Protection circuit puzzle. - UK Vintage Radio Repair and Restoration Discussion Forum

yestertech · 21st Feb 2019, 4:21 pm

I'm repairing an Audiolab amp. Having replaced 2 faulty regulator transistors in the Power supply, I have now found the protection circuit is no longer working . Attached is the diagram of the protection circuit which controls the speaker relay I'm struggling with just how this works. I can see there appears to be a series of diode 'or' gates None of the levels where the didoes merge look to be at an 'active' level and it had been working, so I am convinced the fault lies somewhere in the circuitry to the right of this, but I can't fathom what's going on.
Any ideas ?

Andy

GMB · 21st Feb 2019, 4:52 pm

I don't think anyone will fathom what is going on unless you know what feeds into this. It is just a load of logic, combining signals some directly and some with a time delay aspect, to eventually work a relay.

So when you say it doesn't work, which of its many functions does it not do?

Lucien Nunes · 21st Feb 2019, 6:32 pm

When the amp is powered up, R530 || R531 charge C509 towards halfway between (514) and (516). When C509 reaches ~13.4V, Q515 and hence Q516 begin to conduct, energising the relay. R532 provides hysteresis to make it snap on promptly. This section appears to provide a power-on delay before the relay closes.

Diodes D513 - 516 each conduct in case of one fault condition, discharging C509 rapidly via R539 and tripping the relay off, by turning on either Q511 and hence Q512 or Q513 and hence Q514.

A) If (513) does not sink enough current to hold D512(k) <1.4V, R529 forward biases D513. This seems to be some kind of status monitor with unknown function.

B) If the average of the voltages on (511) and (512), weighted by the ratio 47:56 due to R526 and R527, charges C506/7 >2V positive, D514 becomes forward biased.

C) As per B), but when C506/7 >1.4V negative, D516 becomes foward biased. This may be supply voltage symmetry monitoring.

D) Either Q507 or Q508 charge C505 >7V, Q509 and hence Q510 conduct, with hysteresis from R521, forward biasing D515. This appears to be DC / output anomaly detection for both channels.

JohnBHanson · 21st Feb 2019, 6:34 pm

The diodes DS13m DS14, DS15 form an or gate. I high on any of the anodes of these cause the relay to trip i.e. de-energise.

Also a negative voltage on DS16 will cause a trip. (With respect to connection 516).

Similarly DS09 and DS10 form an or gate.

yestertech · 22nd Feb 2019, 7:09 pm

My thanks to Lucien for a detailed description. As to what this unit monitors :

501 and 509 monitor the +ve and -ve rails to the amp
503 and 507 monitor the 'top end' of the emitter resistors in the o/p stage
505 monitors the LS output ( mid resistors )
513 monitors the AC secondary of the power transformer

I'm still tracing what 511/512 monitor
I wil also take some voltage readings to clarify

A.

mhennessy · 26th Feb 2019, 3:00 pm

511 and 512 will come from the speaker outputs - it's a standard DC protection circuit that disconnects the speakers in the event of a DC offset in either of the amplifier channels.

505 is just to give a connection to the junction of the two current-sense resistors in-between the output transistors. An overload could occur in either transistor.

501 and 509 aren't monitoring the power supply rails as such - they are just brought over to this circuit to allow the respective transistors to do their work.

yestertech · 27th Feb 2019, 2:20 pm

Many thanks to Mark and all other contributors to this thread. I have now got my head around the protection circuit– upon testing, the base voltage of Q515 was not rising above 8 V. It appears that due to an error in the readout diagrams for Q514 ( BC5xx ) which I had initially replaced due to its 'short' indications on test, I managed to get the replacement transistor 180 degrees out. Worse, getting to the underside of the PCB involved an enormous amount of dismantling as the connections are hidden under a chassis sectionr ! Once reassembled yet again, the protection circuit operates properly.
The rub was that all this time I was looking in entirely the wrong place for a power supply fault !
Having done some more tracing of the PCB layout I located the two power regulators for the 30 V and 10 V rails, Both of which were defective.
So , FINALLY the amplifier is now working.
All that remains os to replace the font panel including reattaching the LED connections which are integral to the panel , two covers and a huge pile of screws.
Hey ho....

A.

21st Feb 2019, 4:52 pm	#2
GMB Dekatron Join Date: Aug 2003 Location: near Reading (and sometimes Torquay) Posts: 3,099	Re: protection circuit puzzle I don't think anyone will fathom what is going on unless you know what feeds into this. It is just a load of logic, combining signals some directly and some with a time delay aspect, to eventually work a relay. So when you say it doesn't work, which of its many functions does it not do?

21st Feb 2019, 6:32 pm	#3
Lucien Nunes Rest in Peace Join Date: Apr 2008 Location: London, UK. Posts: 2,508	Re: protection circuit puzzle When the amp is powered up, R530 \|\| R531 charge C509 towards halfway between (514) and (516). When C509 reaches ~13.4V, Q515 and hence Q516 begin to conduct, energising the relay. R532 provides hysteresis to make it snap on promptly. This section appears to provide a power-on delay before the relay closes. Diodes D513 - 516 each conduct in case of one fault condition, discharging C509 rapidly via R539 and tripping the relay off, by turning on either Q511 and hence Q512 or Q513 and hence Q514. A) If (513) does not sink enough current to hold D512(k) <1.4V, R529 forward biases D513. This seems to be some kind of status monitor with unknown function. B) If the average of the voltages on (511) and (512), weighted by the ratio 47:56 due to R526 and R527, charges C506/7 >2V positive, D514 becomes forward biased. C) As per B), but when C506/7 >1.4V negative, D516 becomes foward biased. This may be supply voltage symmetry monitoring. D) Either Q507 or Q508 charge C505 >7V, Q509 and hence Q510 conduct, with hysteresis from R521, forward biasing D515. This appears to be DC / output anomaly detection for both channels.

21st Feb 2019, 6:34 pm	#4
JohnBHanson Heptode Join Date: Aug 2009 Location: Worthing, Sussex, UK. Posts: 662	Re: protection circuit puzzle The diodes DS13m DS14, DS15 form an or gate. I high on any of the anodes of these cause the relay to trip i.e. de-energise. Also a negative voltage on DS16 will cause a trip. (With respect to connection 516). Similarly DS09 and DS10 form an or gate.

22nd Feb 2019, 7:09 pm	#5
yestertech Nonode Join Date: Dec 2002 Location: Coulsdon, Surrey, UK. Posts: 2,135	Re: Protection circuit puzzle. My thanks to Lucien for a detailed description. As to what this unit monitors : 501 and 509 monitor the +ve and -ve rails to the amp 503 and 507 monitor the 'top end' of the emitter resistors in the o/p stage 505 monitors the LS output ( mid resistors ) 513 monitors the AC secondary of the power transformer I'm still tracing what 511/512 monitor I wil also take some voltage readings to clarify A.

26th Feb 2019, 3:00 pm	#6
mhennessy Dekatron Join Date: Jul 2007 Location: Evesham, Worcestershire, UK. Posts: 4,244	Re: Protection circuit puzzle. 511 and 512 will come from the speaker outputs - it's a standard DC protection circuit that disconnects the speakers in the event of a DC offset in either of the amplifier channels. 505 is just to give a connection to the junction of the two current-sense resistors in-between the output transistors. An overload could occur in either transistor. 501 and 509 aren't monitoring the power supply rails as such - they are just brought over to this circuit to allow the respective transistors to do their work.

27th Feb 2019, 2:20 pm	#7
yestertech Nonode Join Date: Dec 2002 Location: Coulsdon, Surrey, UK. Posts: 2,135	Re: Protection circuit puzzle. Many thanks to Mark and all other contributors to this thread. I have now got my head around the protection circuit– upon testing, the base voltage of Q515 was not rising above 8 V. It appears that due to an error in the readout diagrams for Q514 ( BC5xx ) which I had initially replaced due to its 'short' indications on test, I managed to get the replacement transistor 180 degrees out. Worse, getting to the underside of the PCB involved an enormous amount of dismantling as the connections are hidden under a chassis sectionr ! Once reassembled yet again, the protection circuit operates properly. The rub was that all this time I was looking in entirely the wrong place for a power supply fault ! Having done some more tracing of the PCB layout I located the two power regulators for the 30 V and 10 V rails, Both of which were defective. So , FINALLY the amplifier is now working. All that remains os to replace the font panel including reattaching the LED connections which are integral to the panel , two covers and a huge pile of screws. Hey ho.... A.