Denon PMA 737 schematic needed

Denon PMA-737

I hope you have found it, ivanlukic! Otherwise, let me know! :)

Taking advantage of this open thread, I have a question regarding this model.

In the schematics we can see only one opamp, the M5218. Here below, highlighted.

An externally hosted image should be here but it was not working when we last tested it.


But in the printed board, we see two! Unless we have one opamp to each channel... :scratch2:

An externally hosted image should be here but it was not working when we last tested it.


If so, are they in charge of tone control?? ;)
 
Administrator
Joined 2007
Paid Member
Looks like they are just using one opamp in the package. No idea why they didn't use a single opamp... cost probably if they had these available. Or they were deemed most suitable but the board layout wouldn't accommodate using just one.
 
Looks like they are just using one opamp in the package. No idea why they didn't use a single opamp... cost probably if they had these available. Or they were deemed most suitable but the board layout wouldn't accommodate using just one.

Looking at the component numbering it looks like one per channel.

Looking still in the first picture, the lower highlighted part could refer to pins 5, 6 and 7 of that same opamp, but it still has "IC2 M5218P" marked and it's inside a blank box-diagram... Strange...

In the PMA-757 model there are 4 of these Mitsubishi "dual low noise" opamps. A "pair" (IC301, IC303) shown here. The other "pair" is IC302, IC304 and is not shown in the schematics... :scratch1:

An externally hosted image should be here but it was not working when we last tested it.
 
Bias: temperature and stability

I'm sure this have already been discussed many times here, but in this specific case...

Photos below show the right pair of 2SA1011/2SC2344 drivers in this model. I know they were designed like this (without heat sinks for dissipation)...

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.


But due to extreme sensibility even to air flow variations, setting the bias has been a funny thing: voltage goes from the 16mV down to 10mV, or up to 22mV within some 30 minutes. And the drivers are pretty "hot" even with no signal applied.

General question: adding a heat sink to these transistors would improve on stability?
 
Last edited:
Hard to say really. The real problem could be that the vbe multiplier doesn't track well or that it is influenced by heat from the other channel.

:) The amp's heat sink has plenty of space to accommodate the 4 transistors/drivers. I'll put it aside for now. And see if I can do something later...

... I just recapped the PMA-757 with some 25 Epcos film caps instead of most 1uF electrolytics spread in this amp.

34oyq75.jpg


IC301 and IC302 (one opamp per package) I left as they are.

16k7051.jpg


IC303 and IC304 use both opamps in the package. I had several pairs of opamps to test in this amp. This time none of them was a problem. Easy bias setting and pretty stable after some 30 mins.

zsuwaw.jpg


I liked here the OPA2134. It seems to have a huge "3D" sound. :)
 
Administrator
Joined 2007
Paid Member
I've always liked the OPA2604, in fact its still one of my favourites.

Bias stability in commercial amps isn't always perfect, often you can't really do much about it either.... but its a non problem really from an audible point of view. Manufacturers like to play safe and they take into account typical customer abuse such as covering the amp up and so on.
 
Denon PMA-737 drivers

I'm sure this have already been discussed many times here, but in this specific case...

Photos below show the right pair of 2SA1011/2SC2344 drivers in this model. I know they were designed like this (without heat sinks for dissipation)...

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.


But due to extreme sensibility even to air flow variations, setting the bias has been a funny thing: voltage goes from the 16mV down to 10mV, or up to 22mV within some 30 minutes. And the drivers are pretty "hot" even with no signal applied.

General question: adding a heat sink to these transistors would improve on stability?

(...)

The amp's heat sink has plenty of space to accommodate the 4 transistors/drivers. I'll put it aside for now. And see if I can do something later...

Upgrade.

I took the heatsink off and made place for the drivers, etc.

An externally hosted image should be here but it was not working when we last tested it.


But measuring one of them, the 2SA1011, it shows very low voltage drop, something like .050, instead of the usual 450-550...

I have 2 pairs of 2SA1837/2SC4793 and would like to know if they can be used for the replacement.

The 2SA1011 is -180 Vcbo, whereas the 2SA1837 is -230Vcbo. So good.

But... the A1011 is -6V Vebo and the A1837 is -5V Vebo. Big deal?

Ic of 1.5A for the A1011 and 1.0A for the A1837...

Pc of 25W for the 2SA1011, and 20W 2SA1837...

The 2SA1837/2SC4793 are case isolated and are not Ft simmetrical. One is 70MHz (2SA1837) and 100MHz for the 2SC4793.

Possible?

..
 
Last edited:
Administrator
Joined 2007
Paid Member
There shouldn't be a problem using your alternative devices. The reverse base bias breakdown voltage isn't a problem, its a region the devices never enters in an amp like this... its always forward biased.

The one that measures 0.050. Are you saying its failed... amp not working, or just that it reads strangely ?
 
...

The one that measures 0.050. Are you saying its failed... amp not working, or just that it reads strangely ?
I believe it's failed. It was VERY hot. Hotter than the others. Amp was working but bias drifting all the time...

One more question, Mooly: If I use the available trannies I have, they are case isolated. They don't need to be attached to the heatsink, right? Even if they dissipate less (on datasheet) than originals, these transistors do not need heat sink?
 
Administrator
Joined 2007
Paid Member
If the amp is working then I'd be surprised if its failed. There is only transistor that you can get away with failing (going leaky/short and that is the vbe multiplier.

Case isolated... these are less good at dissipating heat when free standing compared to a metal tabbed device (and less good on a heatsink as well). If the devices are running close to their limits for being non heatsinked then it could be a future reliability issue. Around 1 to 2 watt absolute maximum is about the limit for free standing at an ambient of 25C.