Parts Connexion has 220uF 100V Cerefines available
Elna Capacitor 220uF 100Vdc ROA
I used them in my build
Cheers
Pete
Elna Capacitor 220uF 100Vdc ROA
I used them in my build
Cheers
Pete
Man, those things have really gone up in price. Like Panelhead, I'm running some old BlackGate NXs in 2 of the 4 channels of my 4 channel FE. I only paid $5 for those in about 2015. The Audio Note Kaisei caps are about the same price, although I haven't heard them in comparison to any other cap. I guess in the scheme of things, $5 for a cap isn't that bad. I have certainly spent more money for other components. It does make the BlackGate PKs I bought for $1 each seem like a good deal.
Its funny how the topic of caps keeps coming up every few years. It seems that our choice gets discontinued and the remaining stock get higher prices. I would have thought the capacitor business would be more stable.
Last edited:
Through hole components are the issue. Except for snap-in caps and maybe rectifiers all new builds are surface mount.
Hello. Thank you for your reply.
I just received v3 kit and ordered a pair of PCB from Dario. This spring I'll try to build this pair (hope my family will bring me some free time
)Thank you too!
Yes and yes
I will slow down the relay. But going to 270ufd will have minimal impact. I am sure 470 ufd has been used by someone.
I will slow down the relay. But going to 270ufd will have minimal impact. I am sure 470 ufd has been used by someone.
Being a current gain amp the output impedance is high. You can measure. The amp impedance is in series with load.
The with the same source settings with a 4. 6, and 8 ohm resistor. Compare the voltage across each. You can solve for amp impedance.
This works with a voltage gain amp. Should work for for current amp also.
The with the same source settings with a 4. 6, and 8 ohm resistor. Compare the voltage across each. You can solve for amp impedance.
This works with a voltage gain amp. Should work for for current amp also.
Uhm, I would like to correct.. While the output devices actually are in a current pump configuration, then they are included in a global feedback loop, governed by the first stage in the composite topology. (the LM318).
The end result is that the output impedance is decreased by the loop gain. The loop gain is very high, combination of both the first and second stage voltage gain.
On the other hand, the current pumps have not an in finite output impedance; due to their composition, they are more like a dozen ohm. This was a measured choice from Mauro.
This will have to be divided then by the loop gain, which is 90/100 dB at low frequencies.
We are in the mOhm range, though frequency dependent.
I can measure, but will have to put together the setup..
And no, it can not be measured by a multimeter, do not try it, it will not be healthy for the instrument. ��
The end result is that the output impedance is decreased by the loop gain. The loop gain is very high, combination of both the first and second stage voltage gain.
On the other hand, the current pumps have not an in finite output impedance; due to their composition, they are more like a dozen ohm. This was a measured choice from Mauro.
This will have to be divided then by the loop gain, which is 90/100 dB at low frequencies.
We are in the mOhm range, though frequency dependent.
I can measure, but will have to put together the setup..
And no, it can not be measured by a multimeter, do not try it, it will not be healthy for the instrument. ��
You mentioned rev. A. Good point.
There is a difference between original compensation scheme (Evolution original; Myref C, Myref FE original, <> all of these with the latest Evolution rev. A compensation scheme)
The original compensation had a limited open loop gain setup; this resulted in a somewhat higher output impedance ( low frequencies). If I remember, Mauro declared a DF factor of 300.
The rev. A compensation mod has removed this, and it can be calculated with the full loop gain in place.
Ciao, George
There is a difference between original compensation scheme (Evolution original; Myref C, Myref FE original, <> all of these with the latest Evolution rev. A compensation scheme)
The original compensation had a limited open loop gain setup; this resulted in a somewhat higher output impedance ( low frequencies). If I remember, Mauro declared a DF factor of 300.
The rev. A compensation mod has removed this, and it can be calculated with the full loop gain in place.
Ciao, George
The original compensation had a limited open loop gain setup; this resulted in a somewhat higher output impedance ( low frequencies). If I remember, Mauro declared a DF factor of 300.
The rev. A compensation mod has removed this, and it can be calculated with the full loop gain in place.
If I understand, but as always I doubt given my skills with electronics and English, in My ref. we have an output impedance in the order of 0.1-0.15 Ohm and in Rev.A it is even less (or more ... this is my difficulty in interpreting) since we can calculate with the full loop gain.
Thank you George
Ciao a tutti
Giacinto
The rev. A compensation mod has removed this, and it can be calculated with the full loop gain in place.
If I understand, but as always I doubt given my skills with electronics and English, in My ref. we have an output impedance in the order of 0.1-0.15 Ohm and in Rev.A it is even less (or more ... this is my difficulty in interpreting) since we can calculate with the full loop gain.
Thank you George
Ciao a tutti
Giacinto
The compensation changed. Also the output resistor value dropped. The compensation adjustment allowed for use of a larger value input coupling cap without stability concerns.
Sonically, it makes excellent better.
If you look at the FE thread on group buy there are tests run by Joseph on a FE Rev A. An order of magnitude lower noise.
I never had issues with stability with My_Ref A, C, or Evo. But seen it mentioned when operated under extreme conditions. The Evo A compensation addresses this potential issue.
Far as I can tell the changes are all positive sonically and operationally.
This is not easy because the LM3886 is not unity gain stable. The circuitry used resolves this basic issue. All variations of My_Ref are successfully stable with the low voltage gain of LM3886.
Sonically, it makes excellent better.
If you look at the FE thread on group buy there are tests run by Joseph on a FE Rev A. An order of magnitude lower noise.
I never had issues with stability with My_Ref A, C, or Evo. But seen it mentioned when operated under extreme conditions. The Evo A compensation addresses this potential issue.
Far as I can tell the changes are all positive sonically and operationally.
This is not easy because the LM3886 is not unity gain stable. The circuitry used resolves this basic issue. All variations of My_Ref are successfully stable with the low voltage gain of LM3886.
I cant decide on proper transformer rating.
I know, that Dario and others recommend 160+ VA for toroid.
I can order 150VA (2x 22V x 3.4A) transformers for ~35$/1pcs or 200VA (2x 22V x 4.5A) for ~42$. The difference is only 14$ for pair, but do we realy need it?
I mean not pushing limits of this amp, but everyday listening.
P.s. I will use industrial BOM.
I know, that Dario and others recommend 160+ VA for toroid.
I can order 150VA (2x 22V x 3.4A) transformers for ~35$/1pcs or 200VA (2x 22V x 4.5A) for ~42$. The difference is only 14$ for pair, but do we realy need it?
I mean not pushing limits of this amp, but everyday listening.
P.s. I will use industrial BOM.
I went with a single 300VA Antek with dual 22V secondaries and it works great - super quiet no hiss/hum at all. If it were me, increase the VA and use only 1. If you are set on dual-supply, save your money and get two 150VAs.