Upgrading Vintage Marantz 1030 Power Supply to Zener Transistor

[Alan0354]

In this situation, the zener works. This is because the current drawn is low ( less than 20mA). Let's just hope the 42V will never drop below 37V and all will be ok.

But case in point, if you require high current to...say power the power amp output stage, it is not as easy. I actually layout a board like this but with voltage divider like the original circuit using resistor. If I don't leave enough headroom, the raw voltage dip below the regulated voltage and cause problem. BUT BUT, if you give more headroom for the raw voltage to dip, the pass transistor gets hot really really fast. I could not find a reasonable middle ground. I already use a 12W heatsink for the transistor already, it heated up like in 1minutes!!!! I am not going to use a big heatsink just to do regulated supply!!! I think that's the reason all the talks about filter caps and ripple on the rails, people still use good old rectifier/cap and put up with the ripple. Finally, I just go back to rectifier and tons of caps!!!!

This idea sounds like a good idea until you try to implement it for high power application. I am going to study SMPS design when I have time, that will solve the problem with a SMPS.

 

[benhodgson]

In this situation, the zener works. This is because the current drawn is low ( less than 20mA). Let's just hope the 42V will never drop below 37V and all will be ok.

But case in point, if you require high current to...say power the power amp output stage, it is not as easy. I actually layout a board like this but with voltage divider like the original circuit using resistor. If I don't leave enough headroom, the raw voltage dip below the regulated voltage and cause problem. BUT BUT, if you give more headroom for the raw voltage to dip, the pass transistor gets hot really really fast. I could not find a reasonable middle ground. I already use a 12W heatsink for the transistor already, it heated up like in 1minutes!!!! I am not going to use a big heatsink just to do regulated supply!!! I think that's the reason all the talks about filter caps and ripple on the rails, people still use good old rectifier/cap and put up with the ripple. Finally, I just go back to rectifier and tons of caps!!!!

This idea sounds like a good idea until you try to implement it for high power application. I am going to study SMPS design when I have time, that will solve the problem with a SMPS.

Something else occurred to me today thinking about this. When the Marantz was designed it was for 240V (and the transformer has remain wired as such). UK Mains is now 230V and actually if the transformer ration remains the same then the 42V input is 42 * (230/240) = 40.25V ... even less headroom!

Is it worth changing the spade connectors to the 220V setting? Then the 42V rail is 42 * (230/220) = 43.9V. Although this may give too much voltage elsewhere in the amp.

I see your point Alan0354... I've obviously been pretty lucky on tolerances and the components I've put in!

 

[Alan0354]

No, don't change the setting!!! Really bad things will happen!!!! Even 40.25V is ok.

Actually, I would not even modify the circuit. The output voltage of the original circuit is not going to be as constant as the zener, it will drift a little depending on the 40V raw voltage, but it will not affect the sound quality as the drift is in order of less than a few Hz. But the original circuit will not run into problem if the 40V dips too much as the 35V will drop following the 40V. With the zener, the 35V will not droop with the 40V, BUT if the 40V dips below 34.4V, the transistor base-collector diode will be forward bias and large current will flow and might hurt the transistor.

Bottom line, I help you in the calculation, BUT I would not have modify the circuit if this is my amp.

 

[benhodgson]

No, don't change the setting!!! Really bad things will happen!!!! Even 40.25V is ok.

Actually, I would not even modify the circuit. The output voltage of the original circuit is not going to be as constant as the zener, it will drift a little depending on the 40V raw voltage, but it will not affect the sound quality as the drift is in order of less than a few Hz. But the original circuit will not run into problem if the 40V dips too much as the 35V will drop following the 40V. With the zener, the 35V will not droop with the 40V, BUT if the 40V dips below 34.4V, the transistor base-collector diode will be forward bias and large current will flow and might hurt the transistor.

Bottom line, I help you in the calculation, BUT I would not have modify the circuit if this is my amp.

Thanks, I understand.