Hi prasi!
As far as I know there has been one attempt to build a double MOSFET version, and it was successful; although it was not the design I uploaded in those posts.
I myself have been a bit too busy with other things to get the V2 boards printed. However, I am trying to print a couple dozens next month, and hope to get them singing as soon as I get them.
A cap-multiplier supply will be great for this amplifier.
And, welcome to the PeeCeeBee thread!
shaan
edit: !!!!!!! looks like wiljj78 already have answers to your questions !!!!!!!
Last edited:
zOMG
Great job man!A fresh double-output design!!
Power to you!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
shaan
zOMG
A fresh double-output design!!
Power to you!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
shaan
Thanks shaan... Currently im playing with bias.
The amp playing a bit distorted..I played with 1k pot istalled instead of 470R but no changes...
Taking off board 1x4148 and reducing down the VAS resistor down to 10K then works better BUT VAS transistor extremely HOT....
Any suggestion?
Regards...
Hi wiljj.
The layout is not fully visible in the pictures you previously posted. Can you upload one without the markers for the VAS heatsinks? Also, what is your supply voltage?
You can remove that 470R resistor near the 4148 diodes if you plan to run this amp in Class-AB mode. This resistor is used only if no diodes are used in the biasing section or a variable or high bias is needed.
The layout is not fully visible in the pictures you previously posted. Can you upload one without the markers for the VAS heatsinks? Also, what is your supply voltage?
You can remove that 470R resistor near the 4148 diodes if you plan to run this amp in Class-AB mode. This resistor is used only if no diodes are used in the biasing section or a variable or high bias is needed.
Hi Shaan,
Here is the clear picture with bottom view.
My rail is +/-38Vdc, having the default resistor 18K on VAS I'm getting 189mV across 10R resistor (VAS heat is acceptable) but sound is distorted.
I tried to install 20K parallel with 15K sound is perfect but VAS is too HOT.
I guess I need to keep with 18K and play around with bias to stabilized the output.
waiting your inputs..
Regards,
Attachments
No Shaan, and this is what im working right now...
Having the 18k itself onboard offset is about 56mV having paralleled with any value up to 2M offset remain unchanged.
When i replace 18K to 20K paralleled with 15K off set drops to 16mV.
Seems like 18K is enough for VAS, what i need is to increase the output bias.
This is very strange! With 18mA through those diodes and even 100mV offset you should not have distorted sound (unless there is a wiring fault).
Well, if it is distorting then it IS distorting. Can you please test the default MOSFET bias with 18K? And also the voltage around the 4148 diodes?
Well, if it is distorting then it IS distorting. Can you please test the default MOSFET bias with 18K? And also the voltage around the 4148 diodes?
I have pack up my things shaan and will continue tomorrow, its getting late now in my place..
I remember the voltage across single 4148 was .842V with default 18K.
I will do some test again tomorrow and will update the progress..
Thanks shaan.
Hi Wiljj,
I had the same issue with my PEECEEBEE TO-3 version by Jason.
First, 189mV is too high unless you have a good heatsink, try to increase the resistor to about 22-24K, aim for 12-15mA (120-150mV across the 10ohm), mine takes about 20 minutes to become stable (heat wise) so wait a bit so you choose the right value.
Once your VAS has the proper and lower current, I did the following (thanks to Fab on this forum): I removed both 1N4148 and replaced them with a single resistor, for me 130 ohm did it and provides a bias current of 155mA. If you remove the diodes, make sure to always have a resistor in place.
Another method is to add a third 1N4148 in series with the 2 other one. My NOS LatMOS 2SK134/j49 are a bit lazy so the voltage across 2 diodes wasn't sufficient for my needs.
Hope this helps,
BR,
Eric
Last edited:
my 4 channel PeeCeeBee t0-3 by Jason
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
That's some nice wiring there!
Thanks for sharing. Keep us posted!
And, Welcome to the PeeCeeBee thread. ^^
Last edited: