Man I've been poking at this one for awhile..
This amp does not appear to be going into protection but does not start up reliably. The output stages do not start up and go into oscillation - at first..
After many power cycles and repeated attempts to boot it, it eventually does start firing up and works as expected on all 4 channels. It then starts reliably until I let it sit for some hours/day.
I eventually found that if I discharge the 4 bias capacitors (150u), I can then get the amp to immediately go back to malfunctioning.
I have struggled to find a similar design schematic but believe I finally found something similar in the uCD Super Lite design.
There are no off the shelf half bridge drivers in the amp, it is essentially a discreet design.. but as I said the Super Lite design seems quite close at least for the high and low side drivers. I need to dig further obviously..
Things I have eliminated:
Took the amp from a non working state, sat for days. Tried to boot with a very high amperage supply. Same.
Added speaker loads to outputs while trying to boot from a bad state. Same.
Looked over board very thoroughly etc etc.
Attached basic pic from internet, can grab more pics later.
Attached pic from Super Lite schematic. C20 is similar to the bias capacitors on the focal.
This amp does not appear to be going into protection but does not start up reliably. The output stages do not start up and go into oscillation - at first..
After many power cycles and repeated attempts to boot it, it eventually does start firing up and works as expected on all 4 channels. It then starts reliably until I let it sit for some hours/day.
I eventually found that if I discharge the 4 bias capacitors (150u), I can then get the amp to immediately go back to malfunctioning.
I have struggled to find a similar design schematic but believe I finally found something similar in the uCD Super Lite design.
There are no off the shelf half bridge drivers in the amp, it is essentially a discreet design.. but as I said the Super Lite design seems quite close at least for the high and low side drivers. I need to dig further obviously..
Things I have eliminated:
Took the amp from a non working state, sat for days. Tried to boot with a very high amperage supply. Same.
Added speaker loads to outputs while trying to boot from a bad state. Same.
Looked over board very thoroughly etc etc.
Attached basic pic from internet, can grab more pics later.
Attached pic from Super Lite schematic. C20 is similar to the bias capacitors on the focal.
Attachments
Are you saying that, if you power the amp up and have a drive signal reaching the input to the class D circuit that it will not oscillate?
The focal has a clock generator circuit that seems different from the Super Lite.
The driver circuitry seems similar.
Attached Super Lite schematic.
The driver circuitry seems similar.
Attached Super Lite schematic.
Having a clock would change everything.
After it oscillates, is the carrier frequency rock-solid the same frequency regardless of how hard it's driven (loaded or not)?
Or is the carrier frequency variable?
After it oscillates, is the carrier frequency rock-solid the same frequency regardless of how hard it's driven (loaded or not)?
Or is the carrier frequency variable?
The waveform moves quite a bit with signal at the output drivers, but I'd expect that.
Lets see if I can track back the clock and probe there.
Lets see if I can track back the clock and probe there.
A class D amp with a clock shouldn't have problems starting unless the clock isn't starting. Does the clock run reliably?
Where is the connected on the diagram?
Clocked class D amps don't vary in frequency. They only vary in pulse width.
If you added bleed resistors to the capacitors you drained (circuit designation on diagram?), would the amp start reliably? Use the highest value that would make it reliable.
Where is the connected on the diagram?
Clocked class D amps don't vary in frequency. They only vary in pulse width.
If you added bleed resistors to the capacitors you drained (circuit designation on diagram?), would the amp start reliably? Use the highest value that would make it reliable.
It seems like the amp does not like to start once those caps are bled, I'd think that would make it worse?
Yes the output stages frequency is steady at ~370K. Pulse width moves. Audio behaves as expected on the Dscope III.
Yes the output stages frequency is steady at ~370K. Pulse width moves. Audio behaves as expected on the Dscope III.
Clock is there at c955 and c954 even on bad boot (which is shipped out to all 4 channels). It seems like something in the driver circuitry is latching up or not starting like it should.
Last edited:
Alright.. tried new caps since i did have some 150u's...same.
R31 in the schematic on post1 is a 12k in the focal. Each channel has a diode then a 12k feeding the bootstrap/high driver on the focal.
I'm going to try 10k or so on those resistors.
R31 in the schematic on post1 is a 12k in the focal. Each channel has a diode then a 12k feeding the bootstrap/high driver on the focal.
I'm going to try 10k or so on those resistors.
That didnt work, back to 12k.
So....
If I remove the bias/bootstrap caps on the high side drivers, it starts up dependably.
Must be some other circuit that isn't happy unless the bootstrap caps are charged.... I'm thinking? Weird.
So....
If I remove the bias/bootstrap caps on the high side drivers, it starts up dependably.
Must be some other circuit that isn't happy unless the bootstrap caps are charged.... I'm thinking? Weird.
My mistake. I misread.
Is the 12v 'floating' (it's not) supply 12v as soon as the amp is powered up? Is it very near 12v?
I don't see 150uF caps on the diagram. What is their corresponding designation on the diagram?
Is the 12v 'floating' (it's not) supply 12v as soon as the amp is powered up? Is it very near 12v?
I don't see 150uF caps on the diagram. What is their corresponding designation on the diagram?
That's extremely high for a bootstrap capacitor. Did you try going lower?
uCD is popular, have you looked for other diagrams to see what they use?
uCD is popular, have you looked for other diagrams to see what they use?
I've looked a bit at uCD schematics, yeah seems 100u is pretty standard.
Just tried 100u x4, amp is still behaving the same.
Just tried 100u x4, amp is still behaving the same.
Does the signal level that you drive into it affect whether it starts or not?
Where is the clock signal connected to the diagram?
Is it possible that there is one channel causing the problem? Did you try removing the capacitor only one channel at a time to see if it behaves differently for one channel?
Where is the clock signal connected to the diagram?
Is it possible that there is one channel causing the problem? Did you try removing the capacitor only one channel at a time to see if it behaves differently for one channel?