mount some fans under the sinks that blows air up. it's not the heat inside the chassis that is critical. it's the sink temp.
Backed off my bias to 225mV levels and now the amp is stable with it's bias and temperature. Also did some listening tests. The V3 is terrific. Absolutely quite and no noise or hum. Sounds wonderful with very tight hold on music and bass. Little more warmer than original F5.
For my setup at the moment, I am Ok with what it is delivering. I may add a fan little later to drive up the bias. For the moment, I will enjoy the amp.
Thanks for all the help.
Cheers.
For my setup at the moment, I am Ok with what it is delivering. I may add a fan little later to drive up the bias. For the moment, I will enjoy the amp.
Thanks for all the help.
Cheers.
you MAY want to try 1.5ohm source resistors for a total of 0.75ohm with that low bias pr device. then you can take advantage og the diodes. with 0.75ohm source resistance and 0.45A bias(drop of 225mV over 0.5ohm) pr device you get 337mV drop over source resistors.
or just use 2 output pairs. and 0.8-0.9A bias pr device.
or just use 2 output pairs. and 0.8-0.9A bias pr device.
Last edited:
Audiosan
Thanks! I get a little condused here. In the original F5, each device was biased at ~1.3 A and the rail voltage was set at 23V. NP mentioned that with these parameters, the dissipation per device was kept at ~30 watts, about 50% of its max rating.
Going to the V1, the rail voltage goes upto 32V, NP dropped the bias to ~ 1A, in order to operate within the 30 watts per devices target. If for the V3, with a rail voltage getting up to 45 V, the bias should decrease further, right (or am I missing something here)? Is the 2.5 A the current bias set for each output device?
I am not trying to be nitpicking here. I understand that one can change the rail voltage and/or bias around as see fit. I just would like get a better understanding of what are the "optimum" parameters for this V3 amp. This will be the starting point of any changes/experimentation.
I have sourced a bunch of big caps for the PSU but these are rated at 50V only (I assumes the rail voltage for the V3 is teh same as V2). With a rail voltage at 45V, it is getting a little too close for comfort.
Regards,
the 1A bias is pr device. but that is a litle maximum with 0.5ohm source resistance do to the diodes starts conducting around that level. maybe even before. and then you got a runaway problem.
the 2.5A bias is total for all output pairs.
for the v3 you must have minimum 45V rails to get to the 100W mark. with 32V rails you only get 50W at 8ohm. even with 10A bias.
the 2.5A bias is total for all output pairs.
for the v3 you must have minimum 45V rails to get to the 100W mark. with 32V rails you only get 50W at 8ohm. even with 10A bias.
Folks:
While I understand that it would be better to cool the heatsinks themselves instead of focusing on managing the temperature inside of the chassis, I trust that forced air convection inside of the chassis would still benefit an F5T V3. I've been thinking about mounting an 80mm super quiet pc fan to the rear panel of one of the HiFi2000 5U enclosures that diyaudio's store is selling. Is there a reason why this would not be a wise choice?
Regards,
Scott
While I understand that it would be better to cool the heatsinks themselves instead of focusing on managing the temperature inside of the chassis, I trust that forced air convection inside of the chassis would still benefit an F5T V3. I've been thinking about mounting an 80mm super quiet pc fan to the rear panel of one of the HiFi2000 5U enclosures that diyaudio's store is selling. Is there a reason why this would not be a wise choice?
Regards,
Scott
it can not hurt
it easier on transformer and caps. but it will not help the fets much.
Can cause an airflow throughout the interior, the rear one meter the air out, a fan lid on the input (on the inside). Or the base on the outside (not seen), if the base with holes (air intake).
For example the fans of SilenX (IXP-64-14 with 92mm) have 14db and 42 CFM, practically no hear.
Use on my PC and dropped his temps + - 20 degrees.
in a PC, all that creates heat is INSIDE the chassis. so thats a whole other ball game. if you had a heatsink outside the PC chassis with all hot part mounted, you would not get much help from a fan in the chassis.
Correct, but I have done experiments with fan and without fan dento F5 and temps are going fall, that's a fact.
Of course the ideal would be cool sinks, but it is complicated, so we can try to cool the heat that causes this, the mosfets, so an air stream always gives a helping hand.
I am thinking of putting aluminum feet fixed in heatsinks, then F5T I'm building will land on a 10mm aluminum plate that will be cooled by two or three fans, almost as gadjets to cool portable pcs...
well. no i have all the parts in house. exept 2 bloody resistors i forgot the 47.5Kohm input resistors. o well. i found some 47Kohm in my bind not dale, but heck but have not heard anything from variac regarding testing of the boards. and i think i'm getting the sinks cut tomorrow, so i'll just have to use the prototype boards that i have. regardles of the tiny foults. to night there will be some soldering.
one question for buzzforb. have you tried different cascode voltages?
i'm wondering if i chould go for 16.4V or 20.6V.
i forgot the 47.5Kohm input resistors. o well. i found some 47Kohm in my bind not dale, but heck but have not heard anything from variac regarding testing of the boards. and i think i'm getting the sinks cut tomorrow, so i'll just have to use the prototype boards that i have. regardles of the tiny foults. to night there will be some soldering. one question for buzzforb. have you tried different cascode voltages?
i'm wondering if i chould go for 16.4V or 20.6V.
