Big Farnell PSU SCR control circuit problem

[See_Mos]

OK, I think you are mistaking the AC wafeform and the gate pulse. The gate drive is only a couple of volts of narrow pulses, not always visible on USB scopes.

Start by unplugging or disconnecting the SCR control card then measure the resistance of the gate to cathode of each thyristor. They should be roughly the same and usually between 18 and 27 ohms depending on the type.

Reconnect the board and check the voltage across Z5?

Turn the voltage and current controls up a little then measuring from termianal 13 what is the voltage on pin 1 of IC1. does the voltage change of the voltage and current controls are set to minimum?

Turn the voltage and current controls up a little and connect the scope earth to terminal 13. Have a look at the collector of VT4 there should be a narrow negative going spike.

Have a look at pin 7 of IC1 there should be a squarewave or narrow pulse. When working correctly the high low ratio varies with output demand.

OK, lunchtime, I'll be back!

[See_Mos]

D23 is just a protection diode to prevent reverse potential across the transistor array. Basically when the power is turned off the voltage across the bulk storage capacitors is discharged by the parallel resistors and D23 discharges the output capacitor C36 via the same resistors. In this way the reverse voltage across the array can never be more than the 0.7 voltage drop across the diode.

Try connecting a normal houshold filament light bulb across the output, set the current low and the voltage high then measure the voltage across one of the bulk storage capacitors C30....C33. I think it should be about ten volts higher than the required output voltage.

Also measure the voltage across Z3 then Z5 on the SCR control card and report back

[dave cox]

R4 has a different value for the 30V / 60V boards. I guess the volts adjustment control(s) have the same value in both 30V & 60V versions. R4 should have 6.2V across it when in voltage regulation mode.

dc

[Chris Wilson]

I'm back on this later, I had got to the throw it on the tip level, so had a break from the damned thing, I really appreciate the input. I'll post later now I have calmed my nerves and left it alone in disgrace for a while, thanks!

[Chris Wilson]

OK, back to basics. Just the original Auxiliary supplies board fitted. Pins 17 and 18 in relation to 0V pin 15 all from centre tapped transformer MT2 show 40V

Across Z4 I see 5.32V. There's a chance the cases of VT9 and VT10 were touching, made sure they are well spaced now, but they still feel pretty warm, as they did before. 15V output pins 2 and 8 in relation to 0V out pin 5 show 16V each. Anyone see issues here? This is with JUST this one board fitted.

As an aside the thing drives me mad as nearly every time it's turned on or off by its main power switch the 30mA RCD in the consumer unit trips. Am I right in saying old PSU's with input filter caps across live and neutral, and their centre to ground can trip RCD's or is there a basic fault here as well? the other supply has never done this, AFAIK?

[See_Mos]

The tripping is almost certainly caused by the filter. If it is a metal cased one the solution might not be so easy but you could try isolating the earth pin at the filter. I have also experienced the same problem with big transformers where the live connection is closest to the transformer core. The capacitance between the winding and the core was causing the trip. in these cases the ‘fix’ was to reverse the live and neutral connections, at the transformer if necessary.

I don’t think the problem is with the auxiliary card. The case of the 2N3053 is connected to the collector and as they are joined on the schematic there would be no problem if they did touch.

I would expect VT9 and 10 to be quite warm so as long as there is 15 volts across Z5 and Z6 and the trip light is off the fault is likely to be elsewhere.

As a matter of interest, have you checked the smaller 'lytics? Low value ones tend to loose their value, especially the blue ones made by Philips

[Chris Wilson]

The filter components are all exposed, so I could snip a lead of the cap that goes to the earth wire, and if that doesn't work I will try reversing the primary wires on the main transformer, thanks. it sounds trivial but I have to go up and down a flight of stairs and across the house to reset the trip, and it gets tedious at my age

So 16V across the Zeners near enough and is OK?

I have removed the SCR housing metalwork and that allowed me to access the snubber resistor R99 / capacitor C25.
On my Peak ESR meter the C25 0.1 uF , 1000V cap shows as open circuit or low capacitance. As it's accessible should I change it? If so with what type of cap, and is the 1000 figure on it a DC rating please? I wouldn't want to do the surgery to get to it twice. I'll test the SCR's out of circuit later. Thanks!

[Station X]

I assume you mean Peak rather than Peal?

ESR Meters are meant to be used for checking Electrolytic Capacitors and a 0.1uF capacitor is unlikely to be an electrolytic.

[Alistair D]

The subject of RCD tripping is discussed here.

http://golbornevintageradio.co.uk/fo...d.php?tid=6201 See post #15 in particular.

One possible fix that I have not got round to trying

http://www.vintage-radio.net/forum/s...d.php?t=119636

Al

[See_Mos]

No need to mess around with the SCR's, if there was a fault there the unbalance would blow the mains fuse / circuit breaker every time.

This would be so easy if the PSU was in front of me, it's the type of work I have been doing for the last twenty years!

We still need to know if you can get at least 55 volts across C30.

First off, I would be a bit nervous of the voltage rating of the main capacitors C30...C33 because 1.4 * 64 volts = 89.6 volts which is higher than the 75 volts rating of the capacitors. I know they are supposed to withstand at least 10% over voltage but it's too close for comfort.

This isn’t quite how I would proceed but considering what you have done so far here is what I would suggest:-

Refit the SCR control card. Disconnect the positive of the bridge rectifier. Connect a filament light bulb of at least 60 watts across the positive and negative of the bridge. Connect a variable power supply, across D23 or at the SCR control card – positive to cathode (terminal 5) – negative to anode (terminal 13) Set the variable supply for about 8 volts. Turn on the power for the PSU then by adjusting the variable supply around 5 to 12 volts you will find a point where the lamp will either be on or off. You will not be able to control the intensity because there is no DC feedback around IC1A. If the SCR card is working correctly you should have about 80 volts across the lamp.

Alternatively I could test the SCR card for you.

[Chris Wilson]

Marvellous info, I am working today and tomorrow but at the first chance will test as you suggest above, thank you! Can I use a 60W UK 240V bulb as the load? Whilst i had access to the SCR's I tested them on 12V with an automotive 10W, 12V bulb as the load and as you and others have said, they seem fine, although one seemed reluctant to conduct for a couple of times, but then did and held until the supply was removed.

[See_Mos]

60w 240V should be fine, it's just to provide a light load (pun intended) to help the SCR's to fire properly and give an instant visual output as well.

[Chris Wilson]

Thanks See_Mos. Update, did exactly as you suggested and the original SCR board, fitted as the ONLY board in the PS, did not work. Fitted SCR board from sister 25 Amp at 60V supply and bulb lit and went out with a few volts hysteresis. The swing between on and off was roughly between 7.5 and 9V.

Refitted original SCR board and I can only measure 0.94V across Z1, but I see the correct 10V across Z2. It's hard to say with both these Zeners being right next to the WW ceramic resistor R27, which does run hot on both boards, but I was fairly sure Z1 was also running hot. Not sure if it can with 0.94V across it though.... I swapped Z1 for another 15V Zener at 5W rating and see the same 0.94V across the new one. Thanks.

[Chris Wilson]

Originally Posted by Alistair D

The subject of RCD tripping is discussed here.

http://golbornevintageradio.co.uk/fo...d.php?tid=6201 See post #15 in particular.

One possible fix that I have not got round to trying

http://www.vintage-radio.net/forum/s...d.php?t=119636

Al

Thanks Al, useful links, appreciated.

[See_Mos]

No volts across Z1 and 10. volts across Z2 I suspect VT5 is short circuit.

If it is a 2N2905(A) would be OK and easy to find

[Chris Wilson]

Thanks See_Mos I renewed VT5 not long ago and the old one tested good out of circuit. Just using diode test on my multi meter I now see no difference between the VT5 on the working board and VT5 on the none working, low V across Z1 board.

[Chris Wilson]

BTW, I see 63V across the output of the bridge (no cap smoothing caps connected) when bulb is ON and about 125mV when OFF.

[dave cox]

Is C12 a dip-tantalum cap ?
If it's shorted then VT6 would also get beaten up

dc

[See_Mos]

Originally Posted by Chris Wilson

BTW, I see 63V across the output of the bridge (no cap smoothing caps connected) when bulb is ON and about 125mV when OFF.

That's good.

[See_Mos]

Originally Posted by dave cox

Is C12 a dip-tantalum cap ?
If it's shorted then VT6 would also get beaten up

If it is then beware of replacing those BC182L, use only L because the pinout is different from BC182A ,B or C