Burning Amp BA-3

It seems like some of you are also having this thermal drift problem. I had already replaced the single turn pots to multi-turns type and setting the dc offset is very much easier now except that it will drift off due to thermal change. Even after a warm up of more than 1 hour, it will drift off after putting back the top cover. Looks like the only way is to have bigger heatsinks for the FETS.

Check the heat of your sinks and see how hot they are getting. Try it after you have the lid on in the situation where it drifts. It could be heat related. What is the Idss of your input fets? what current are you showing on the Jfets when the amp is on.
 
Member
Joined 2004
Paid Member
Let's hear from more builders of the BA3. See what they find with output offset stability.

It's just possible that a DC servo is required to control the offset, but at this early stage, further investigation is required to find out why offset is drifting so much.

The DC offset drift in my follower stage is minimal, less than +-.005mV.

The FE drifts all over the place that's why it's capacitivly coupled to the follower. Any DC on the input would exacerbate this drift.
 
The DC offset drift in my follower stage is minimal, less than +-.005mV.

The FE drifts all over the place that's why it's capacitivly coupled to the follower. Any DC on the input would exacerbate this drift.


If I don't understand wrongly, the FE's dc offset should be tuned without hooking up with the preamp and source. The main problem is that when I hooked it with the preamp, the FE's dc offsets will drift away by as much as 160mV on the right channel and about 40mV on the left channel (immediately). That really puzzled me. The dc on the input (DCB1 for my case) is rock solid at 3.5mV on the right and 0.1mV on the left.
 
By adding just a little bit of DC to the FE of the BA3, you are basically affecting its bias. Normally the gate would be pulled to ground and the bias would be fully open if not for Rs, which create a positive voltage and dynamically reduce the current through the fet. with even a small DC at the gate, you change the bias point, and as a result everything that depends on that throughout the rest of the chain. Hopefully someone will confirm or shoot down this explanation and make it clearer.:headshot:
 
By adding just a little bit of DC to the FE of the BA3, you are basically affecting its bias. Normally the gate would be pulled to ground and the bias would be fully open if not for Rs, which create a positive voltage and dynamically reduce the current through the fet. with even a small DC at the gate, you change the bias point, and as a result everything that depends on that throughout the rest of the chain. Hopefully someone will confirm or shoot down this explanation and make it clearer.:headshot:


I am no expert in electronics and hope you won't mind my stupid question here. Will a dc component be amplified also if it is present together in a signal? If it does, then it is more logical to explain that the dc component is amplified to 170mV at the FE stage from DCB1's dc offset of 3.5mV although the amplification factor is about 48 times.

The Idss for the Jfets is 9.1mA. Currents on the Jfets are 7.79mA and 7.74mA measured at R6 and R7 respectively on the right channel with the top cover fixed up. Without the top cover, its 7.62mA and 7.57mA.
I had measured them with the power amp hooked up and unhooked up with the DCB1 with both situations showing almost the same readings, only a slight difference of 0.01mA

I can't measure the temperature as I do not have a thermometer with me now. I guess its about 60 ~ 65 degree Celcius or 140 ~ 149 degree Fahrenheit.
 
Hi,

The BA3 Front End can be connected either with BA-1 or BA-2 output stages.

The BA3 Front End with BA-1 output stage (+ / -32 volt rails) : output power limited to 50W into 8 ohms, 25W into 4 ohms, 12 W on 2 ohms.

I think that the balanced BBA3 Front End (+ / -32 volt rails) must be connected with the official BA # 2 push-pull output stage.pour get close to 200W into 8 ohms.

I think that the balanced BBA3 with official output stage BA-1 (+ / -32 volt rails) will always be the power of 50W into 8 ohms, 25W into 4 ohms, 12 W on 2ohms.

I'm right?

Thank you for your answer

Regards

Darry
 
Small theory question. Using Ba3 FE as an example, what affect does the bias current of the second stage have on the drive capability and performance of the circuit. It is well known that more bias increases linearity, but what affect does it have on driving different impedances. Are there any side effects, other than heat, to increasing the bias.