First Post - Telequipment D75 Oscilloscope

[willmac]

Thanks everyone for this site, which I have already found very useful. This is my first post so introducing myself ---

I am retired but spend a lot of my time in my workshop and at a museum where I am a volunteer. My original background is basically mechanical engineering/manufacturing, with very little electronics, but I am in the process of educating myself. Given my state of ignorance please go easy on the explanations of electronics.

I have been given two oscilloscopes, a Telequipment D75 and a Tektronix 468. Both of these have been stored in a damp shed for perhaps 10 years. Neither seemed to be in working order, but they would be useful tools if they could be repaired, and that is what I have decided to do.

The Tektronix 468 looks a bit intimidating as a first serious repair attempt, but the D75 looks much simpler so I started with it.

There was no trace on the D75 screen even with the use of the beam finder at first, but a really good clean up/dry out and switch cleaner has got it to the stage where both channels seem to work, the calibration signal gives about the right signal on the screen and the controls all do what I think they should.

I am very pleased with progress, but the next stage looks much more difficult. The main power supply has obvious problems with capacitors, one of which has leaked and contaminated the circuit board underneath. My plan is to replace all of the electrolytic capacitors on this board. The large capacitors have three legs, and I intend replacing them with two leg capacitors of the correct voltage and capacitance. Is using two leg capacitors a problem apart from ensuring that they are mechanically stable and correctly polarised?

The more serious problem is the high voltage power supply. I notice that after a few minutes with the oscilloscope powered up, I get a faint smell coming from this board. Left longer I get a suspicion of smoke, which seems to be coming from one of the polyester capacitors (C307 or C310). Now I was planning to just replace these, together with the resistors which are a known problem on this board, but have hit a problem.

The polyester capacitors are rated at 5kV, and I can't find capacitors of this type rated for such a voltage. The best I can find are for 3kV and I assume that they would not be usable. Looking at the circuit diagram, I make a wild guess that these capacitors are acting as bypass capacitors to take out any high frequency noise on the high voltage line. Does that seem reasonable? If so I would further assume that the capacitance value is not ultra critical?

Can I replace the polyester capacitors with ceramic capacitors with the correct capacitance and voltage rating? I can find ceramics with a 10kV rating and very close to the same capacitance (20nF). These are not cheap but perhaps they would be more robust than the original?

I have no means of measuring these voltages safely, but I read that I should. Is making up a high voltage probe still the best appraoch? If so what voltage rating should the resitors have? Shoild they be rated for 5kV or better?

Any help you can provide on this would be very welcome

[dave cox]

In the power supply, as well as meeting voltage and capasitance specification, you need something capable of meeting the ripple current rating. This is probably not so difficult in this case but beware if the replacement is a tiny fraction of the size of the original! Materials & manufacturing is a lot better today than when the scope is made so ~ 1/2 the size could be fine.

dc

[willmac]

OK, I can see that could be a problem. The ceramic caps I have found are physically much smaller than the original polyester caps. How do I know what ripple current is likely to be present? I can't find a data sheet for the original caps so I can't just match the original specification unfortunately.

[MrBungle]

You can use two 3kv in series as an alternative. You will need twice the capacitance though.

New capacitors are definitely smaller - I've replaced some in a Tek 465B with embarrassingly small ones with the same specification.

[trh01uk]

"Is using two leg capacitors a problem apart from ensuring that they are mechanically stable and correctly polarised?" - you asked.

Well that depends on exactly what these three legs are doing. I see two possibilities:

1. Sometimes a single capacitor has two legs connected together - often the negative terminals. The other leg being the positive terminal. The two negative terminsals are just for mechanical stability usually. Replacing one of these with a two leg capacitor will be fine.

2. The other possibility is that there are two capacitors in the one can, with one leg being a common negative connection for both capacitors, and the other two legs being the positive connections for each capacitor.

In this case, you need to replace the whole can with with two separate capacitors - or find a double capacitor in one can like the original.

The manual should make it entirely clear which case is the correct one.

Richard

[trh01uk]

"I have no means of measuring these voltages safely, but I read that I should. Is making up a high voltage probe still the best appraoch? If so what voltage rating should the resitors have? Shoild they be rated for 5kV or better?"

The way to make up such a probe is to have a string of lower voltage rated resistors in series. Typically an axial leaded resistor has a voltage rating of 500V - though really you should check out the exact specification. Radiospares and Farnell carry such detailed specs.

So to make a 5kV rated probe, I would take 10 x 500V rated individual resistors, each resistor having a value of 1/10 of the total resistance you need. For the sake of safety the entire chain needs to be encapsulated in some thick plastic insulating tube. Some very thick hose pipe might well do the job.

Richard

[mhennessy]

The 3rd leg isn't connected to anything - it's just for mechanical stability.

When I replace these, I buy modern types with 3 legs. But, the unconnected leg has moved, meaning you have to cut tracks and carefully make connections under the PCB with wire. I don't mind doing this because it preserves the look, but others may prefer to use regular radial caps. From memory, the current taken on the 24V rails is less than half an amp (about 400mA IIRC), so I'd be surprised if ripple current becomes an issue.

I've never had to work on the EHT section of the D75/83, but have been inside the similar DM63 many times. There's 2 electrolytic capacitors in there that should be changed. I've not had a problem with the other types, so wonder if the damp storage is part of the problem?

Given the voltages with this one (~2.5kV), I'd recommend a proper HT probe rather than a DIY one. I did my own for the DM63, but that was for 1400V or thereabouts. I'm a bit of a coward! Every now and again, ex-MOD units come up for sale - I got a Fluke one that had never been out of its case, and it was only a few 10s of pounds. Worth every penny IMHO. The trouble is, quite apart from safe operation at those sorts of voltages, resistors have a voltage coefficient that leads to inaccuracies. Lots of smaller resistors in series is the best way to minimise the errors and improve safety, if you must

Metal oxide resistors are much, much better at high voltages. E.g., Mullard/Vishay VR37 series.

[HowMuch!]

CPC do a range of high voltage resistors, not expensive either, if you want to make your own probe
http://cpc.farnell.com/c/electronic-...ge-rating=15kv

I expect others do them too.

[MrBungle]

Good spot. Quite cheap too!

[willmac]

Thanks for all the help with this project so far. I have ordered the new parts I need to fix the faults so far found and will start to replace parts soon.

However --- as soon as I sent the order, I found some more problems and I am puzzled by both the failure modes and the function of these parts so I am hoping that the experts can enlighten me.

In the main power supply, TR401 has only 2 legs left. The third seems to have dissapeared, possibly blown off or just dropped off. This transistor is part of the +105V supply and apparently provides short circuit protection. Could the missing leg have been caused by providing such protection? The supply seems to work without it and measures ok so a bit strange. TR401 is a BF305.


On the plug for the CRT there is a Zener diode D305 soldered between the cathode and the grid pins. This diode is cracked right through. Despite this the tube seems to work fine. What is the function of a zener diode in this position? I can't think of any obvious purpose. The zener is supposed to be rated at 130V. This seems a very low voltage given that the cathode voltage should be around 2.5kV and the grid presumably less than this? Does this diode control the difference between the cathode and grid voltages?? wild guess there!

My knowledge of vacuum tubes is near zero so any explanation would be helpful.

[mhennessy]

The voltage across the zener is what matters, not the two absolute voltages with respect to earth.

The schematic for the very similar DM63 suggests that there is 50V across the zener - that detail doesn't seem to be given for the DM75. Both ends are sitting down in the -2500V region. The zener is no-doubt there to limit how far they actually can be separated, limiting the maximum beam current.

TR401 is not switched on until the current limit is reached on the 105V rail. That needs to be replaced ASAP. The circuit is quite subtle - as soon as the current limit is hit, TR404 is turned off, and TR401 stays on until the 'scope is "rebooted". This prevents TR416 from overheating and causing further damage. But as soon as the 105V collapses, then so does the +/- 24V rails, which again stops a trail of destruction from occurring.

[Diabolical Artificer]

Re high voltage caps, some can be found on ebay and some here - http://hvstuff.com/high-voltage-capacitors . these arn't cheap but the price comes down if the dollar is weak against the pound.

Also on ebay and other places you can find HV resistors that have a blue body. These are usually in the 10M and up range like these http://www.ebay.co.uk/itm/10M-Ohm-5-...0AAOSwMNxXXXRD which can be cheaper than going to Farnell, Digikey etc - http://uk.farnell.com/vishay/vr68000...ial/dp/2396017. I have a box full of HV ( 1kv + I think) 120M resistors, you'd be welcome to a few, but I don't think are suitable. From what I remember I used 10 x 10M and one 11M for my EHT measuring lead. Helps to have croc clips on each end and not a probe on one end. The spark you get when taking readings can be a bit off putting and obviously a fixed test lead attached prior to switch on is safer.

Andy.

[willmac]

Latest progress with this project.

I have replaced all the failed parts and all electrolytic caps associated with the main power supply.

I then made what was probably an error, but I would like your comments to confirm. With the power supply board out of the oscilloscope frame, but properly grounded, I set the supply voltages as accurately as I could. The unregulated voltage ended up a bit high, but I thought that would not be a big problem. I assembled the scope and tested it. I got a spot on the screen, but no timebase and only limited control over the spot position. I have to admit that I thought that I had destroyed the scope after all that effort, but I decided to have a think about the problem

I decided to test the voltages again with the power supply in circuit. The +105V was ok, as was the 30V unregulated (much closer to nominal now), but the +24V was down about 6V. The -24V was also down a bit. I did worry a bit about what was causing the drop, but I adjusted the voltages back to nominal. The scope now worked correctly in all the ways I could easily test.

I have two questions based on this experience.

1) Is it generally correct to test a power supply in circuit with a full load to set the voltages?

2) Is it sufficient to get correct voltages and no smoke or hot parts in the power supply or should I also check for ripple or anything else? If I should check for ripple, what would be reasonable?

The scope now has a noticeably more stable, clean and bright trace. The sizzling sound and smell from the high voltage supply seems to have stopped. Given this I am tempted to just use the scope without delving more deeply into this area.

Many thanks for the help with this project, which I have found interesting and rewarding to date.

[John Caswell]

I suspect the sizzling sound is corona discharge in the EHT area, and the smell is ozone coming from it.
Try looking at the area in darkness to see if you can see a blueish discharge, possibly caused by dust or a sharp point/solder joint. cleaning the area and "ball jointing" solder joints will help.

John

[willmac]

The smell definitely was more of a smokey burning nature, but it has stopped after the repairs I have made. I wonder whether replacing the zener diode that controls the difference between the cathode and grid voltages has helped with this.

The sizzling sound has also gone.