Yes i have seen the diodes So they have nothing to do with possible spikes coming from the power supply
Let me elaborate a little on the subject of power supply noise and ask questions
I was watching the lab testing of a 50 years old Sansui During the THD test with a 1kHz sine there were spikes at 60-120- etc.
Those spikes could not come from other place than the power supply and the PSRR of the amp was not high enough to suppress them
My question is ... could better and newer caps, some caps bypass or other kind of filter suppress those spikes ?
If the answer is yes then this motivates the intervention
I do not have the instruments to check this thing here But as i said before many manufacturers sell a cap upgrade for some of their models
I would prefer to place some low esr caps directly on the Voltage pins of the output devices to ground
They could provide "local" energy for them
We can discuss about uF quantity and cap quality of course
I really do not like all the big capacitance only placed at the entrance of the circuit
Again many companies offer caps upgrade Many There must be a reason
Those spikes come from the rectifying diodes most likely. To solve that, I made a Quasimodo tester, which experimentally determines the snubber circuit on each secondary individually, and that noise disappears. You need an oscilloscope and that Quasimodo tester. The device itself was developed here on the forum and is very useful.
Hi thanks a lot and yes that would be the easiest action But i am thinking to something more drastic ... like taking out the transformer entirely and put in a separate box No vibrations no EMI nothing Then enter the case with VAC-0-VAC a new robust diodes bridge and two big caps like 20-33mF each of the screwed typed Maybe with some additional bypassing
I will use some short wires to connect the caps to the board It would be the cleanest/safest way
I do not know why people hate external transformers I love them
CRC or CLC filtering is much more effective than putting in large electrolytes. With larger electrolytes, you have even more noise from the bridge rectifier. But that's not the main problem, those amplifiers generally sound like that according to those schematics. Maybe the problem is the input capacitors. Some know how to blur the sound a lot. Try to find a original Wima MKS4 100V of the same vaslue & dimensions as the blue ones on the PCB. There is room there, maybe some bigger ones can fit.
Last edited:
Because there are very strong current pulses from transformer to rectifier to electrolytics. The magnet field you create by using long connections is much worse than all of the effects of internal transformers.
Edit:
However, by including long(er) wires from transformer to rectifier+caps you also increase resistance and thus you reduce the current pulses and slightly increase ripple.
External transformers (plus rectifier and caps) are ok for low current devices.
Last edited:
thank you very much This is very important I thought the spikes were due to the aged caps
This is how things should be done
But some design principles make a lot of sense to me Like placing thicker wires where the currents are higher (i get this advice from Mr Ohm)
placing big and fast caps close to the output devices expecially in class AB designs (with class A the current draw is always the same i understand)
I am learning ... slowly
If there is no audible 50/60Hz hum, the transformer should not be moved. It is mostly forgotten that the secondary current in the case of a transformer+diode bridge+capacitor combination is not a sinusoidal but a series of strong pulses at the peaks of the secondary voltage.
Thanks ! i did not think of that but you are right of course I have to remove this board If the weather will allow next week i will do and take some pics
I've been shocked a couple of times by the difference in sound just by replacing those coupling capacitors. And sometimes I didn't hear any difference. If you had DC speaker protection, I'd tell you to bridge those input capacitors. You can try (for test) to bypass them with a smaller but high-quality capacitor, around 100nF. Nothing expensive, Wima FKP, or whatever.
even with the rectifier mounted very close to the electrolytics ? Like this solution for instance
i guess that you refer to the wires inside the case ? i could try to shield them using some aluminum foil
What about entering the case with DC instead ? and make a 1st rectifying stage externally with the transformer ?
Naim Cyrus do that with their premium outboard power supplies
Depending on the current draw, the rectifier may not be able to cool sufficiently.
Hugo
Long wire from transformer to cap (with rectifier somewhere between, that's not relevant) with it's current peaks creates magnet field, with mains frequency and strong uneven harmonics.
You cannot shield magnet field with aluminium.
But I guess you will have to try it to know it!
Theory, theory and nothing else.
Practical test::
Look at old well-praised japanes amps. Some of them didn´t even have 6A diodes, but still had 20/30/40000 µF capacitor banks.
If they didn´t blow over 40 years.... why should these??
That might be fine...... And it might NOT.
An old saying: "If it works, don´t fix it".
Audio brand engineers certainly have a budget limit. That doesn´t mean, that they don´t know, what they´re doing.
To think, that without proper education and the only valid arguement to be "I don´t like the looks of it", seems a bit far fetched to me.
As some other wrote (I think)..... rather send the amp to me, than dstroying it 😊 .
On the "separate transformer box", I would at least build in the rectifying circuit, and have the connecting wires run DC-0-DC instead.
The problem is to discharge the gate capacitances of all the mosfets. A pair of current drivers should be added.
Attachments
Hi and thanks again I think i have understood Is very important to keep AC power out of the case ? then i could place the bridge and bigger caps just very close to the rear panel and go from the big caps to the caps on board with DC That could work ?
I have indeed noticed that all the brands that offer upgraded external power supply like Naim and Mission to name two they provide DC to the amp
And people report of great benefits And those external power supplies are also very expensive
I really like the mains transformer out Very much indeed For preamps would be easier
Thanks a lot but better be safe than sorry I do not understand why oversizing is so demonized I can understand for shoes but for electronic parts ?
I have already bought some I really do not know if they will make any difference i understand they are for solar panels
It will be my way to go green ... they are so beautiful Marked 20A10
The transformer in that amp just isn’t very big. That’s why 6 amp diodes don’t blow. I’ll bet the short circuit secondary current is under 50 amps.
this say it all Good thing cannot come cheap
yes but incidentally the more expensive parts are those more important for sound Mains transformer and big power supply caps
A good transformer can easily cost 100 euro and the same for two high grade 20mF caps That makes 200 euro
If we consider that the ratio between parts cost and MSRP could be 5 times or more this means that in the end the amp should cost overall 2000 euro
I could end spending 200 euro and have a sound worth 2000 euro
The very only thing that still stop me are those bloody mosfet ... But i understand that i could do it carefully At the end of their mounting it will be only to check for the electrical insulation with the chassis I could even sand a little the chassis to get a better contact
In the meantime i have completely changed my mind I understand now the very good reason about this kind of mounting For instance this save the heatsink cost that is not little Very very smart Never underestimate British people ... after all they have conquered almost all the world