0.18 ohm are the ones 
and being so low they would be hard to measure accurately because that resistance is comparable to typical test lead resistance and contact resistance. That said, still worth a measure. Amp OFF of course for this.
The voltage is the voltage measured across each resistor with the amp on and speakers connected. Firstly with no music playing and then with it playing under the fault condition.
and being so low they would be hard to measure accurately because that resistance is comparable to typical test lead resistance and contact resistance. That said, still worth a measure. Amp OFF of course for this. The voltage is the voltage measured across each resistor with the amp on and speakers connected. Firstly with no music playing and then with it playing under the fault condition.
Just tested the volt drop across left and right channel with and without music playing, here are the results;
Left channel no music Right channel no music
R35 - 16mv R35 - 16mv
R36 - 3mv R36 - 3mv
Left channel with music Right channel with music
R35 - varies between 15 - 47mv R35 - varies between 18 - 51mv
R36 - " " 9 - 27mv R36 - " " 7 - 31mv
Left channel no music Right channel no music
R35 - 16mv R35 - 16mv
R36 - 3mv R36 - 3mv
Left channel with music Right channel with music
R35 - varies between 15 - 47mv R35 - varies between 18 - 51mv
R36 - " " 9 - 27mv R36 - " " 7 - 31mv
Thanks. Those are certainly in the right ballpark and so a little inconclusive in one sense, and yet also at face value seem to show both channels behaving similarly.
16mv across 0.18 ohms is 88ma. All of that should be flowing in TR7 and there should be no current flowing in the output pair. TR7 would be dissipating around 4 watts
Silly question (maybe) but when you say the heatsink get hotter on one side... is it the output transistors getting hot or the Class A driver ? Probably not easy to tell.
The Quad can operate up to 1 watt output (that is more than you might think volume wise) without the outputs doing anything... in fact they only come into play at levels over this and so in themselves should be cold up to that point.
It looks like we have drawn a blank without a scope to look at what is really going on under dynamic conditions I'm afraid. Its an interesting fault though.
16mv across 0.18 ohms is 88ma. All of that should be flowing in TR7 and there should be no current flowing in the output pair. TR7 would be dissipating around 4 watts
Silly question (maybe) but when you say the heatsink get hotter on one side... is it the output transistors getting hot or the Class A driver ? Probably not easy to tell.
The Quad can operate up to 1 watt output (that is more than you might think volume wise) without the outputs doing anything... in fact they only come into play at levels over this and so in themselves should be cold up to that point.
It looks like we have drawn a blank without a scope to look at what is really going on under dynamic conditions I'm afraid. Its an interesting fault though.
One final thought... post #7 and that diode D6. The voltage drop across that diode is used to bring the output stage closer to the point of beginning to conduct. Earlier models didn't use it (effectively it was a wire link if you look in the manual) and only the later models had it added. The voltage drop is fairly critical, if its a little high then the outputs will conduct more than they should, effectively running in a high bias Class AB condition. If the diode is suspect and its forward voltage a little more than it should be when passing current then it would give rise to these symptoms music and levels depending. It would be a dynamic rather than a static problem.
Do you have a general purpose silicon diode like a 1N4004 that you could tag across D6 ? No need to remove the old one. For the sake of a diode it has to be worth a try. Its not critical either what the part is.
Do you have a general purpose silicon diode like a 1N4004 that you could tag across D6 ? No need to remove the old one. For the sake of a diode it has to be worth a try. Its not critical either what the part is.
While running the amp without its lid when checking R35/36 I felt the temperature of the class a drivers and output transistors. Both class a drivers were warm and the output transistors cold when the amp had no input signal, the latter warming up once music above 1 watt was played. The right side output pair becoming warmer than the left the longer music was left on.
I have a suitable silicone diode to replace D6 but it tests exactly the same as the one presently fitted. Could D6 be failing under load ? Can the voltage drop be checked in situ` ?
It does seem to be an interesting problem which requires further investigation with a scope etc, but many thanks for your time and effort in attempting to find the cause. I`ll be very interested to eventually learn what the issue was.
I have a suitable silicone diode to replace D6 but it tests exactly the same as the one presently fitted. Could D6 be failing under load ? Can the voltage drop be checked in situ` ?
It does seem to be an interesting problem which requires further investigation with a scope etc, but many thanks for your time and effort in attempting to find the cause. I`ll be very interested to eventually learn what the issue was.
Diodes can fail in strange ways that relate to the current through them, and although its not really a 'stressed' location you do have to think of the initial fault and any collateral damage that might have occurred at the time the original outputs failed.
You can measure volt drop in circuit although it doesn't tell the full story of what is happening in response to a music signal under the varying currents (a DVM just doesn't respond quickly enough) and so it is a more certain test to just eliminate the part by substitution.
The diode can even be shorted out as a test which would force a zero bias condition at all times on the output stage. It was added to bring the stage nearer to the point of conduction and so reduce the crossover area that the Class A stage 'fills in'.
Voltages across D6 of more than around 1.2 volts will cause the output stage to begin to conduct
You can measure volt drop in circuit although it doesn't tell the full story of what is happening in response to a music signal under the varying currents (a DVM just doesn't respond quickly enough) and so it is a more certain test to just eliminate the part by substitution.
The diode can even be shorted out as a test which would force a zero bias condition at all times on the output stage. It was added to bring the stage nearer to the point of conduction and so reduce the crossover area that the Class A stage 'fills in'.
Voltages across D6 of more than around 1.2 volts will cause the output stage to begin to conduct
Hi Mooly (and all),
My apologies for the very late update on the overheating Quad but here it is as promised.
Before sending the amp back to Quad themselves a friend suggested using Dada in Holland which is what I did as their quoted lead time was half that of Quad.
So after initially contacting the helpful and knowledgable Joost at Dada the amp was sent via UPS to him (arrived in the same time frame as a UK to UK parcel) for repair. After a week Joost contacted me to say the problem was indeed being caused by high frequency oscillation in the right channel and he had repaired the issue. He says the use of MJ15003G high gain version coupled with MJ15031 class A drivers was the issue as "405-2`s circuits are very sensitive to open loop gain". He has replaced all 4 MJ15003`s with the standard gain version and changed all 4 class A drivers to BD242C`s which are his preferred transistors. The amp now has a balanced heatsink and sounds wonderful.
I was very impressed with Joost who kept me informed all the way through the repair, which was completed faster than I expected. He also replaced all the screws holding the boards in place with the correct Philips type and even fitted 4 new rubber feet ! His charge for all this work was less than half the price of the original "repairer" and also included the 8 new transistors. He charged 1 hours labour plus the parts whereas the original "repairer" wanted 4 hours labour @ £45 per hour to swap 4 transistors plus the price of the parts - and this guy`s reputed to be a competent engineer !!!
If anyone requires repair/upgrade work to their Quad equipment I would wholeheartedly recommend Joost @ Dada.
Thankyou to all the forum members who contributed to this post, it was much appreciated.
David.
My apologies for the very late update on the overheating Quad but here it is as promised.
Before sending the amp back to Quad themselves a friend suggested using Dada in Holland which is what I did as their quoted lead time was half that of Quad.
So after initially contacting the helpful and knowledgable Joost at Dada the amp was sent via UPS to him (arrived in the same time frame as a UK to UK parcel) for repair. After a week Joost contacted me to say the problem was indeed being caused by high frequency oscillation in the right channel and he had repaired the issue. He says the use of MJ15003G high gain version coupled with MJ15031 class A drivers was the issue as "405-2`s circuits are very sensitive to open loop gain". He has replaced all 4 MJ15003`s with the standard gain version and changed all 4 class A drivers to BD242C`s which are his preferred transistors. The amp now has a balanced heatsink and sounds wonderful
.I was very impressed with Joost who kept me informed all the way through the repair, which was completed faster than I expected. He also replaced all the screws holding the boards in place with the correct Philips type and even fitted 4 new rubber feet ! His charge for all this work was less than half the price of the original "repairer" and also included the 8 new transistors. He charged 1 hours labour plus the parts whereas the original "repairer" wanted 4 hours labour @ £45 per hour to swap 4 transistors plus the price of the parts - and this guy`s reputed to be a competent engineer !!!
If anyone requires repair/upgrade work to their Quad equipment I would wholeheartedly recommend Joost @ Dada.
Thankyou to all the forum members who contributed to this post, it was much appreciated.
David.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.