A little help/guidance on fixing an old oscilloscope

Thread Starter

Tealc

Joined Jun 30, 2011
140
So got the new oscope.

Here's some of the waveforms. With TR814 and TR815 still disconnected.

This before connecting, an unloaded signal.


On wires 13, 17 and 16.



On wires 12, 14 and 15 (14 is marked as 11 in the schematic).



I don't really understand what the four primaries (or are they centre taps) are doing in the circuit but the top half is totally loading down my signal to nothing while the bottom half is not affecting it so much.

I should point out that this is with the secondaries all attached to their respective points. Is it worth disconnecting them?


This is the oscillation on Position 13. It's a lot slower than I expected. The amplitude is around what I thought it was from last weekend's experiments with that USB scope thing that worked poorly.



And position 14



It's stable like this and only wobbles a very small amount. Nothing is heating up.

Swapped in a new (probably old stock) MJE520 at TR813 (to replace the doofah BD435) and there's no change to oscillation speed or amplitude.
 
Last edited:

ian field

Joined Oct 27, 2012
6,536

ian field

Joined Oct 27, 2012
6,536
Well 330Ω is 51 milliamps, which should be plenty to drive the inverter / transformer primary, so the bsic LV rail provided by Tr813 appears to be working. 17 volts seems a tad low - I guessed 18 volts, but not significantly.

I'm sorry to say we seem to be working towards a transformer fault, but keep trying tomorrow as you said.
Direct link to PDF:


http://members.ozemail.com.au/~bobpar/k7205.pdf
 

Thread Starter

Tealc

Joined Jun 30, 2011
140
Yes Ian thanks.

I've bought some 4015s, some BC328, and should be able to cobble the rest together from my parts bins and will be building this tomorrow, probably just onto a stripboard as I haven't quite perfected toner transfer yet.

The only cap I'm struggling with is 0.047uF MKT but I do have a few of 0.1uF MKT so I wonder if they are suitable. Will build up with them anyway and see how it goes.
 
Last edited:

ian field

Joined Oct 27, 2012
6,536
Yes Ian thanks.

I've bought some 4015s, some BC328, and should be able to cobble the rest together from my parts bins and will be building this tomorrow, probably just onto a stripboard as I haven't quite perfected toner transfer yet.

The only cap I'm struggling with is 0.047uF MKT but I do have loads of 0.1uF MKT so I wonder if they are suitable. Will build up with them anyway and see how it goes.
2 equal capacitors in series is half the value.

2 equal capacitors in parallel is twice the value.
 

Thread Starter

Tealc

Joined Jun 30, 2011
140
Yes I had thought of that, thanks Ian. Not sure if I have quite that many (8 required in that case). The bypass C7 I'm sure can be changed to a single 0.1uF.

Will have a dig as I know I stripped down quite a few junk PCBs from work which used 0.1uF MKT as bypass but they were quite tricky to get off the board without damaging them.
 

ian field

Joined Oct 27, 2012
6,536
Yes I had thought of that, thanks Ian. Not sure if I have quite that many (8 required in that case). The bypass C7 I'm sure can be changed to a single 0.1uF.

Will have a dig as I know I stripped down quite a few junk PCBs from work which used 0.1uF MKT as bypass but they were quite tricky to get off the board without damaging them.
I'd try the correct values of whatever type I had to hand - you can always order the right type if it doesn't work.
 

Thread Starter

Tealc

Joined Jun 30, 2011
140
I've ordered the right type in the meantime but should be able to at least put something together in the meantime.

It seems no matter how many parts I scavenge and hoard I never seem to have the right ones for a project.
 

studiot

Joined Nov 9, 2007
4,998
Tr813 forms an emitter follower voltage supply derived from the 20 volt rail.
What voltage does this rail measure?

If I am right, Tr814 and Tr815 and T801 form a single transformer saturable core oscillator, with the transistors acting alternately. They appear to be in common base connection via R818 and L801.

This type of oscillator does not depend upon a tuned LC circuit or an RC relaxation time constant for its frequency.

Basically one transistor turns hard on, with the rising curent in the windings providing positive reinforcement of the base drive. The windings to the other transistor of the pair, being reversed, reduce its base drive, turning it off.
The CE current of the on transistor builds up (rises) until the transformer core saturates, which happens quite suddenly. Since no further rise in current is possible the base drive is removed, causing the CE current to fall rapidly. The falling CE current induces further turn off voltage in the on transistor but now induces turn on voltage in the off transistor, which then turns on and the transistors swop roles.

The comments of others would be appreciated on this description.


Edit Second thoughts.

R818 and C 813 have a time constant = 390 x 1 x 10E-8 = 3.9 microseconds

From the settings of your first scope picture the period of the sawtooth is about 10 milliseconds (ie f = about 100Hz) with a rise of 0.4 volts in 2 ms and the fall in 8 ms. Please confirm you have not got a x20 switch operated on the timebase. Otherwise this is classic 2 x mains frequency ripple. What do you see at the top of C812?

In trying to understand the rectangular waves in the rest of the pictures, have you changed the timebase setting?
 
Last edited:

Thread Starter

Tealc

Joined Jun 30, 2011
140
The rail measures 20.6v currently.

Here's some other voltages measured with TR814 and TR815 removed. They were lower at TR813 emitter when they were attached, I think the 20v rail was also pulled down but that may have been due to non MJE520 being fitted.

In the morning I will reattach TR814 and TR815. The PCB traces have come away a bit down there so I will have to be creative with transistor legs.



As for the LOPT tester. Isn't it just typical I have loads of 472 film caps, several 474 but no 473 film caps. I have a small selection (4x) of 473 ceramics. I have (2x) 223 caps I can put in parallel and a few 104s I can put in series.
 

studiot

Joined Nov 9, 2007
4,998
See the edit to my last post.

If you have a couple of spare diodes and resistors I suggest making up a temporary auxiliary board (its only low voltage) to accomodate Tr814 and Tr815 until you have finished experimenting.
 
Last edited:

Thread Starter

Tealc

Joined Jun 30, 2011
140
It was definately 100Hz hence the surprise it was so slow. I checked the calibration after that with the 1kHz signal and that was near enough to correct. Just needed a slight tweak and was ok across several time bases. So the 100Hz is from the full wave rectification and basically is our ripple from the main power supply section.

I'm thinking then that, especially from your description on how the oscillator circuit works, that TR814 and TR815 are vital to the actual oscillation and nothing is basically happening without them. If I recall the oscillation I had on the USB scope had more amplitude and was much faster but still had a lot of DC content.

Sorry the first three signals were with a 20kHz ish PWM signal from my 555 circuit, with scope powered down of course. The other signals are from the scope itself.

...

Re the auxiliary board, that's a great idea. Will deffo do that. Yes I have plenty of 1N4148 and 470 Ohm resistors. Will see what the waveform is like with that attached tomorrow.

Would a small bit of breadboard be ok do you think? I have one of those 17 row breadboards that came with an Arduino kit that would be super simple to set up and probably won't be that different to going with stripboard for stray capacitance.

....

C812 shows the same as emitter of TR813, 17.1v, which coincidentally seems to vary by +/- 0.1v or so every time I measure.
 
Last edited:

studiot

Joined Nov 9, 2007
4,998
I meant is the ripple showing on the top of C812.
This is a smoothing electrolytic.
O.4 volts of ripple seems far too high to me.

Yes, breadboard or veroboard or whatever. Pop some sleeves over the legs of the transistors and leave them long.

Yes I think the actual oscillator that drives the transformer primary is formed around TR814 and TR815.

I also think that there are not 4 primaries, but that they are wound in series or there are only two.
 

Thread Starter

Tealc

Joined Jun 30, 2011
140
Yes the ripple is the same at C183 and on the +20 rail.

Here's the text from the manual regarding the oscillator..

The EHT supply circuitry as shown in Fig. 10 is contained within the EHT box assembly. It comprises an oscillator driving a transformer with three secondaries, each rectified to provide the three lines of approximately —1580 volts, —1500 volts and +9kV.

The oscillator is formed by the centre tapped primary winding, of the transformer in the collectors of transistors, TR8I4 and TR815, with feedback windings taken to the bases via protection diodes, D813 and D814. The centre tap of the primary is taken through the choke, L801, to the supply and that of the secondary being returned to the supply through R818.

The waveform at the collectors is approximately alternate half waves of 55 volts amplitude at 30kHz.

The independent grid supply from one high voltage secondary winding is rectified and smoothed by D805 and C807. The cathode supply from a further secondary is rectified and smoothed by D804 and C804, and is stabilised by a feedback circuit which controls the supply switched to the primary of the transformer.

The current through R803, (the focus control potentiometer) and R821 is compared with the reference current taken from the +15 volt line through the 'set EHT preset (R854) potentiometer on the power supply P.C. assembly and R816.

The difference signal is amplified by TR811 which controls TR812 and in turn the series regulator, TR813, which controls the supply line for the oscillator at approximately 14 volts. Overload protection is provided when the voltage across R814 causes D811 to conduct, thereby reducing the drive into TR812. The output from TR813 is limited to approximately 600mA and if this current is maintained, the fuse FS832 on the power supply P.C. assembly will blow.

The third high voltage secondary winding is connected in series with the cathode winding to give approximately 3kV pp a.c. input into the tripler circuit, D801, D802 and D803, with capacitors, C801, C802 and C803. The ripple present on this output is removed by the additional filter circuit, R801 and C806, and the resultant +9kV used as the P.D.A. potential for the CRT.
The breadboard is in place.

Oops wired it wrong. Didn't notice that the diode and resistor are the other way around in the circuit, on the TR185 side.

Now the components are getting warm, especially TR183 which is providing too much output at the moment.

This is the waveform at collector of TR184


and at collector of TR815


Timebase in both cases was 2uS/div and vertical 2v/div with ground at the bottom. DC coupled. So much faster than it should be and nowhere near the 55v. I've yet to put the original electrlytics back in and the BC212 at TR816 hasn't been put back in, mainly because I can't find it :(

TR815 is getting warmer than TR814 but the warmest is TR813 which even though it's case mounted is hot to the touch, so I'm only running it for short periods, hence the rubbish photos.

The base of TR814


The base of TR815


I'm thinking these really should be the same. The amplitude and waveform is about the same, but there is that DC offset. Same timebase as before but the volts/div is now 0.5v.
 
Last edited:

studiot

Joined Nov 9, 2007
4,998
Those transistors are just amplifying the ripple.

Looks more and more like the transformer.

Unfortunately it will be a 'special'.
How easy would it be to change it if you could get once from a scrap unit?

The operation frequency stated in the manual is useful because it cannot be deduced from the circuit, as it is a function of the magnetic parameters of the transformer.

Stewarts of Reading
or
Music Rainbow in Taunton may be able to help with scrap parts.
 
Last edited:

Thread Starter

Tealc

Joined Jun 30, 2011
140
I'm thinking it is too. When it failed it was the nearest component that went with it, the TR815. That makes sense I suppose.

The transformer itself looks to be a fairly simple affair and should be simple enough to remove. I was planning on doing so anyway for when I get the LOPT tester up and running so I can totally isolate it.



I've been checking eBay for OS3000 and there aren't any at the moment. I've got email alerts set up for any Gould oscilloscope so can keep an eye on new listings.

I think Stewarts prices would be more than I actually paid for the whole scope a year ago, considering the prices of their other stuff.

Looking on the internet the last few weeks and it does seem that Goulds had a bit of a reputation for failed EHT transformers.

So after checking the transformer I won't bin it just yet. Will keep it and see what happens.
 
Last edited:

Thread Starter

Tealc

Joined Jun 30, 2011
140
I've built the LOPT tester and it seems to be, sort of, maybe, working. I have a small encaspulated transformer, CT 18V 8VA and all LEDs light up and strobe towards LED 8. I try another 12V x 2 1.5VA and only the second LED lights up.

I tried a few linear power supplies and they lit all LEDs. A SMPS transformer lights up about 6 of them.

My soldering iron only lights up two.


In the EHT which is still attached to the scope if I touch the wires on the primary the LEDs go totally out. If I touch other wires sometimes I get maybe the second LED flashing. A similar thing occurs on the secondary.

I think I should wait until the film 47nF caps arrive. Maybe it will be more reliable.

I did find a few solder bridges so will have to go through the thing again I think.
 

ian field

Joined Oct 27, 2012
6,536
I've built the LOPT tester and it seems to be, sort of, maybe, working. I have a small encaspulated transformer, CT 18V 8VA and all LEDs light up and strobe towards LED 8. I try another 12V x 2 1.5VA and only the second LED lights up.

I tried a few linear power supplies and they lit all LEDs. A SMPS transformer lights up about 6 of them.

My soldering iron only lights up two.


In the EHT which is still attached to the scope if I touch the wires on the primary the LEDs go totally out. If I touch other wires sometimes I get maybe the second LED flashing. A similar thing occurs on the secondary.

I think I should wait until the film 47nF caps arrive. Maybe it will be more reliable.

I did find a few solder bridges so will have to go through the thing again I think.
Sounds like the tester is doing pretty much what its supposed to - isolate the transformer and test again.

Best to test between the 2 points where the collectors go.
 
Top