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 Audio (record players, hi-fi etc)

Notices

Vintage Audio (record players, hi-fi etc) Amplifiers, speakers, gramophones and other audio equipment.

Closed Thread
 
Thread Tools
Old 12th Dec 2018, 7:42 am   #1
GW4FRX
Pentode
 
GW4FRX's Avatar
 
Join Date: Aug 2005
Location: Shrewsbury, Shropshire, UK.
Posts: 149
Default Radford SPA.50 resistor values?

For the last year or so I’ve been restoring a very tired Radford SPA.50 amplifier. Amongst many other things the 2N4906/4915 complementary pairs in the amplifier were defective and have been changed for MJ15003/15004s.

Recently I’ve been doing some preliminary tests on the audio channel boards and run into a bit of an oddity. On the original schematic (attached) there’s a resistor chain from the collector of Ts2 to ground in which P2 sets the quiescent current of the output devices. According to the schematic this chain totals 7K4. However, with the values of the resistors actually found on the PCBs the actual values total 31K4. Specifically, R8 is 2K2, R9 is 18K and R10 is 8K2.

The value of quiescent current for best IMD performance in this amplifier is supposed to correspond to 18mV of bias developed across R25 and R26 (i.e. measured between the collectors of Ts10 and 11). However, I can’t get anywhere near that. Using a Farnell 2 x 30V current-limited supply for test purposes, the achievable control range via P2 is about 230 to 500mV. This gives an overall quiescent current of about 300mA, which seems decidedly on the high side for a 50W Class AB audio amplifier.

I’ve been trying to work out from first principles what resistors to change to give a control range of say 0 to 100mV or thereabouts but have failed miserably. Can anyone skilled in the art please advise how to go about it?
Attached Thumbnails
Click image for larger version

Name:	SPA.50 original.jpg
Views:	702
Size:	75.0 KB
ID:	174423  
GW4FRX is offline  
Old 12th Dec 2018, 8:54 am   #2
ukcol
Rest in Peace
 
ukcol's Avatar
 
Join Date: May 2003
Location: Harlaxton, Lincolnshire, UK.
Posts: 3,944
Default Re: Radford SPA.50 resistor values

Do the resistors R8, R9 & R10 look like factory fitted originals?

I would do a cold "diode" test on all the semiconductors. An open circuit base/collector junction in Ts3 could be responsible for example.

Also check that none of the carbon resistors have changed value as a result of the previous fault.
ukcol is offline  
Old 12th Dec 2018, 11:00 am   #3
GW4FRX
Pentode
 
GW4FRX's Avatar
 
Join Date: Aug 2005
Location: Shrewsbury, Shropshire, UK.
Posts: 149
Default Re: Radford SPA.50 resistor values

Quote:
Originally Posted by ukcol View Post
Do the resistors R8, R9 & R10 look like factory fitted originals?
Yes, on both boards.

Quote:
I would do a cold "diode" test on all the semiconductors. An open circuit base/collector junction in Ts3 could be responsible for example. Also check that none of the carbon resistors have changed value as a result of the previous fault.
In fact Ts3 has been changed on both boards. On one there was a control range of only about 4mA; changing Ts3 fixed that. The other one was changed because it was an unmarked plastic device and clearly wasn't a BC108. I have also changed the BFY60 and BFY90 on one board in view of the unreliability of Lockfit devices. The two SE.1 catching diodes on the output were both o/c and have been changed for BYV27s. All other transistor and diode junctions test OK.

All resistor values have been checked and on one board the four electrolytics have also been changed. Both boards exhibit the same control range and P1 sets the half-rail voltage as expected.
GW4FRX is offline  
Old 12th Dec 2018, 11:36 am   #4
Herald1360
Dekatron
 
Herald1360's Avatar
 
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,528
Default Re: Radford SPA.50 resistor values?

Just in case (ignore this if it's superfluous), it is possible for transistors to go C-E short or leaky whilst still showing "good" readings E-B and C-B. This can upset bias conditions somewhat!
__________________
....__________
....|____||__|__\_____
.=.| _---\__|__|_---_|.
.........O..Chris....O
Herald1360 is offline  
Old 12th Dec 2018, 1:20 pm   #5
GW4FRX
Pentode
 
GW4FRX's Avatar
 
Join Date: Aug 2005
Location: Shrewsbury, Shropshire, UK.
Posts: 149
Default Re: Radford SPA.50 resistor values?

Just to clarify, my strategy was to change various components on one board that had failed or fried and leave the other one as near original as possible. So Ts3, 4, 6 and 7 were changed on one board for new or NOS components together with two or three resistors 4, the four electrolytics and D3 and 4. All semiconductors have been tested with the 'cheap Chinese tester' and measure OK.

Both boards seem to perform very similarly and I'm inclined to think they're performing more or less as they should apart from high quiescent current (i.e. too many millivolts between the collectors).

I'm wondering whether the problem might be associated with the change to MJ15003/4.
GW4FRX is offline  
Old 12th Dec 2018, 2:20 pm   #6
Herald1360
Dekatron
 
Herald1360's Avatar
 
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,528
Default Re: Radford SPA.50 resistor values?

It's possible- they may have somewhat lower VBE ON than the originals. It may be worth carefully dropping the value of their B-E resistors (collector loads of the drivers) to see whether this brings things back into range quiescent current wise.

It would be interesting to simulate the circuit with various different output devices to see how sensitive the quiescent current is with different types.
__________________
....__________
....|____||__|__\_____
.=.| _---\__|__|_---_|.
.........O..Chris....O

Last edited by Herald1360; 12th Dec 2018 at 2:28 pm.
Herald1360 is offline  
Old 12th Dec 2018, 4:52 pm   #7
GW4FRX
Pentode
 
GW4FRX's Avatar
 
Join Date: Aug 2005
Location: Shrewsbury, Shropshire, UK.
Posts: 149
Default Re: Radford SPA.50 resistor values?

Quote:
Originally Posted by Herald1360 View Post
It's possible- they may have somewhat lower VBE ON than the originals. It may be worth carefully dropping the value of their B-E resistors (collector loads of the drivers) to see whether this brings things back into range quiescent current wise.
That's a thought. Would that be R24 and R27 on the schematic?

Quote:
It would be interesting to simulate the circuit with various different output devices to see how sensitive the quiescent current is with different types.
It would indeed, but alas that sort of thing is well outside my limited skill set! If any kind soul on here is skilled in such arts I'd be exceedingly grateful for any assistance.

I'm slightly curious as to why the faulty Ts3 initially gave a very restricted control range of 4-8mV which nevertheless was a lot nearer the required value of 18mV than the minimum of 230mV currently available. The Chinese tester declined to indicate an hfe value for it (which is why I changed it) but something about it evidently had some influence over the proceedings.
GW4FRX is offline  
Old 12th Dec 2018, 5:15 pm   #8
Herald1360
Dekatron
 
Herald1360's Avatar
 
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,528
Default Re: Radford SPA.50 resistor values?

Yes, R24 and R27.

A shorted Ts3 would reduce the standing current since it's the voltage between the bases of Ts6 and Ts7 which controls this. Turning Ts3 on reduces it and off increases it. Ts3 is actually living up to its name as a transfer resistor.
__________________
....__________
....|____||__|__\_____
.=.| _---\__|__|_---_|.
.........O..Chris....O
Herald1360 is offline  
Old 12th Dec 2018, 5:18 pm   #9
vidjoman
Dekatron
 
Join Date: Jun 2013
Location: East Sussex, UK.
Posts: 3,315
Default Re: Radford SPA.50 resistor values?

Back in the late 60's to mid 70's I worked for a shop that sold Radford amps. There were numerous 'versions' of the amps. Can't remember the model numbers but I think they had a letter assigned to them to work out which version you had. I know the boss said enough is enough and stopped selling them by the time it got to version G. Servicing was always a nightmare to try and work from a diagram as they were all different.
vidjoman is offline  
Old 12th Dec 2018, 5:32 pm   #10
GW4FRX
Pentode
 
GW4FRX's Avatar
 
Join Date: Aug 2005
Location: Shrewsbury, Shropshire, UK.
Posts: 149
Default Re: Radford SPA.50 resistor values?

That's interesting. The PCBs on my example have several extra resistors not shown on the schematic, not to mention some with wildly different values. Oh, and Ts6 and Ts7 were different as between my boards (40594/5 on one and the stated 40361/2 on the other, the latter now changed because they were both o/c and I had 40594 and 5 available. By all accounts they were slightly better devices).

I'm determined to get it going though, having built a completely new power supply board for it with a 723 and 3423 crowbar. Both were just about contemporary with the amplifier so I feel mildly justified. And the overcurrent trip works, unlike the original...

Last edited by GW4FRX; 12th Dec 2018 at 5:38 pm.
GW4FRX is offline  
Old 12th Dec 2018, 5:36 pm   #11
GW4FRX
Pentode
 
GW4FRX's Avatar
 
Join Date: Aug 2005
Location: Shrewsbury, Shropshire, UK.
Posts: 149
Default Re: Radford SPA.50 resistor values?

Quote:
Originally Posted by Herald1360 View Post
Yes, R24 and R27. A shorted Ts3 would reduce the standing current since it's the voltage between the bases of Ts6 and Ts7 which controls this. Turning Ts3 on reduces it and off increases it. Ts3 is actually living up to its name as a transfer resistor.
Many thanks. I'd very much like to know why there's a huge disparity in resistor values between the boards and the schematic in this area. Would changing the ratio of R9 and R10 achieve anything useful?
GW4FRX is offline  
Old 12th Dec 2018, 6:40 pm   #12
Herald1360
Dekatron
 
Herald1360's Avatar
 
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,528
Default Re: Radford SPA.50 resistor values?

I've gone for a lengthy mouse clicking session and loaded the circuit into a simulator (Simetrix v3.0, ancient but useful). The most interesting finding is that if the values of R8, 9 and 10 are as per circuit, the quiescent current sets up nicely for the 18mV figure but with the values you gave for your board, it sits much higher with only limited adjustment range. Sound familiar?

I have no idea what's going on with the circuit differences, though in principle the higher values of the resistor chain just don't seem right for their application.

I had to pick arbitrary transistors, not having models for the actual types, but I did find that changing output types had very little effect on dc conditions.

Types I used were BFY51, BFX29, BC108, BD139, BD140, MJ15025, MJ15024.

I left out the output protection clamping circuits.
__________________
....__________
....|____||__|__\_____
.=.| _---\__|__|_---_|.
.........O..Chris....O
Herald1360 is offline  
Old 12th Dec 2018, 6:46 pm   #13
GW4FRX
Pentode
 
GW4FRX's Avatar
 
Join Date: Aug 2005
Location: Shrewsbury, Shropshire, UK.
Posts: 149
Default Re: Radford SPA.50 resistor values?

Quote:
Originally Posted by Herald1360 View Post
I've gone for a lengthy mouse clicking session and loaded the circuit into a simulator (Simetrix v3.0, ancient but useful). The most interesting finding is that if the values of R8, 9 and 10 are as per circuit, the quiescent current sets up nicely for the 18mV figure but with the values you gave for your board, it sits much higher with only limited adjustment range. Sound familiar?
Horribly. Very many thanks for doing that -- I'm immensely grateful.

Quote:
I have no idea what's going on with the circuit differences, though in principle the higher values of the resistor chain just don't seem right for their application. I had to pick arbitrary transistors, not having models for the actual types, but I did find that changing output types had very little effect on DC conditions. Types I used were BFY51, BFX29, BC108, BD139, BD140, MJ15025, MJ15024. I left out the output protection clamping circuits.
That's exceedingly interesting and suggests a way forward. I'll break out the trusty kettle-mender after supper and change the resistor chain back to the original values. Watch this space...
GW4FRX is offline  
Old 12th Dec 2018, 7:54 pm   #14
Trevor
Octode
 
Join Date: Aug 2006
Location: Stoke-on-Trent, Staffordshire, UK.
Posts: 1,082
Default Re: Radford SPA.50 resistor values?

Very interesting read
Re the fake transistors about two years ago i repaired an Sp50 fitted with Sp60 boards
When purchased it was a right mess wrong transistors reduced operating voltage But the main problem was fake transistors never purchase from that auction site repaired tested ok switched it off then switched it on and both channels blew out
went back to my old stocks refitted and perfect operation
Sp60 boards are very similar but have base drive dampers fitted and a min distortion preset
Paid to much for the amplifier considering the state of it when delivered but it does sound nice
just fit the schematic value resistors in the bias chain and I am sure it will work well

Trev
Trevor is offline  
Old 12th Dec 2018, 8:17 pm   #15
GW4FRX
Pentode
 
GW4FRX's Avatar
 
Join Date: Aug 2005
Location: Shrewsbury, Shropshire, UK.
Posts: 149
Default Re: Radford SPA.50 resistor values?

Thanks very much for that, which inspired me to have a look at the SPA.60 schematic. As you say it's quite similar but the resistor values around the Vbe multiplier are much more like those on the original SPA.50 drawing rather than those on the PCBs in my amplifier. All very strange.

Rather than attack the rather fragile board immediately, I'll tack some parallel resistors across those currently on the PCB and see what happens.
GW4FRX is offline  
Old 12th Dec 2018, 9:41 pm   #16
GW4FRX
Pentode
 
GW4FRX's Avatar
 
Join Date: Aug 2005
Location: Shrewsbury, Shropshire, UK.
Posts: 149
Default Re: Radford SPA.50 resistor values?

Quote:
Originally Posted by Herald1360 View Post
I've gone for a lengthy mouse clicking session and loaded the circuit into a simulator (Simetrix v3.0, ancient but useful). The most interesting finding is that if the values of R8, 9 and 10 are as per circuit, the quiescent current sets up nicely for the 18mV figure but with the values you gave for your board, it sits much higher with only limited adjustment range. Sound familiar?
Further to my last -- you da man!

I tacked 110R across R8, 4K3 across R9 and 1K3 across R10 to produce the approximate values given in the schematic. In terms of differential voltage across Ts10 and Ts11 collectors, the differential is now variable from 4mV to about 500mV. At 18mV or thereabouts, the board quiescent current is about 90mA. I've no way of knowing whether that's normal for this amplifier but we'll see what happens with audio testing.

The only minor issue now is that P2 is exceedingly critical in adjustment. Even touching it with a screwdriver is apt to shift the differential by 10mV or so and there's really far more control range than necessary. Any suggestions as to how to make it slightly more tractable? Something like say 0-100mV would be ideal.

Again my very grateful thanks for your help with this. One wonders what Mr Radford was playing at...
GW4FRX is offline  
Old 13th Dec 2018, 12:34 am   #17
Herald1360
Dekatron
 
Herald1360's Avatar
 
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,528
Default Re: Radford SPA.50 resistor values?

Glad to be of some assistance!

You'll need to figure out/measure the exact values of the components of the bias chain for the right setting (including each "half" of the pot), then reorganise the chain resistor values so that the 18 mV comes up somewhere around mid travel of a 100R pot, whilst maintaining the total chain value the same as before. That should give you a more limited adjustment range around the correct value.

Or you could fit a ten turn 1k trimmer to give you better adjustment resolution.
__________________
....__________
....|____||__|__\_____
.=.| _---\__|__|_---_|.
.........O..Chris....O
Herald1360 is offline  
Old 13th Dec 2018, 10:37 am   #18
GW4FRX
Pentode
 
GW4FRX's Avatar
 
Join Date: Aug 2005
Location: Shrewsbury, Shropshire, UK.
Posts: 149
Default Re: Radford SPA.50 resistor values?

A nice thought but I was rather hoping to keep the 'look' of the PCB reasonably original and there isn't really the space for a ten-turn. Also, P2 does seem to want to sit at about half its travel to give 20mV or so. But I noticed late last night that bringing a hand near the meter leads affects the differential voltage and the setting of P2 is unbelievably twitchy. A bit of messing about with decoupling capacitors didn't immediately change anything but possibly there's some RF pickup or instability that needs investigating.

At least the differential voltage is controllably in the right range now, thanks to your good self, so we're getting somewhere. Onwards and upwards...
GW4FRX is offline  
Old 14th Dec 2018, 12:19 am   #19
Herald1360
Dekatron
 
Herald1360's Avatar
 
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,528
Default Re: Radford SPA.50 resistor values?

OK, I've had another play. With my sim and the devices I used, I reckon you could try R9 3k3, shunt P2 with 120R across ends and make R10 1k5. In the sim this gives an adjustment from about 5mV to 85mV in the Ts10c to Ts11c voltage difference.

18mV gives about 17mA in Ts10 and about 8mA in Ts11. I think the disparity is caused by the dc feedback path loading via R13, R6 and R7.

The amp does seem to be very nearly in class B. I wonder if quiescent current was increased at some point to get better low level crossover distortion figures? Upping the standing current to 50 mA or so would only give standing dissipations of around 2W for each output device.

With the actual devices in the amp, you may need to play around a bit with R9 and R10 to get the 18mV point near central on the pot travel- just increase one and decrease the other by the same amount if needed.

I hope the practice comes out somewhere near the theory!
__________________
....__________
....|____||__|__\_____
.=.| _---\__|__|_---_|.
.........O..Chris....O
Herald1360 is offline  
Old 14th Dec 2018, 6:41 am   #20
GW4FRX
Pentode
 
GW4FRX's Avatar
 
Join Date: Aug 2005
Location: Shrewsbury, Shropshire, UK.
Posts: 149
Default Re: Radford SPA.50 resistor values?

You're a megastar -- very many thanks. Investigations will proceed along those lines over the weekend!

I did wonder about the differential voltage and the standing current to which it gave rise. The 18mV figure came from someone who'd apparently optimised the THD performance using an analyser but quite what Radford used in production is presumably lost in the mists of time.

The PSU in this amplifier was a wreck. Pic of posh new one attached for amusement.
Attached Thumbnails
Click image for larger version

Name:	JHN final build -- 28 Nov 18.jpg
Views:	124
Size:	115.5 KB
ID:	174556  
GW4FRX is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 6:26 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.