there are only 4 diodes listed on the diagram - D16, 17, 18 and 19, where printed on the board its D15, 16, 17, 18.
Anyway, the +150v point after the diode D19, as listed in the schematic, is the one that reads only 89.3.


I'm not sure if this extra info helps in diagnosing, but thought I'd mention it all just in case.

I haven't had a chance to test it at night for arcing... although I'm guessing that may not be relevant as it doesn't seem to arcing anymore, but I'll try tonight.
Given the poor solder on the earth connection, I think I'll try resoldering a few more points on the EHT board today, and check any connections in case I'm missing something.

Thanks again for your help.
Any other suggestions greatly appreciated.

 

[joshAU]

Just to work through your suggested likely faults spec:
(1) Damp
(2) Dirt
(3) Dry/corroded joint
(4) Insulating screen dislodged/ corroded
(5) Component dislodged and is too near to the chassis or another component.
(4) Cracked printed circuit board
(5) Faulty EHT cable/tube connector
(6) Faulty inverter transformer
(7) Faulty capacitor, especially the high voltage capacitors- this is a common fault. The capacitors tend to crack.

1. damp - its in my livingroom, which doesn't suffer from damp, and the heaters often going, so it should be dry air.
There's no evidence of any signs of moisture, but I guess I cannot completely rule it out.
Might try a hairdryer on the insides to rule it out completely.

2. Dirt. yes, well, I cleaned out the system, and the EHT connector on the tube - which made the hiss disappear.

3. Bad joints - that's what I'm going to check next.

4. Insulating screen dislodged/corroded. How do I tell?

5. Component dislodged and too near chassis.
I've checked and it all looks ok.

4. I've inspected both sides of the pcb board for obvious signs of cracks - looks good.

5. faulty EHT cable/connector. I checked with a dmm and it shows no resistance.

6. faulty inverter transformer - This I am beginning to suspect - as the only voltage I can safely read is only 89.3, when it should be either 150 or 165 volts, depending on whether you use the manual or the printing on the pcb - located below D19 on the schematic (D15 on the board).
I might have to invest in a high voltage probe to check the -1000, -1080 and 3000 volt lines.
The fact that the tube doesnt seem to be getting charged makes me suspect the transformer, or perhaps the diodes after it - D15 & 16, if using the printing on the circuit board, or D19 & D16 from the schematic.

7. faulty capacitor.
I don't have an ESR meter to test them with, but i may try swapping out C78 & C79 - both 1000v 330uf.

Is there an easy way to test the Diodes labelled D16 & D19 on the schematic?
The parts list show them as 15KV 2.5mA diodes. part number BY140, which, from reading, seem to be a diode string - there are 10 diodes or more.
Testing with a multimeter shows infinite resistance both ways, so I'm assuming I can't test it the same as a standard diode.

 

[joshAU]

Another update.
While I no longer hear any hissing... I've noticed that the 2n3055 is still getting as hot as before - ie. almost burning my skin when I touch it.
I'm thinking a short in the transformer? - but I'm just guessing.
Any ideas?
Thanks again.

 

[spec]

Hy josh,

You seem to be making some progress with your scope.

OK can you do this:

(1) Remove
(1.1) C75
(1.2) C87
(1.3) D19
(2) Clean PCB which was under these components
(3) Inspect PCB to ensure that it is not burnt/ carbonized anywhere.
(4) Connect C17 cathode (left) to D16 anode (left) CORRECTION Should read, Connect D17 cathode (left) ....

(5) Connect your meter to the 150V point
(6) Turn scope on.
(7) Does the 150V point read around 150V?
(8) Is the 2N3055 still getting hot?
(9) Does the tube display a trace? You may have to turn the brightness up and even then the trace may still be very faint and it may be out of focus.

spec

​

 

[spec]

About the BY140 rectifier diodes here are the main characteristics as far as I can tell:
(1) approximate year of manufacture around the 1970 mark
(2) Peak Inverse Voltage (PIV) 15KV
(3) Maximum forward current (IF) 2.5mA (a scope typically has around 150 micro Amps screen (post deflection anode [PDA]) current at full brightness.
(4) Maximum forward voltage (VF) 50V

These old EHT diodes have always been a bit weird but they generally work quite well. If you need to replace the diodes a modern type would be superior.

They are easy to test but you would need a 60V power supply which would probably be difficult for you to get. Also 60V is above the generally accepted safe voltage of 30V, so there is a danger.

What you can try though is a 9V battery in series with a 4K7 resistor and your DMM set to around 10mA. Connect the battery both ways and see what happens. With any luck with the positive of the battery connected to the anode of the diode you will get a small current flowing and the other way around you should get no current flowing. If you do get any current flowing in the reverse direction the diode is faulty.
A better test would be to connect three nine volt batteries in series, to make 27V, with a 15K or so resistor in series with the diode and multimeter, if you had the batteries that is. You won't damage the batteries or flatten then so you could borrow them from other equipment.

You could do a more meaningful test with seven 9V batteries in series to give 63V but there is a danger. In this case you should have a resistor around 33K in series. For a good diode, you would probably get a reading of 330micro Amps in the forward direction and virtually zero in the reverse direction.

Incidentally, you can check the capacitors for leakage/shorts this way too, but adjust the voltage so as not to exceed the voltage of the capacitors and with electrolytic capacitors only put a positive voltage on to the positive of the capacitor.

spec

spec

 

[joshAU]

Hi again spec.
Yes, some progress.
Thanks again for your help.
I'm just about to remove the components as you suggested, but wanted to check one thing first.

You refer to removing C87 - I'm guessing you mean C81 - the cap right near C75?
Its hard to tell I know from the text in the manual, but its labelled C81 on the board, so I assume its the correct one.
I've gone ahead and removed it.
No sign of burnt marks or damaged traces under them.
The PCB boards are translucent, so its fairly easy to inspect them for damage, but I've double checked closely with a torch and it looks good.

I'm just about to connect C17 cathode (left) to D16 anode (left), but I can't seem to find C17?
Is it on the EHT supply board, or on another board?

Also, when you say to connect them, do I need to remove the C17 cathode and D16 anode from the board first, then connect them, or just run a wire between them as they are?

Sorry for the additional questions, but I prefer to ask now than discover later.

 

[spec]

Hi again spec.
Yes, some progress.
Thanks again for your help.
I'm just about to remove the components as you suggested, but wanted to check one thing first.

You refer to removing C87 - I'm guessing you mean C81 - the cap right near C75?
Its hard to tell I know from the text in the manual, but its labelled C81 on the board, so I assume its the correct one.
I've gone ahead and removed it.
No sign of burnt marks or damaged traces under them.
The PCB boards are translucent, so its fairly easy to inspect them for damage, but I've double checked closely with a torch and it looks good.

I'm just about to connect C17 cathode (left) to D16 anode (left), but I can't seem to find C17?
Is it on the EHT supply board, or on another board?

Also, when you say to connect them, do I need to remove the C17 cathode and D16 anode from the board first, then connect them, or just run a wire between them as they are?

Sorry for the additional questions, but I prefer to ask now than discover later.

Hi josh,

I have just woken up (9:37 am UK BST).
The mods I suggest are shown in red on the attached picture. The components to be removed are circled in red and the wire link is also shown in red.

Basically, I am suggesting that the the three voltage doubler components to be removed from the circuit and preferably physically from the PCB, so that their effect on the 2N3055 inverter can be determined.

With the voltage doubler removed the EHT on the tube PDA will be only half what it should be but this may be sufficient to get a visible trace on the screen.

No wonder you were not sure what I was rabbiting about. I made an error (it was 4am) in post #8 which has now been corrected in red. I make mistakes all the time though.

No problems about asking questions- ask away. By the way I like scopes.

spec

​

 

[joshAU]

Hi spec.
Ah, different time zones...that's easy to forget.
I just had dinner. Nearly 8 pm here.

I can't seem to see any attached pic?


oh, just read the "To be completed" bit... that's probably why.

 

[joshAU]

Ha ha!
I'm amazed that you didn't make more errors given it was 4am.
No worries at all - I too make plenty of errors on a regular basis, and not just around 4am.

Its funny though, some of my most productive times have also been well after midnight...

OK, so I've removed the components, and have just run the wire between.

Powered on, no smoke, but also no traces, despite turning graticule up and the intensity on both traces up to max and adjusting the vertical shift on both channels fully up and down.
The 2n3055 is still getting too hot to touch.

Still getting 88.5volts or therabouts on the +150/+165 point.

Just in case it means anything, the tube does get a faint glow at the rear of the shaft, as it did previously as well, so no change, but absolutely nothing on screen.
Just to be sure I went through the manual for the initial setup in case I missed anything, but all seems good.

I'm guessing that its looking more like the transformer?
I really hope not... as I've never seen one like it before, its pretty weird looking, and I doubt its a standard, easy to replace part.

One a side note, and completely unrelated to this problem, is I think I may have fixed another issue I had with the channel A trace - I found a broken solder joint and a component hanging loose in the A compartment while checking this issue, which is what I suspect was causing the A trace to be set too low vertically.

Thanks again for all your help.

 

[spec]

Ha ha!
I'm amazed that you didn't make more errors given it was 4am.
No worries at all - I too make plenty of errors on a regular basis, and not just around 4am.

Its funny though, some of my most productive times have also been well after midnight...

OK, so I've removed the components, and have just run the wire between.

Powered on, no smoke, but also no traces, despite turning graticule up and the intensity on both traces up to max and adjusting the vertical shift on both channels fully up and down.
The 2n3055 is still getting too hot to touch.

Still getting 88.5volts or therabouts on the +150/+165 point.

Just in case it means anything, the tube does get a faint glow at the rear of the shaft, as it did previously as well, so no change, but absolutely nothing on screen.
Just to be sure I went through the manual for the initial setup in case I missed anything, but all seems good.

I'm guessing that its looking more like the transformer?
I really hope not... as I've never seen one like it before, its pretty weird looking, and I doubt its a standard, easy to replace part.

One a side note, and completely unrelated to this problem, is I think I may have fixed another issue I had with the channel A trace - I found a broken solder joint and a component hanging loose in the A compartment while checking this issue, which is what I suspect was causing the A trace to be set too low vertically.

Thanks again for all your help.

No probs about help josh.

OK, that test shows that the voltage doubler components are not causing an excessive load on the 2N3055.
I will have a look at the manual to see where the 150V comes from, or do you know.
If you like, best return the circuit back to normal.

I am not that sure that the transformer is causing the problem.

Can you confirm that the inverter is oscillating- be handy if you had a scope
If the inverter is not oscillating:
Is it easy to pull the 2N3055 and do you have a replacement or a power transistor with a similar specification?
Or do you know how to test a transistor?

The other two transistors associated with the 2N3055 are also suspect. Do you have any electronic spares?

spec

 

[joshAU]

Thanks again spec, I appreciate it.

Yes, it seems like it is something else, but glad to hear that you aren't sure its the transformer.

Hmm, the only scope I have available is the one in question.

I should be able to desolder the 2N3055 easily enough, its only two pins, with the casing being the collector, held on with a couple of screws through the board.
Although I don't have a replacement, I can probably order one in.
I should be able to test it, its a larger beast than I've come across before, but I assume the testing is the same for any other pnp transistor?
I've had a bit of practice testing transistors, although not for a few years.
I'm guessing I'll have to desolder it first to check it properly.

The other two transistors I don't have spare, but I can hopefully source them easily enough.

 

[spec]

Thanks again spec, I appreciate it.

Yes, it seems like it is something else, but glad to hear that you aren't sure its the transformer.

Hmm, the only scope I have available is the one in question.

I should be able to desolder the 2N3055 easily enough, its only two pins, with the casing being the collector, held on with a couple of screws through the board.
Although I don't have a replacement, I can probably order one in.
I should be able to test it, its a larger beast than I've come across before, but I assume the testing is the same for any other pnp transistor?
I've had a bit of practice testing transistors, although not for a few years.
I'm guessing I'll have to desolder it first to check it properly.

The other two transistors I don't have spare, but I can hopefully source them easily enough.

Hy josh,

Is it the case that the 2N3055 is not mounted on a heat sink?

You should be able to test the 2N3055 power transistor like a small signal transistor but that is not always the case. It depends on the leakage current of the particular 2N3055.

This is not a very scientific approach, but it is rather difficult to fault-find remotely, especially as you don't have a bag of spare parts, but I suggest removing all three transistors in the inverter and testing them in isolation.

Let me know how you get on.

spec

 

[spec]

Few more tests while the 2N3055 is removed:
(1) transformer T2 primary coil upper measure resistance. Should be 31 Ohms
(2) Transformer T2 primary coil lower measure resistance. Should be 15 Ohms

spec

 

[joshAU]

Ahhrrrg... I just typed a heap and lost it hitting the wrong key.

OK, attempt two...

I'll have a look over them and report back.
I'll desolder all three transistors and give them a test, and do the other tests while the 2N3o55 is removed.
It may take a while.... work, etc resuming after the Easter break tomorrow, hopefully tomorrow night.

There is a picture of the transistors in the manual, just after the parts list and before the circuit diagrams.
The 2N3055 is huge in comparison to the other two.
It has a much larger metal casing, more than an inch across, with 2 bolts that go through the pcb board with nuts on the other side.
I think the housing itself is the heatsink, given the size of it.
There does appear to be a bit of brown discolouration around it on the pcb, like its been overheating for a while, although it may have just started since I lost the traces and it began arcing....

Anyway, I have to get to bed. Work tomorrow.
Thanks again spec.
Speak soon.

 

[spec]

Ahhrrrg... I just typed a heap and lost it hitting the wrong key.

OK, attempt two...

I'll have a look over them and report back.
I'll desolder all three transistors and give them a test, and do the other tests while the 2N3o55 is removed.
It may take a while.... work, etc resuming after the Easter break tomorrow, hopefully tomorrow night.

There is a picture of the transistors in the manual, just after the parts list and before the circuit diagrams.
The 2N3055 is huge in comparison to the other two.
It has a much larger metal casing, more than an inch across, with 2 bolts that go through the pcb board with nuts on the other side.
I think the housing itself is the heatsink, given the size of it.
There does appear to be a bit of brown discolouration around it on the pcb, like its been overheating for a while, although it may have just started since I lost the traces and it began arcing....

Anyway, I have to get to bed. Work tomorrow.
Thanks again spec.
Speak soon.

Good idea- get some sleep.
spec

 

[joshAU]

Hi again spec.

I've had a look at a few replacement options for the 2N3055, but just wanted to ask your advice as to which on to buy.
If I search for the part number 2N3055, virtually everything I find is a NPN transistor, but the manual clearly states, but it could be wrong of course, given the other errors I've seen in it, that Q42 is a PNP transistor.
The parts list shows it as -60v 1A 30W PNP part number TIP30A

Searching on ebay, if I search for the words 2N3055 and PNP - it only finds 6 sellers, selling 2N3055 NPN paired with MJ2955 PNP transistors.
If I search with NPN instead it finds 62 results, all of which are NPN.

going on a wikepedia search, not always the most reliable source I know, it looks like it certainly is a NPN:
https://en.wikipedia.org/wiki/2N3055

Do you think the manual just got it wrong?

I checked a site I regularly buy components from - rscomponents, again using 2N3055 as the search term, it only finds two NPN transistors.
eg:
https://au.rs-online.com/web/p/bipolar-transistors/5452210/
https://au.rs-online.com/web/p/bipolar-transistors/7743212/

Both are NPN, but have differing specs,
ie,
Maximum Operating Frequency is 2.5MHz for the 2N3055G as opposed to the 2N3055AG, which is 1MHz.
Also, the Maximum Collector Emitter Saturation Voltage is 3Volts and 5 Volts respectively.

BUT, assuming the manual did actually get it wrong, would either of those to transistors be a suitable replacement?

Its all rather confusing I must say.
Speak soon.
Josh