Q303 Revision

  • No one logged in.

10-Jun-2017 10:55 AM

QuadOAP

QuadOAP

Posts: 9

I have just completed the 303 V1.9 revision and experiencing a problem in the initial set-up. The 67 volts slowly increases.

I allowed the amp to settle for a couple of hours (without signal) and set it to 67v. Left it overnight and checked it on switch-on. 64.9v. After 1 min = 65.5. 2min = 66v. 3min= 66.2v. 4min = 66.3v. 5mins = 66.5v. It stayed at 66.7 and 66.8 for several mins. It eventually reached 67v at 40 mins after switch on.

The voltage across the 12v zener on regulator board started at 11.6 but quickly went to 11.7 where it remained constant. The set-up instructions says set to exactly 67v.

Can you tell me what are the tolerances for the 67v and voltage across the zener. I want to get this right before proceeding.

  Reply

10-Jun-2017 05:43 PM

1p21gigawatt

1p21gigawatt

Posts: 21

Hey there,
The slow increase you describe here is completely normal. I'm not sure about the tolerance of the supply voltage, but one volt more or less should not have any negative effect on the amplifier performance. Just make sure that the "middle voltage" of both boards (Pin 5) is half the supply voltage.
Best regards, Marc
  Reply

13-Jun-2017 07:46 PM

QuadOAP

QuadOAP

Posts: 9


Thanks Marc,

Ran the amp (without signal) for a couple of hours then set 68v, 33.5v and quiescents. Ran again for a few hours with signal and made some fine adjustments. Fitted cabinet and ran on tuner for a few hours, all seamed Ok then I noted a smell of overheating. It was running at a Volume of about 3 +, but the cooling fines were very hot indeed. Too hot to touch. Don't think this is related to the original post.

Will investigate later , but thanks for your input

  Reply

14-Jun-2017 11:31 AM

EJP

EJP

Posts: 1412

It's overbiased. Whatever you have set it to, it's too much.

Over more than ten years, I have found the correct bias point for any given 303 to be extremely variable, anything between 3mV and 12mV across pins 4 and 6. The later ones with the bias spreader transistor tend to need more than the early ones with the four diodes instead.

I set the bias with a distortion analyzer, which most readers won't have. In practice you should probably set it to the highest value that doesn't run away thermally when observed for several minutes. This is quite difficult to achieve, as you're measuring a positive thermal feedback loop with a long time constant, so just find the point where it has neither increased nor decreased on its on initiative over at least two minutes.

EJP
  Reply

17-Jun-2017 11:44 PM

QuadOAP

QuadOAP

Posts: 9

To recap: I had both channels set to near 6.5mv. and no overheating occurred. I had tried higher settings but the measured voltage fluctuated, so I set them to where I felt they were more stable. The overheating occurred only when I applied a signal.

I reduced both channels to approx. 5mv - still overheating when a signal is present. Reduced further to near 4mv - same result. Same again, when I tried 3mv, although the left channel fluctuated a lot more than the right at this voltage 

I have only recently inherited this amp and played it for a few hours before starting the revision. No signs of overheating at that point. It did linger on for a while after switch off, but I hoped this would be rectified during the revision.

Not sure where to go from here?

Can overheating occur when the bias is low as well as high.

Should I run on a single channel to see if both channels are affected or which one is overheating. 

I have ordered a USB Scope, so I should be able to see more into the circuit shortly. 

  Reply

18-Jun-2017 04:52 AM

EJP

EJP

Posts: 1412

Running on one channel is a good idea. 6.5mV should be OK.
  Reply