Thanks 6L6, I thought I tried that, but now that I tried again, that seems to be the issue. Duh, another rookie mistake.
Many thanks. You've been a big part of getting me this far. Chassis should arrive next week and soon I hope to have a pair of V3 monoblocks to enjoy!
Now time to apply some juice. Fingers crossed.
-Jeff
Many thanks. You've been a big part of getting me this far. Chassis should arrive next week and soon I hope to have a pair of V3 monoblocks to enjoy!
Now time to apply some juice. Fingers crossed.
-Jeff
Hi all!
I just collected the parts to start building the boards however I have some doubts regarding the efficiency of thermistors when it comes to thermal tracking.
Suppose I set and stabilize the bias not that high, and during class AB operation the amp gets hotter than idle. Will the thermistor solution be able to keep bias, or could it runaway then?
Thank you
Daniel
I just collected the parts to start building the boards however I have some doubts regarding the efficiency of thermistors when it comes to thermal tracking.
Suppose I set and stabilize the bias not that high, and during class AB operation the amp gets hotter than idle. Will the thermistor solution be able to keep bias, or could it runaway then?
Thank you
Daniel
Help needed
Hi,
a friend of mine is building the F5T V.2. Everything was running fine until he fired up the whole amp without the light bulb. After 2 seconds ALL P-channel fets explode.
To me this looks like a systematical failure, he've checked all the connections but still have no clue.
Would a wrong + and - to the amp boards cause such a failure damage? For me this is the only logical explanation...
Any hint?
Thanks,
Mallard
Hi,
a friend of mine is building the F5T V.2. Everything was running fine until he fired up the whole amp without the light bulb. After 2 seconds ALL P-channel fets explode.
To me this looks like a systematical failure, he've checked all the connections but still have no clue.
Would a wrong + and - to the amp boards cause such a failure damage? For me this is the only logical explanation...
Any hint?
Thanks,
Mallard
Assuming that was the case, what would you say how long does it take to get them explode?
it can easely happen in a few sec.
measure resistans across R5 to confirm if this is what happened.
Both Pots P1 and P2 were set to max resistance instead of min resistance.
Ok, easy root cause
just need to buy a new set matched fets.
Ok, easy root cause
just need to buy a new set matched fets.
there is nothing wrong with little reading exercise :
http://www.firstwatt.com/pdf/prod_f5_man.pdf
http://www.firstwatt.com/pdf/art_f5_turbo.pdf
try other pdfs too , you'll see - they're pretty informative
http://www.firstwatt.com/pdf/prod_f5_man.pdf
http://www.firstwatt.com/pdf/art_f5_turbo.pdf
try other pdfs too , you'll see - they're pretty informative
Hi,
Looking for feedback on an attempt at a super-compact PCB for a hybrid amp based off a mix of concepts from the F5 and the F5T. Adapted to work with my specific speakers and a specific cooling solution. Just 3"x4", these are designed for use with a copper heatspreader. All the footprints are unusual (Caddock and Holcos for the build) and the dimensions extremely compact (reason why below) so I could not go with any currently available PCB. The design is optimised for being etched by hand - so no vias or complex routing. I have never really developed PCBs before so some feedback would be welcome.
The circuit is very similar to what was being sold on DIYA store as the V2.0 of the original F5, and P3 and the new thermistor arrangement from the Turbo. Which is to say dual pair output, cascoded input, current limiting and P3. I guess you could also see it as a F5T V 1.0 with added current limiting and cascodes. I am debating the routing of the thermistors. Not yet sure whether to route them to the source resistor of one output pair, or to the supply. Or have two thermistors, one for each pair. I am not really sure what is happening to the unmonitored pair/s when the thermistor begins to operate.
All grounds are not connected together on the board - cascode and signal ground are deliberately left unconnected, there will an off-board star ground. We have two amplifiers to a case and two cases. This allows for a biamped system or balanced/bridged drive with a few small modifications to the board and connections.
You can see the cases here, these were finished almost a year and a half ago but I could not focus on this project due to a bad back, and that doesn't play well with heavy amplifiers. Just ask Mr. Pass, he knows that one well http://www.diyaudio.com/forums/pass-labs/207103-f5-turbo-builders-thread-49.html#post3208407 The front is 14" wide but due to the thick frame I get only about 10" of clear space across the sinks' width. And the contact patches are really narrow, just about 4" each, which is one reason for the very small boards.
Going to be creating one prototype board with transistor sockets for the input JFETs and TO-220 IRF540/9540 for the outputs (cost abut $1/pr). That way I can destroy a bunch of stuff without having to worry about financial consequences or worse, re-soldering boards.
Thanks for reading
Looking for feedback on an attempt at a super-compact PCB for a hybrid amp based off a mix of concepts from the F5 and the F5T. Adapted to work with my specific speakers and a specific cooling solution. Just 3"x4", these are designed for use with a copper heatspreader. All the footprints are unusual (Caddock and Holcos for the build) and the dimensions extremely compact (reason why below) so I could not go with any currently available PCB. The design is optimised for being etched by hand - so no vias or complex routing. I have never really developed PCBs before so some feedback would be welcome.
The circuit is very similar to what was being sold on DIYA store as the V2.0 of the original F5, and P3 and the new thermistor arrangement from the Turbo. Which is to say dual pair output, cascoded input, current limiting and P3. I guess you could also see it as a F5T V 1.0 with added current limiting and cascodes. I am debating the routing of the thermistors. Not yet sure whether to route them to the source resistor of one output pair, or to the supply. Or have two thermistors, one for each pair. I am not really sure what is happening to the unmonitored pair/s when the thermistor begins to operate.
All grounds are not connected together on the board - cascode and signal ground are deliberately left unconnected, there will an off-board star ground. We have two amplifiers to a case and two cases. This allows for a biamped system or balanced/bridged drive with a few small modifications to the board and connections.
You can see the cases here, these were finished almost a year and a half ago but I could not focus on this project due to a bad back, and that doesn't play well with heavy amplifiers. Just ask Mr. Pass, he knows that one well
http://www.diyaudio.com/forums/pass-labs/207103-f5-turbo-builders-thread-49.html#post3208407 The front is 14" wide but due to the thick frame I get only about 10" of clear space across the sinks' width. And the contact patches are really narrow, just about 4" each, which is one reason for the very small boards.Going to be creating one prototype board with transistor sockets for the input JFETs and TO-220 IRF540/9540 for the outputs (cost abut $1/pr). That way I can destroy a bunch of stuff without having to worry about financial consequences or worse, re-soldering boards.
Thanks for reading
I am spreading them out over the sink I have to fit two boards/eight devices, edgewise, in a 10" space. That means the boards can be no more than 5", and that is pushing it because the devices on the outer edges will be on the edge of the conduction surface.
There will be a 12mm copper heatspreader on which the devices will mount, so though heat concerns are not totally mitigated I don't foresee too much of an issue. I am not targetting the bias level of a full fledged F5 Turbo, about 20W per device should be more than adequate for my needs.
I have to fit two boards/eight devices, edgewise, in a 10" space. That means the boards can be no more than 5", and that is pushing it because the devices on the outer edges will be on the edge of the conduction surface.There will be a 12mm copper heatspreader on which the devices will mount, so though heat concerns are not totally mitigated I don't foresee too much of an issue. I am not targetting the bias level of a full fledged F5 Turbo, about 20W per device should be more than adequate for my needs.