Hey guys,
So after thinking about it for years I have built my first ever amplifier, well half of it for the time being. It's a Mugen Hybrid (https://www.hobbielektronika.hu/forum/getfile.php?id=19235), it's very well documented. After powering it on, adjusting and testing the necessary voltages as the article said I had connected some cheapo speaker to test it and it worked, I could not had been happier. If I remember correctly initially there was no distortion or I just could not hear it, but eventually it became very significant, now whenever I power it on it's there. It becomes better as I raise the signal strength, but it's still there. Also, to my surprise the transistors attached to the heat sink remain very cold to the touch though I was not running the amp extensively.
My question would be, what should I be looking for? I have used the best quality parts I could find, but there are still some cheap parts which I could not source for the moment.
Here is a video to demonstrate the problem:
Mugen Hybrid
Here are the schematics of the amp:
So after thinking about it for years I have built my first ever amplifier, well half of it for the time being. It's a Mugen Hybrid (https://www.hobbielektronika.hu/forum/getfile.php?id=19235), it's very well documented. After powering it on, adjusting and testing the necessary voltages as the article said I had connected some cheapo speaker to test it and it worked, I could not had been happier. If I remember correctly initially there was no distortion or I just could not hear it, but eventually it became very significant, now whenever I power it on it's there. It becomes better as I raise the signal strength, but it's still there. Also, to my surprise the transistors attached to the heat sink remain very cold to the touch though I was not running the amp extensively.
My question would be, what should I be looking for? I have used the best quality parts I could find, but there are still some cheap parts which I could not source for the moment.
Here is a video to demonstrate the problem:
Mugen Hybrid
Here are the schematics of the amp:
The problem is that your Vbe multiplier is overcompensating, lowering the bias into full class B, the Issue is that it will need to be set in a way it mantain a minimal bias for the amplifier to stay in AB, if it lowers to class B again then it may be needed to change its fisical position to it doesnt have the same thermal feedback (more heat less VBE that fixes the quiesent current/bias) it will need lots of experimentation it will also cause havoc with the bias stability when heavy loading the amplifier if the VBE multiplier isnt doing its job correctly.
Member
Joined 2009
Paid Member
This amplifier looks like a very nice design, definitely worth investing in.
I agree that it may very well be a thermal control issue affecting proper bias. The simplest option maybe to set your bias only after the amplifier has warmed up (e.g. 20 minutes or more). Set the bias more generously at first so that it can at least reach a higher operating temperature and then fine tune it until it's correct. Yes, it will draw 'too much' current when cold but this will only help it to warm up faster. This approach would likely be the approach the original designer would have used.
Another trick to allow you to tune the responsiveness of the thermal (bias) control is to put a resistor in the emitter of Q1. This is how it's done in the Naim
TGM10 - based on NAIM by Julian Vereker
I agree that it may very well be a thermal control issue affecting proper bias. The simplest option maybe to set your bias only after the amplifier has warmed up (e.g. 20 minutes or more). Set the bias more generously at first so that it can at least reach a higher operating temperature and then fine tune it until it's correct. Yes, it will draw 'too much' current when cold but this will only help it to warm up faster. This approach would likely be the approach the original designer would have used.
Another trick to allow you to tune the responsiveness of the thermal (bias) control is to put a resistor in the emitter of Q1. This is how it's done in the Naim
TGM10 - based on NAIM by Julian Vereker
Thank you for your replies! I let the amp warm up for around 45 minutes, then started from a higher DC offset and bias, started lowering the bias and now it has improved a lot! Still far from perfect, but I will keep fine tuning it as you have said it.
Thank you again!