UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Test Gear and Workshop Equipment

Notices

Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders.

Closed Thread
 
Thread Tools
Old 14th Jul 2018, 12:49 pm   #1
mole42uk
Octode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Bristol, UK
Posts: 1,905
Default Gould 4072 bad PSU

A first visual skirmish around the dead power supply of this Gould 4072 oscilloscope produced this clutch of hot & bothered capacitors:
Attached Thumbnails
Click image for larger version

Name:	leaky&legless.jpg
Views:	180
Size:	34.4 KB
ID:	166047  
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 14th Jul 2018, 1:26 pm   #2
AC/HL
Moderator
 
AC/HL's Avatar
 
Join Date: Nov 2003
Location: Heckmondwike, West Yorkshire, UK.
Posts: 8,426
Default Re: Gould 4072 bad PSU

Time was it was the Mk1 nose, these days the eyes have it. Always a welcome sight!
__________________
Bill, BVWS member
AC/HL is offline  
Old 15th Jul 2018, 2:38 pm   #3
mole42uk
Octode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Bristol, UK
Posts: 1,905
Default Re: Gould 4072 bad PSU

I also found this RIFA capacitor which had clearly given of it's best.....

There was an o/c 100R resistor in the supply to the 3524A SMPS chip, a replacement quickly fried..... I tried putting a 12V supply on to the chip which only drew about 10mA so I have left it on the bench while I think about the next move.
Attached Thumbnails
Click image for larger version

Name:	RIFA 470nF.jpg
Views:	86
Size:	32.1 KB
ID:	166082  
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 22nd Jul 2018, 9:33 pm   #4
Noopy2014
Tetrode
 
Noopy2014's Avatar
 
Join Date: May 2014
Location: Eichstätt, Germany
Posts: 72
Default Re: Gould 4072 bad PSU

Hello Richard,

are you sure the 3524 was fully working while you supplied it with 12V?
At it“s shutdown pin (10) the 3524 gets a high signal as long as the transistor Q2 doesn“t get activ and pulls the potential low. In this Situation the 3524 is only standing by. Q2 checks the input voltage. The input voltage has to be above 12V to be sure the discrete voltage regulator can do it“s job. Perhaps exactly 12V isn“t enough to enable the 3524 and the meassured 10mA was only the standby current.

A guess from my side: If the 3524 didn“t fail on it“s own, i would interview Q3 and Q4. Q5 and Q6 can die without much smoke but because of the gate resistors they shouldn“t kill the 3524.

Regards,

Richard
Noopy2014 is offline  
Old 23rd Jul 2018, 8:48 pm   #5
mole42uk
Octode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Bristol, UK
Posts: 1,905
Default Re: Gould 4072 bad PSU

Hello Richard,

I am working on the pre-regulator because that is indicated in the Service Manual as the next section to check after the AC mains rectification. Tonight I found an o/c 2µ2 tantalum and two s/c diodes in the circuit that derives power to the 3524A from the transformer when the chip is running.

I can see that the 3524A has a typical SYNC output on pin 3 but nothing on pin 11 or 14 which, from my understanding of the datasheet, is because the voltage on the shutdown pin 10 is too low. Because I see a SYNC output I think the 3524A may be okay. I have several 3524 parts in my spares box which might serve to prove the remainder of the pre-regulator circuit.l

I think there are a few failed parts in this supply and I'm tracking them down part by part using low voltage test supplies until the measurements start to make sense. I am fortunate to have another similar supply which also does not work but allows me to cross-reference. When I have the first one working I'll fix the other as a spare.
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 26th Jul 2018, 8:45 pm   #6
mole42uk
Octode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Bristol, UK
Posts: 1,905
Default Re: Gould 4072 bad PSU

I wonder if anyone can help with this bit. I might be having a mental block, but I can't see how the circuit works.

Given that the 3524 shutdown pin is active low, how does the circuit around Q10 work? It is supposed to shut the PWM down if there's an overvoltage output. D16 feeds the 12V start-up votage to pin 10, and R41 holds it low unless something else holds it high.

But how does the circuit arount Q10 act as an over-voltage shutdown?
Attached Thumbnails
Click image for larger version

Name:	Screen Shot 2018-07-26 at 20.37.23.jpg
Views:	97
Size:	60.1 KB
ID:	166600  
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 27th Jul 2018, 5:16 pm   #7
dave cox
Octode
 
dave cox's Avatar
 
Join Date: Jan 2009
Location: Bristol, UK.
Posts: 1,412
Default Re: Gould 4072 bad PSU

It doesn't seem to make sense !
There is nothing to pull pin 10 towards ground from +12V !

Is one of the voltages -Ve ?

dc
dave cox is offline  
Old 27th Jul 2018, 7:38 pm   #8
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 3,769
Default Re: Gould 4072 bad PSU

Where does the '12V' line on your circuit come from? A rectifier on the chopper transformer? If so, it will not be present until the PSU is running.

I don't think it's an over-voltage shutdown at all. What it seems to be doing is preventing the PSU from starting until the voltage across the mains smoothing capacitors (the '340V' in the diagram I assume) is high enough for Q10 to turn on, thus pulling shutdown high. At this point the PSU rattles into life, the 12V line comes up and keeps the supply running no matter what Q10 does.
TonyDuell is offline  
Old 27th Jul 2018, 8:24 pm   #9
mole42uk
Octode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Bristol, UK
Posts: 1,905
Default Re: Gould 4072 bad PSU

I agree Dave, I can't make sense of it either.

The only thing pulling pin 10 down is R41, a 5k6 resistor but if D16 has 12V on the cathode, then the 5k6 won't do much. there is a transistor which, if turned on, will shunt the D16 voltage towards ground but I can't see why it would be turned on.

I've looked at several datasheets for the 3524 chips, and I used them in a commercial item I designed years ago, all seem to indicate the pin 10 is active low, in other words, if it is low the chip shuts down. Perhaps I've misread the data and it's actually active high - in other words, if pin 10 goes above 0.5V the chip shuts down. That makes more sense, the internal circuits I've seen for the 3524 show an NPN transistor with it's base connected to pin 10, collector to the output of the error and current sense ampolifiers and the emitter to ground.

If the pin 10 sense is active high, then the circuit around Q10 makes more sense, if the line voltage, nominally 340V, goes too high, then the 75V zener-controlled base of Q10 will allow the voltage across R40 to rise. D16 is still a fly in the ointment, though, since it would seem to deliver 12V to pin 10, high enough to shut down the 3524, at all times.

I think I'll have to do some more testing and probing although I don't want to use 340V until I'm reasonably certain how the circuit is supposed to behave.
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 27th Jul 2018, 8:30 pm   #10
mole42uk
Octode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Bristol, UK
Posts: 1,905
Default Re: Gould 4072 bad PSU

Quote:
Originally Posted by TonyDuell View Post
Where does the '12V' line on your circuit come from? A rectifier on the chopper transformer? If so, it will not be present until the PSU is running.

I don't think it's an over-voltage shutdown at all. What it seems to be doing is preventing the PSU from starting until the voltage across the mains smoothing capacitors (the '340V' in the diagram I assume) is high enough for Q10 to turn on, thus pulling shutdown high. At this point the PSU rattles into life, the 12V line comes up and keeps the supply running no matter what Q10 does.
The 12V is the initial start-up voltage derived from the bridge-rectified, capacitor-smoothed mains input by a zener-controlled BU810 transistor. The 340V nominal line is the supply to the PSU and is regulated by the pre-regulator that I'm working on right now.

I think I'll upload a larger section of the circuit.
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 27th Jul 2018, 8:34 pm   #11
mole42uk
Octode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Bristol, UK
Posts: 1,905
Default Re: Gould 4072 bad PSU

Here's a larger section of the PSU
Attached Thumbnails
Click image for larger version

Name:	Screen Shot 2018-07-27 at 20.32.42.jpg
Views:	74
Size:	61.4 KB
ID:	166664  
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 27th Jul 2018, 8:37 pm   #12
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 3,769
Default Re: Gould 4072 bad PSU

I can't find an SG3524 data sheet that says that Shutdown is active-low and as you say the block diagram in said data sheets implies it's active high (high to shut the IC down, low or floating to let it run).

Is D16 certainly shown the right way round? If not then it could be to clamp the voltage on the shutdown pin to 12.6V to prevent damage to the IC. And where does that 12V line come from.

That said, I think Q10 will turn on at a lot less than 340V on the supply line. Which implies the IC will never run.
TonyDuell is offline  
Old 29th Jul 2018, 8:00 pm   #13
Noopy2014
Tetrode
 
Noopy2014's Avatar
 
Join Date: May 2014
Location: Eichstätt, Germany
Posts: 72
Default Re: Gould 4072 bad PSU

In my understanding it works like this: High-Level on Pin 10 disables the 3524.

There are two causes that disable the chip:

1. If the input-voltage (PLD 7,8) is to low for a proper 12V-supply the zener D15 doesn“t conduct, than there is low on Q12 base, than there is high on D16, an the 3524 is of cause of low supply.

2. For the second cause you have to realize that the upper switching supply works with the input voltage minus the voltage of the lower switching supply. That means that a too low voltage on the output of the lower supply gives a too high voltage on the upper supply.
The second path around Q10 switches the 3542 off whenever the voltage on the output get“s too low. With switched off lower supply there will be no voltage on the upper supply because the power fet Q14 is (hopefully) off.

Sorry for my English. I“m in a hurry...

Best regards, Richard
Noopy2014 is offline  
Old 30th Jul 2018, 7:52 am   #14
mole42uk
Octode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Bristol, UK
Posts: 1,905
Default Re: Gould 4072 bad PSU

Thank you for all your thoughts. I see now that I mis-read the datasheet where pin 10 was labelled "shutdown" with a line above, that means it has to be low for the 3524 to run, not that it has to be low for the chip to shutdown.

I may be chasing a non-existent fault now. Initially, R34 (100R) smoked after I'd replaced an open circuit component, and I was trying to find a higher current path which would make it smoke. Having discovered that D17 & D18 were short circuit, that would give a path to ground through R34 and Q11 and the transformer primary.

I also thought that the 12V zener was affecting the base of Q11 but if it is only controlling the base of Q12 then it's likely that the control of pin 10 is correct and the circuit just needs a high enough voltage on PLD7,8 for the circuit to run. I haven't found any other failed components on the pre-regulator sub-panel. I may install the chip in the socket with pin 10 bent outwards to check that the circuit will do something sensible on a lower voltage, my high-voltage power supply isn't current limited at a low enough level to use without some precautions!
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 30th Jul 2018, 8:08 pm   #15
Noopy2014
Tetrode
 
Noopy2014's Avatar
 
Join Date: May 2014
Location: Eichstätt, Germany
Posts: 72
Default Re: Gould 4072 bad PSU

You are welcome!
Troubleshooting in the power supply can be a pain in the ass. I hope you find every fault and i hope the transformer is still ok!

I“m on page 310 of a teardown regarding the 4074. Coming soon... ...but will be German...
The method of Operation looks like this:
Attached Thumbnails
Click image for larger version

Name:	supply.JPG
Views:	46
Size:	20.0 KB
ID:	166799  
Noopy2014 is offline  
Old 30th Jul 2018, 8:38 pm   #16
mole42uk
Octode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Bristol, UK
Posts: 1,905
Default Re: Gould 4072 bad PSU

Maybe it is fortunate that I worked at Petershausen close to München for two years and that my partner speaks German!

I have a faulty 4072 and most parts of a 4074 - I think you sent me the front panel for the 4074 - because I will use all the parts I need from both instruments to make one working 4074 and then I will have some spare parts!
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 30th Jul 2018, 9:28 pm   #17
Noopy2014
Tetrode
 
Noopy2014's Avatar
 
Join Date: May 2014
Location: Eichstätt, Germany
Posts: 72
Default Re: Gould 4072 bad PSU

Yeah, more than five posts! Now I can reply a little bit faster and can save more than 10 PMs!

I will start a thread here as soon as I have finished the report.
I hope it will be at least entertaining:

Correct, I have sent you the frontpanel. I was lucky to get three 407x in very bad conditions for very little money. For repair and restoration help is always welcome and so I also help wherever I can. Even with my poor English. But I was shocked how expensive it is to send a parcel from Germany to Britain...
Attached Thumbnails
Click image for larger version

Name:	ccd.jpg
Views:	67
Size:	134.7 KB
ID:	166800  
Noopy2014 is offline  
Old 31st Jul 2018, 3:53 pm   #18
mole42uk
Octode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Bristol, UK
Posts: 1,905
Default Re: Gould 4072 bad PSU

That image reminds me of when I worked in silicon wafer testing back in the '90's.
__________________
Richard

Index:
recursive loop: see recursive loop
mole42uk is offline  
Old 31st Jul 2018, 4:15 pm   #19
dave cox
Octode
 
dave cox's Avatar
 
Join Date: Jan 2009
Location: Bristol, UK.
Posts: 1,412
Default Re: Gould 4072 bad PSU

Low to inhibit on pin 10 makes more sense with the rest of the circuit (i think) ...

At startup, Q?? (BU810) turns on and pulls pin 10 up through R25 (?)
As the rail ramps up Z?? (12V) conducts, and the lower BC182 will eventually turn on (C?? slowing it down, as connected B-E) robbing the current that WAS pull pin 10 up ...
By that time, pin 10 is pulled up by Q?? as not much above 75V it will saturate!

But I could be talking rubbish

dc
dave cox is offline  
Old 31st Jul 2018, 4:51 pm   #20
mole42uk
Octode
 
mole42uk's Avatar
 
Join Date: Aug 2010
Location: Bristol, UK
Posts: 1,905
Default Re: Gould 4072 bad PSU

After some more pondering, I think my mistake was to mis-interpret a data sheet. Pin 10 on that sheet is referenced as SHUTDOWN with a line above which I took to mean active low as in logic gate naming. However, it makes more sense in relation to the internal circuit of the 3524 if pin 10 is low to operate the chip, so SHUTDOWN with a line above means that if it is low the chip will operate, if high, it will shutdown. In other words, pin 10, in relation to the whole 3524, is active low.

From there, some of your analysis makes sense, line 1&2 are good, in that the chip is held in shutdown until the power input has stabilised, then the BC182 switches on and shunts the voltage to ground allowing R41 (5k6) to pull pin 10 low. The Service Manual states that this pin should be 0.3V.

Q10, the 2N6520, should be the over-voltage switch. The 75V zener D25 should allow the transistor to switch on at a voltage on R36 & R39 (both 39k) which is above 80V (from the Service Manual).

This supply is specified to run on any input AC from 100V up to 250V but my brain isn't yet able to work out how the power supply takes the 140-340V DC on the rail connected to PLD7,8 and make it into 70V DC to drive the main regulator since both are connected together on the drawing. EDIT, I've just looked at Bavarian Richard's drawing and realised that the power supply is messing with the ground to get the required voltages. How it does that is, as yet, as mystery to me, but if the main regulator ground is not at the same potential as the pre-regulator ground, then it begins to make sense.
__________________
Richard

Index:
recursive loop: see recursive loop

Last edited by mole42uk; 31st Jul 2018 at 5:00 pm.
mole42uk is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 7:51 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2020, vBulletin Solutions, Inc.
Copyright ©2002 - 2020, Paul Stenning.