Finished capacitance multiplier

Status
Not open for further replies.
Jump to Latest
#341
Ricky: Combine how?

edit:

I think I see what your mean, instead of transformer and rectifier, use the SMPS as the unregulated portion fo the power supply... you can do that, but not sure how much is gained. Low frequency (bass) transient response should improve, but can't say until you test. LC reported a considerable sag under high (5A) load, so that may work against you at the same time. You still have to adjust the voltage drop at the pass transistor to the maximum expected ripple from whatever is before the filter, SMPS, or rectifier/reservoir cap.

No transformer, that's a plus... 😀

The ALF's are probably worth the wait, 😉
 
Last edited:
#342
PMI said:
That is a true statement as far as it goes, but it is not the whole story. We may not be able to hear a frequency above 20KHz, but in most cases, we still care about the performance of the amp above 20 KHz, because it has an effect on the accuracy with which the sound is reproduced. The good qualities of VSSA include extended frequency range and high slew rate. If nothing above 20KHz mattered, we would not care about that.

High frequency switching noise will modulate the audio signal in ways that alter what we hear. One of the first things that LC did was modify the Hypex SMPS with larger and better quality capacitors, to reduce the high frequency ripple.
Click to expand...

Yes, my statement was over simplified. I was just suggesting that the higher frequency the ripple the less audible its effects become, so with an good SMPS it will be far cleaner sounding than a linear supply with 120Hz ripple, given similar ripple amplitude for a given load.

Of course reducing the ripple is the name of the game for this thread with the capacitance multiplier. I should be receiving some MOSFETs tomorrow and will be cobbling together a simplified version to test with.
 
#344
I went back for a look. It is certainly in the 'Go Big Or Go Home' category! I think I heard Tim The Tool Man Taylor in the back of my head while checking it out...

I'm just going to do some testing with an artificially soft raw supply to see what kind of performance is possible with the simplified circuit then switch to a stiffer production supply if I'm happy. I'm also looking to test the ability to shut it down rapidly as a form of DC offset protection.
 
#345
jkuetemann said:
...I'm also looking to test the ability to shut it down rapidly as a form of DC offset protection.
Click to expand...
I am pretty sure that can be done, although it might require a circuit which is isolated from the filter, through an optocoupler or something similar.

Do you know how quickly one would have to shut the supply off, to be effective?

What I mean is, a remote shutdown, or remote on/off is one thing, the "rapidly" part I am not so sure of.
 
#346
I already have a shutdown circuit in the simulator. Have a peek at what I'm working with here. The pass transistor cuts off immediately in sim, so the limiting factor becomes how quickly the charge on the output capacitors can be dumped. With a 1000uF capacitor and an 8 ohm load the output is zero in 100ms.

This circuit will trigger on frequencies of <10Hz at full output and DC as low as about 2V in a little over a second (a full rail voltage of either polarity triggers it almost immediately).
 
#347
jkuetemann said:
I already have a shutdown circuit in the simulator. Have a peek at what I'm working with here. The pass transistor cuts off immediately in sim, so the limiting factor becomes how quickly the charge on the output capacitors can be dumped. With a 1000uF capacitor and an 8 ohm load the output is zero in 100ms.

This circuit will trigger on frequencies of <10Hz at full output and DC as low as about 2V in a little over a second (a full rail voltage of either polarity triggers it almost immediately).
Click to expand...
It looks like it would do the job (I am not smart enough to find flaws on inspection though... 😀 ) What I meant above, was that I am not sure what the requirements are for the shutdown to be effective. Aside from that, I think it is a great idea, I like it.
 
#348
PMI said:
Ricky: Combine how?

edit:

I think I see what your mean, instead of transformer and rectifier, use the SMPS as the unregulated portion fo the power supply... you can do that, but not sure how much is gained. Low frequency (bass) transient response should improve, but can't say until you test. LC reported a considerable sag under high (5A) load, so that may work against you at the same time. You still have to adjust the voltage drop at the pass transistor to the maximum expected ripple from whatever is before the filter, SMPS, or rectifier/reservoir cap.

No transformer, that's a plus... 😀

The ALF's are probably worth the wait, 😉
Click to expand...
Thanks for answer Pete.
Intend to try most PSU types on same VSSA modules, before making final build of amps.
I will report later when times come and different setups been tested, i guess it will be at VSSA thread.
Ricky
 
#349
I must report that I had an issue with hum/buzz with my PeeCeeBee version of VSSA that used ordinary PSU but this issue was completely solved with Mr Evil's cap multiplier. At first I was in doubt if I had a ground loop but that was not the case. VSSA version without CCS has lower PSRR and for completely noiseless performance requires very good PSU. Mr Evil's cap multiplier is such circuit.

I am absolutely positive that all versions of VSSA will perform the best with this cap multiplier. It is cheaper than SMPS, has inherent "soft start" and in any sensible application will be completely noiseless. I already told to PMI that there are only few circuits in any category in diyaudio forum that are this good. Unlike SMPS this circuit will not radiate HF polution, it's safe for user and has potential to work for decades without malfunction. (SMPS are reported to expire after relatively short period of time)

This was Mr Evil's finest hour!
 
#352
quan said:
Hi guys, silly questions re CM. I am intending to use two of CMs with my 2X300VA trafos. Do i still need soft start board?.
Quan
Click to expand...

If you use rectifier diodes with appropriate current/voltage rating and main smoothing (bulk) capacitors with appropriate voltage rating you do not need separate soft start circuit that will protect rectifier during turn on of the transformer. MrEvil's cap multiplier will slowly ramp voltage at it's output so that amp circuit will not receive sudden voltage spike and produce unpleasant sound at loudspeaker terminals. When I used ordinary supply there was thump at the turn on, but with MrEvil's cap multiplier absolutely nothing - total silence. Therefore amp circuit is in some way protected from sudden application of supply voltage by the cap multiplier. This is of some importance because with some circuits (VSSA among others) there are some components that are sensitive to this instant supply voltage. Of course, preamplifier and source can produce unpleasant sound at the turn on too, but this is something to deal with separately.
 
#353
quan said:
Hi guys, silly questions re CM. I am intending to use two of CMs with my 2X300VA trafos. Do i still need soft start board?.
Quan
Click to expand...
Not silly at all. Ivan and Andrew are both correct, but they are talking about different things.

Andrew is talking about the mains fuses & wiring before the transformer, especially if transformers are intentionally oversized, in the attempt to make a "stiff" power supply. When using Cap Multipliers (any type, not just this one), there is usually no reason to oversize the trafo, unless you just have them lying about.

A typical Class AB amplifier should not need a soft start with this supply (one or two of them), when using a toroid up to about 500VA. The inrush due to that size transformer is fairly small, and the startup load from 10.000uFd reservoir caps, even a total of 4 on two boards, is manageable, unless the mains wiring in your home is intentionally under-rated. An example of intentionally under-rated mains fuses (breakers) are some University residence rooms.

If you have large toroids (800VA and above), especially two of them, and/or large capacitor banks, you will need a soft start circuit to protect the mains wiring.

Ivan is referring to the diodes, caps, and circuitry after the transformer. The voltage in the Cap Multiplier power supply ramps up very slowly. a couple hundred milliseconds or more, depending on the time constant of the multiplier RC filter. No inrush into the amplifier, and no large transients in, or after the power supply. No turn-on noises from the speakers, etc.

Test Results, Startup and Shutdown

In other words, the only contribution to an inrush is what is before the CM Filter PSU.
 
Last edited:
#355
Soft start circuit is needed with high power sound reinforcement amps that use really big transformers. That's not what we use for home listening. Slow charging is necessary with large capacitor banks that are sometimes used with class A home amplifiers (some Pass amps). The easiest way to achieve this is to use NTC resistor in line between bridge rectifier and cap bank, to prevent bridge damage at turn on. Sometimes NTC resistor is used between transformer secondary and bridge rectifier. In both cases it's hard to decide the power and resistance of the NTC resistor.

With reasonable size transformers and MrEvil's cap multiplier we do not need neither the soft start nor slow charging circuit. That's what makes this circuit so special, great filtration without extra expenses. And I believe we need circuits like this one for the not-so-bright peak oil future.
 
#358
ivanlukic said:
Unlike SMPS this circuit will not radiate HF polution, it's safe for user and has potential to work for decades without malfunction. (SMPS are reported to expire after relatively short period of time)
Click to expand...
Hi all,
excuse, just for clarity (which is considered public) to those who pass by this thread.
It is not the smps which has a short life, but the electrolytic capacitors. you do not use?
I do not give that a smps breaks if it is mounted properly and if it is built well.
The problem is buying small amounts at a retailer, such as electrolytic capacitors. maybe have 2 to 5 years of storage in the warehouse. this is a big problem. but we all like to dream. 🙂

for HF radiation...yes know, especially when you want to use a smps for lamps as audio psu.
I like this Evil regulator, I have always pushed the use of a regulator and an psu for an amp, then no problem.

Regards
 
Last edited:
Status
Not open for further replies.
Top Bottom