Aksa Lender P-MOS Hybrid Aleph (ALPHA) Amplifier

[xrk971]

So we are talking about 15mV of ringing under 10nS rising edge square wave excitation? If square wave was 8vpp, would ringing be commensurate or still in mV range? I don’t think this sounds like an audio issue anymore if it requires a pathological example of excitation.

 

[keantoken]

Output is proportional to input. The simulation used a 100mVpp square wave input. There is more ringing than that because the HP filter attenuates it a bit.

You would get more ringing in proportion to the input square wave level, but it would still be hidden in the root of the wave and undetectable, hence the HP filter to make it detectable.

 

[zman01]

Watchtmouse,
You are moving fast. 100uF 16v is all you need and not 1000uF 50v. That was my fault for over spec. I would use 220uF 25v if you have it. It doesn’t see anything higher than rails / 10 or 2.5v.

Hi X,

Exactly which position is being discussed here? A 100uF to 220uF cap for C103?

IIRC the bill of materials had 47uF for C103 - a change to that is being suggested?

 

[xrk971]

Hi X,

Exactly which position is being discussed here? A 100uF to 220uF cap for C103?

IIRC the bill of materials had 47uF for C103 - a change to that is being suggested?

Hi Zman,
The cap I was referring to is the feedback shunt cap C111 below (220uF 25v is plenty here):

 

[irribeo]

http://www.diyaudio.com/forums/atta...ph-alpha-amplifier-img_20180415_104926978-jpg

Question was about c103 as in picture, not c111.

 

[xrk971]

Thanks, Irrebeo,
My mistake - C103 needs to be 1000uF and at least 35v and the large value is for the anti thump feature.

Folks:
In the future please attach schematic and specify part label when asking questions about specific parts and mods. Hugh’s schematic and JPs64’s Schematics have different labels.
Thanks

 

[watchtmouse]

Thanks, Irrebeo,
My mistake - C103 needs to be 1000uF and at least 35v and the large value is for the anti thump feature.

Folks:
In the future please attach schematic and specify part label when asking questions about specific parts and mods. Hugh’s schematic and JPs64’s Schematics have different labels.
Thanks

Wasted some shipping costs here, yesterday I ordered the wrong value, no big deal.

Then again the question, but with more context ;-)

1). Alpha 20 JPS64 schematic, The C103 cap, Due to the limited size on the Alpha 20 board, a 1000uF capacitor will not fit, the max diameter for the C103 cap is 10 mm and the RM is 5 mm, this gives me a max of 470 uF 50 V (10 times the original of 47 uF (C103 is named C2 in the original AKSA schematic).
Now to the question, is the 1000 uF important or will any bigger cap as 47 uF do?

Sorry for being a nucance

Happy listening

Bert

 

[AKSA]

470uF is fine! This caps affects only turn on and has no influence on sonics!

HD

 

[cl00sed]

Hugh,

Could you please provide possible list of fets for alpha 20 4ohm version
I have the irf’s for 8ohm version

Regards

 

[danny_66]

See here for transistor choices

IXTK88N30P TO264 600W 88A 6.3nF 300V 55S
IXTK90P20P TO264 890W 90A 12nF 200V 50S

 

[cl00sed]

See here for transistor choices



IXTK88N30P TO264 600W 88A 6.3nF 300V 55S

IXTK90P20P TO264 890W 90A 12nF 200V 50S

Thanks Danny, am looking for any alternatives like in Fairchild, etc . I could not able to get the ixy’s locally

 

[xrk971]

Thanks Danny, am looking for any alternatives like in Fairchild, etc . I could not able to get the ixy’s locally

If you look at the bottom of the first schematic by Hugh on Post #1, you will recommendation for Fairchild parts:

FDA38N30
FQA36P20 (but I think Hugh meant FQA36P15 as I could not find a P20)

Capable of 300w dissipation.

 

[cl00sed]

If you look at the bottom of the first schematic by Hugh on Post #1, you will recommendation for Fairchild parts:



FDA38N30

FQA36P20 (but I think Hugh meant FQA36P15 as I could not find a P20)



Capable of 300w dissipation.

Can use this in place of FDA38N30?
TK39J60W5,S1VQ(O Toshiba | Toshiba TK39J60W5,S1VQ(O N-channel MOSFET, 39 A, 600 V DTMOSIV, 3-Pin TO-3PN | 799-5135 | Welcome to RS Online

And the FQA36P15 is
FQA36P15 ON Semiconductor | ON Semiconductor FQA36P15 P-channel MOSFET, 36 A, 150 V QFET, 3-Pin TO-3PN | 671-4938 | Welcome to RS Online

Regards

 

[danny_66]

I would rather select the ones in a TO-264 or SOT-227 package,
for better heat transfer to the heatsinks

 

[maty tinman]

One question:

How much is the phase shift/deviation at 20 kHz? And a 100 kHz?

 

[xrk971]

Hugh or Danny_66 can confirm, but from what I recall of Hugh’s sims - almost zero. It’s flat up to nearly 1MHz.

 

[irribeo]

One question:

How much is the phase shift/deviation at 20 kHz? And a 100 kHz?

20kHz is given for Alpha and BB, 5.3 degrees

 

[maty tinman]

Thank you very much to both.

I read that one of Hugh's goals in his amplifiers was to minimize such shift.

Yesterday I was reviewing, again, the manual of the new IcePower 1200AS2 class D module and I was struck by this graph, where the shift is almost 70º at 20 kHz!

 

[irribeo]

Yes combine with your Kef's phaseshift and you might get lucky.

 

[AKSA]

FR and Phase relationship of ALPHA20

Here it is, pretty good too! 7 degrees at 20KHz, and 33 degrees at 100KHz. Much of this is due to the input LP filter; I have set it a 330R//1nF at the input.

The LF depends on the size of the input cap. I have set this to 10uF but down to 1.5uF is acceptable if you want really good caps (large!)

FDA38N30 OR FQA40N25
FQA36P20 (but I think Hugh meant FQA36P15 as I could not find a P20)

FQA36P15 is correct, thanks X and Danny!

Cheers,

HD