UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV 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 23rd May 2017, 10:39 pm   #41
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: HP1715A repair

The opamp looks OK, pin 2 (- input) is positive and output is -ve.

You are looking at channel B Q8? Where is the +ve collector voltage coming from? Most likely from a fried FET amongst other bits.

Looks like you will need to measure some voltages on the Attenuator boards. If you really have +9V on the collector then I would expect to find multiple faults.

The circuit would be so simple to understand if Q8 were drawn the other way up!

Last edited by PJL; 23rd May 2017 at 10:54 pm.
PJL is offline  
Old 24th May 2017, 7:47 am   #42
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: HP1715A repair

Sorry, that's my mistake. It should be -9.77v on the collector.
WaveyDipole is offline  
Old 24th May 2017, 1:28 pm   #43
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: HP1715A repair

Check that the +ve input, P3, is definitely zero. P3 has a slightly positive bias via R80 but its only about 0.5mV. The ATTEN BAL control can introduce about +/-10mV on P2 but P2 is measuring at 380mV so it looks like a fault on the ATTEN board.

At the moment Q8 is conducting as much as the circuit allows and that should increase the current through Q2 which should lower the grid of Q3 whose emitter provides the -ve feedback through U1 P2.

My calculation puts the output at about +1.75V when it should be 0. You can verify this by measuring the voltage on the connection from the ATTEN board to the pre-amp board marked 3.

This might be explained by a FET which is now a short circuit!
PJL is offline  
Old 24th May 2017, 1:54 pm   #44
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: HP1715A repair

Well it certainly looks like the FETs are blown, but that is only part of the problem. I checked voltages around the semiconductors on the attenuator board. None of them are correct. These readings are for channel A:

Q1 G +6.05 [SIG]
Q1 S +5.26 [+1.5]
Q1 D +7.13 [+10]

Q2 E -10.5 [+0.7]
Q2 B -9.8 [+1.5]
Q2 C -6.0 [+5.4]

Q3 E +2.46 [-13]
Q3 B -3.39 [-12.2]
Q3 C +2.49 [-3.8]

The two transistors seem to test OKish but with an oddity. BC and BE are OK, CE tests like a diode? Both Q2 and Q3 on both boards test the same. Q3 in each case was tested in-circuit, however, Q2 was removed.

I found a FET in my box of bits (SK125) and having checked the pinout I tried swapping it. Voltages are still wayy off what they should be but I now get a signal on the second scope at the input to A1 of that channel although the signal is distorted. On the DUT I get a very slight semblance of a signal although the attenuator seems to do very little.

Q3 is a BFR91A which I have found a couple NOS on eBay. However the other two are more elusive. Markings as follows:

Q1 E213
Q2 752 4-636

In CH B the FET has been replaced with something marked F2860. A quick search turned up this:
https://nationalstocknumber.info/par...2860_012434454
It does not really identify the device although it does provide some parameters.

BTW, VR1 and VR2 - does the 'L' in the symbol denote that these are these tunnel diodes?
WaveyDipole is offline  
Old 24th May 2017, 2:39 pm   #45
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: HP1715A repair

Q1 and Q3 are dead. Q2 seems to be OK. What's the state of the other channel? Any chance Q3 has survived? Probably not as the faulty FET would result in a rather high base current.

PS: Q3 base (-3.39) looks wrong as Q1 source is +5.26. I would guess that should be +3.39.

It would also be good to see the voltages with the replacement FET.

Last edited by PJL; 24th May 2017 at 2:51 pm.
PJL is offline  
Old 24th May 2017, 2:40 pm   #46
factory
Octode
 
Join Date: Jul 2009
Location: Cheshire, UK.
Posts: 1,706
Default Re: HP1715A repair

VR stands for voltage regulator on HP diagrams, VR1 & VR2 look like zener diodes to provide 10V & 5.4V from the +15V supply rail.

David
factory is offline  
Old 24th May 2017, 2:51 pm   #47
factory
Octode
 
Join Date: Jul 2009
Location: Cheshire, UK.
Posts: 1,706
Default Re: HP1715A repair

The voltages in the brackets for Q2 & Q3 in post 44 seem to be mixed up.

David
factory is offline  
Old 24th May 2017, 3:18 pm   #48
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: HP1715A repair

This article should help as it suggests a J309 as suitable and there is an ebay offering from France:
https://www.eevblog.com/forum/repair/hp1725a/
PJL is offline  
Old 24th May 2017, 4:30 pm   #49
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: HP1715A repair

Quote:
Originally Posted by PJL View Post
Check that the +ve input, P3, is definitely zero. P3 has a slightly positive bias via R80 but its only about 0.5mV. The ATTEN BAL control can introduce about +/-10mV on P2 but P2 is measuring at 380mV so it looks like a fault on the ATTEN board.
It varies from 22mV to 275mV depending on attenuator setting. Setting to from DC to GND makes about 2mV difference.

Quote:
Originally Posted by PJL View Post
At the moment Q8 is conducting as much as the circuit allows and that should increase the current through Q2 which should lower the grid of Q3 whose emitter provides the -ve feedback through U1 P2.

My calculation puts the output at about +1.75V when it should be 0. You can verify this by measuring the voltage on the connection from the ATTEN board to the pre-amp board marked 3.

This might be explained by a FET which is now a short circuit!
It was about +0.8v, but definitely not 0.

Quote:
Originally Posted by factory
VR stands for voltage regulator on HP diagrams, VR1 & VR2 look like zener diodes to provide 10V & 5.4V from the +15V supply rail.
I should have realised that! There would have been no need for a silly question. I don't think I have seen that symbol before though which is what threw me.

Thanks also for spotting that I had Q2 and Q3 voltages mixed up. Here is the hopefully corrected table:

Q1 G +6.05 [SIG]
Q1 S +5.26 [+1.5]
Q1 D +7.13 [+10]

Q2 E -10.5 [-13]
Q2 B -9.8 [-12.2]
Q2 C -6.0 [-3.8]

Q3 E +2.46 [+0.7]
Q3 B -3.39 [+1.5]
Q3 C +2.49 [+5.4]

PJL, does this mean that Q2 and Q3 may, in fact be OK?

Quote:
Originally Posted by PJL
This article should help as it suggests a J309 as suitable and there is an ebay offering from France:
https://www.eevblog.com/forum/repair/hp1725a/
Thanks for that reference. I notice that Farnell and RS do them in a SOT23 surface mount package. Any reason why I cannot use these?
http://uk.farnell.com/w/search/prl/results?st=MMBFJ309
http://uk.rs-online.com/web/p/jfet-transistors/6263314/
I think these may be too tiny to fit within the through the hole footprint and through hole would be much more convenient, but theprice is tempting. If the above are best avoided, then I will have to opt for the eBay item you pointed out

I think my substitute FET was also leaky. Without a FET in place, the voltage at the zener VR1 is 10.5v, with my replacement FET it drops to 8.35v. I also tried a jellybean in place of Q2 but it made no difference. Clearly I need to source a couple of FETs before I can proceed any further.

Last edited by WaveyDipole; 24th May 2017 at 4:51 pm.
WaveyDipole is offline  
Old 24th May 2017, 4:45 pm   #50
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: HP1715A repair

Hadn't noticed the error! The circuit voltages won't mean much while the feedback is not working.

What was the answer with the Q3 base as -3.39V looks very improbable. If it's +3.39 it's possible Q3 survives.
PJL is offline  
Old 24th May 2017, 7:24 pm   #51
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: HP1715A repair

Yes, another mistake on my part - I do seem to be making a lot of them . It is +3.39.

I have also just checked all of the resistors in the attenuator section (CH A) and all seem to be fine. It looks like switches may need cleaning. The front panel and knobs do seem to have quite a bit of grime on them. I will sort out the FET first though.
WaveyDipole is offline  
Old 24th May 2017, 8:39 pm   #52
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: HP1715A repair

Q3 may be running lots of base current but you might be lucky. The circuit looks pretty dependent on the FET characteristics but the guy who wrote that article looks pretty competent.
PJL is offline  
Old 25th May 2017, 9:20 am   #53
Humber888
Heptode
 
Humber888's Avatar
 
Join Date: Jul 2012
Location: Bromley, London, UK.
Posts: 702
Default Re: HP1715A repair

I think I have somewhere some E308 JFETs selected for various Idss values. They are the older version in epoxy of the J308s, which were the unselected versions of the 309 and 310 IIRC. If someone can work out the desired Idss to make the circuit work, I may have something near.
Humber888 is offline  
Old 25th May 2017, 2:01 pm   #54
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: HP1715A repair

Yesterday IO ordered some of those SMD J309's. At one pound for 10 from RS and free next day delivery I figured I had nothing to loose by trying. I tried one in today, but the same result. I didn't solder it to the board. Plugged 3 short equal lengths of component wire into the socket and tacked the component to the 'legs', taking note to ensure that the FET pins were connected to the correct socket on the board. It basically required that the FET be tacked onto the wires upside down.

The result, unfortunately, is practically no different.

On CH A, Q3 is soldered in, but on CH B is was plugged in and could be removed. This enabled me to test it outside of circuit and it tested good.

One thing I noticed is that with FET removed, the voltage at U1 P3 is around 2mV. With a FET plugged in, this goes up to around 200mV.

Unless I damaged the new FET while tacking it on to the wires, it would seem there is something more than just problem with the FET going on. The author in the linked thread mentions adding a couple of zenner diodes to get the biasing correct, but that was for a different FET and I don't think that is what is going on here.
WaveyDipole is offline  
Old 25th May 2017, 5:50 pm   #55
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: HP1715A repair

That's annoying. To prove the rest of the circuit is working you could replace the FET with a resistor between S and D. Because the circuit just shows rounded measured voltages, there is a bit of guesswork involved but the current is probably about 8mA which means a 1K resistor should do it.

PS: I guess it might be possible to re-bias the FET so it passes the correct current.

Last edited by PJL; 25th May 2017 at 5:55 pm.
PJL is offline  
Old 26th May 2017, 12:07 am   #56
Chris55000
Nonode
 
Join Date: Aug 2007
Location: Walsall Wood, Aldridge, Walsall, UK.
Posts: 2,853
Default Re: HP1715A repair

Hi!

I do suspect that Q1 was almost certainly "selected-on-test" by HP in the laboratory where the thing was designed - a large equipment producer like H.P. would have the clout to say to Siliconix (the main purveyor of JFETs in those days!) we want "x-hundred-thousand of these selected for this Idss and they've to be coded with our specified number!"

If Sphere or ebay haven't got any of the original house-coded ones, it's going to be a matter of trawling thro' many gen-sheets to try and find a match - if one of the originals has survived you can try it on a breadboard etc., to see what Idss was!

This page might be of help:-

http://stompville.co.uk/?p=112

By the way, I have been off work ill and it might be well into next week before I get back, but rest assured, the remaining parts of the manual will follow when I'm back at my bench!

Chris Williams
__________________
It's an enigma, that's what it is! This thing's not fixed because it doesn't want to be fixed!

Last edited by Chris55000; 26th May 2017 at 12:14 am.
Chris55000 is offline  
Old 26th May 2017, 11:26 am   #57
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: HP1715A repair

Quote:
Originally Posted by PJL View Post
That's annoying. To prove the rest of the circuit is working you could replace the FET with a resistor between S and D. Because the circuit just shows rounded measured voltages, there is a bit of guesswork involved but the current is probably about 8mA which means a 1K resistor should do it.
This test proved to be very useful and I think I am getting somewhere at last!

I think I understand how approximately 8mA was arrived at. I did some calculations and got around 9.5mA across R12 and 8.5mA from Q2E and Q8C although I'm not sure how the current across the FET itself is calculated. In any case, on substituting the resistor, the voltages were approximately correct at Q1, Q2, Q3 and Q8.

I then reviewed all the information I had to hand and discovered that the two datasheets I had on the J309 did not agree on pin assignments! So I started again and followed the layout on the Fairchild one, which proved to be correct.

I now have what looks like a correct signal on the input to A1 and correct voltages around Q1 and Q3:

Q1D +0.43
Q1G +10.13
Q1S +1.49

Q3E +0.73
Q3B +1.49
Q3C +5.10


But voltages around Q2 and Q8 are still incorrect:


Q2E -8.20
Q2B -10.35
Q2C -11.14

Q8E -2.57
Q8B -3.38
Q8C -10.35

Voltages at U1 inputs seem OK:

U1+ve 0.0v
U1-ve -0.03v

I'm mulling over the reason for this but could it still be down to the FET characteristics?

Quote:
Originally Posted by Chris55000 View Post
...if one of the originals has survived you can try it on a breadboard etc., to see what Idss was!

This page might be of help:-

http://stompville.co.uk/?p=112
Thanks for this. I had a read of it and I could certaily perform a couple of these tests on a breadboard. Unfortunately neither of the originals seems to have survived.

Quote:
Originally Posted by Chris55000 View Post
By the way, I have been off work ill and it might be well into next week before I get back, but rest assured, the remaining parts of the manual will follow when I'm back at my bench!

Chris Williams
Sorry to hear that. I wish you a speedy recovery!

Last edited by WaveyDipole; 26th May 2017 at 11:38 am.
WaveyDipole is offline  
Old 26th May 2017, 11:49 am   #58
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: HP1715A repair

I think you have the Q2E and Q2C swapped.

The figures are different because the FET is taking more current than the originals. The feedback loop is working but fairly close to one end of its working range so I would expect the scope to work but it will probably clip prematurely.
PJL is offline  
Old 26th May 2017, 2:31 pm   #59
WaveyDipole
Nonode
 
WaveyDipole's Avatar
 
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,505
Default Re: HP1715A repair

Well spotted!. I have now done the same for CH B. The signal was lost somewhere between one side and the other of C1. At first I thought it was a problem with the coupling switch, but on the DC setting I got a short across the capacitor suggesting that the switch is OK and the problem is on the board. The track between C1 and C2/R3 runs underneath so I couldn't see anything but no continuity between those two points. I soldered a wire link between them and now I had a signal to the FET. CH B behaves almost the same as CH A but the voltages are fractionally different. So I now have a signal at both inputs to A1 - but still nothing on the screen....

As before, shorting the two outputs brings the trace into the centre. Voltages at A1-Ain and A1-Bin are 0.001v and 0.002v, A1-Aout and A1-Bout are 8.1v and 7.2v respectively. Need to have a think about where to look next.
WaveyDipole is offline  
Old 26th May 2017, 3:07 pm   #60
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: HP1715A repair

The sensitivity is .005V/div at the inputs so they should be near centre on the screen. As a line is displayed when shorting the output the timebase must be running.

Back to my earlier post and page 5-14 where you will find the instructions for setting the channel amplifier offsets, Vertical preamplifier balance adjustments. Hopefully someone has fiddled with them. Put your scope on the Aout when twiddling to see if anything is changing.
PJL is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 11:13 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 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.