diyAB Amp The "Honey Badger" build thread

Hello
greetings anybody help me at 55 volt dc supply rails how do i do the biasing
of honey badger how many millivolts on base of output transistors and reading
over emitter resistances maybe someone can help me
warm regards
Andrew

you do not measure the base voltage, measure the emitter resistor voltage drop instead...as Mark Tillotson said...
 
Thanks

Well done. I have been following your progress through this.




Thanks Stuart.. I' haven't put it back in the chassis yet, probably tomorrow. I'll do that and then I need to pull off it and get a website up and running. After that, I'll come back and do some testing we were working on.


It occurred to me that I might get much different readings and Indeed still might not get to where you set yours for testing. We have different rail voltages and diff bias, so I don't think a comparison is even possible, or worthwhile.



I'm going to look at my numbers and they will be what they are. We have to many differences in our amps to do what we were thinking. Hell, I'm not even using the same devices you are....


I do thank you for the help in building the attenuation. ;)
 
Member
Joined 2016
Paid Member
I do thank you for the help in building the attenuation. ;)

No worries, happy to help out. If you can run a test with your amplifier module outside the case and then one in the case and compare. This will show you what influence your internal wiring, power supply loop area etc is having on your measurements.

What load resistor are you using again. Can you post a image?
 
Settled on a bias of 36mv, it was running about 30mV before. It's running a lot hotter. I ran it up to 100W or so and let it rip for 40mins and the highest temp I found was 46.6C.



Once I put it back in and check numbers everything looked good, so I watched it idle for 10mins or so then streamed Audioslave as loud as I could stand and she never burped. :)


Off to build a website, so I won't get back to doing anymore testing for a week or two.



@Stuart I'll get you some numbers after that. As I said though, I'm not going to worry to much about them... the amp sound good. :)
 
Settled on a bias of 36mv, it was running about 30mV before. It's running a lot hotter. I ran it up to 100W or so and let it rip for 40mins and the highest temp I found was 46.6C.

the objective of biasing is really the elimination of notch or zero crossing distortion, into 1 watt 8 ohm dummy load....

you need to bias your amp so that none of this is seen on the scope...

bias enough to get rid of this, more bias current and your heatsink gets warmer, not really required...

if you have a distortion meter it will be interesting to find how much bias lowered THD.....and by how much idle power increase to bring thd down...

An externally hosted image should be here but it was not working when we last tested it.


below is the correct sine wave looks....
Crsdist1.gif
 
Last edited: