Here's a PDF of the schematic -- I blew out one channel so had to trace it out. The transistors used inthe original are 2SK405 and 2SJ115 for the output stage, and 2SA1360 2SC3423 for the va stage.
the one channel which is working sounds spectacular -- I have owned this amp since the 1980's but it was used by my son with a set of Swan's and I think that he overdrove it.
the THD at 1kHz around 60W was 0.02% if that matters.
the THD at 1kHz around 60W was 0.02% if that matters.
replaced the original MOSFETs with IRF'S -- initially the distortion was quite high, 0.50% but by adjusting VR1 and VR2 both channels match at 0.01% to 0.02%.
jackinnj said:
If you are replacing the original parts types, you might try using a better chip for U2 and a better cap for C9.
A 741 is hardly a low-offset IC, and something like an LF411 might be a better option. There are other types, but that might be easier to find and cheap too.
DC servos are used to eliminate the feedback large cap on the gain. But that doesn't mean that no audio is passing through any caps, as it is going through the servo cap C9. This cap should be a high quality one, preferably a film type. As that value is a bit high for such a type and you might have some trouble finding pcb space to hold it, a better eletrolytic should do it.
Carlos
Carlos
Yes, I just wanted to get the amp up and running first -- the opamp's bypass caps are far removed from the opamp itself -- meaning that they are useless -- I can get a non-polar electrolytic and byass it with 100nF MKT for C9 --
anything more is gilding the lilly (but I am tempted) -- it sounds good, the distortion at low levels is 0.03% on both channels, and at high levels a little over 0.01%
Yes, I just wanted to get the amp up and running first -- the opamp's bypass caps are far removed from the opamp itself -- meaning that they are useless -- I can get a non-polar electrolytic and byass it with 100nF MKT for C9 --
anything more is gilding the lilly (but I am tempted) -- it sounds good, the distortion at low levels is 0.03% on both channels, and at high levels a little over 0.01%
jackinnj said:
Getting the amp running is certainly the most important thing.
I didn't refer to the opamp's bypass caps, but C9, which is in series with the chip's output. Changing the chip might bring the greatest gain, but do that using a chip socket, if it's not already provided for.
For the cap it might be better trying several types and values.
These two mods I would not consider gilding the lilly, as they are part of the feedback. If you go for it, change the chip first and then try different cap values and types.
Or write to Buongiorno, who designed the amp and might have something to say on potential improvements. That might be the wise thing to do.
Carlos
Hi
Something is wrong in the schematic , with C9 in series with the output of the Op Amp , the op amp can't do anything to control the offset...
Something is wrong in the schematic , with C9 in series with the output of the Op Amp , the op amp can't do anything to control the offset...
I didn't really bother to check the value of the compensation caps in the feedback network -- I'd have to unsolder the part to get to it.
with respect to C9 -- I will take it apart again and retrace
with respect to C9 -- I will take it apart again and retrace
djk said:
You are right: this was not a James Bongiorno design. But he used FETs on the Andromeda.
Carlos
The original Andromeda was BJT, the later one (not a Bongiorno design) was FET.
The BJT design sounds better and is more reliable (former Sumo dealer and service center).
Not that it matters, but the source and drain connections on the schematic for your Polaris are drawn wrong, as well as the DC servo.
The BJT design sounds better and is more reliable (former Sumo dealer and service center).
Not that it matters, but the source and drain connections on the schematic for your Polaris are drawn wrong, as well as the DC servo.
CR1 and CR8 are drawn wrong.
If I remember correctly they should go to the supply rails (Baker Clamp).
If I remember correctly they should go to the supply rails (Baker Clamp).
The anode of CR8 looks like it goes to the base of Q16, not ground as shown in the diagram. Can't see where the cathode goes.
Is CR1( CR8 ) 1N4148 ? Because Vr of 1N4148 is only 75V (Vrm is 100V) and Polaris's power is +/- 60V.
looks like an IN4148 but there are no markings -- as opposed to the zeners which are clearly marked.
- Status
- Not open for further replies.