X100 backengineered here

minor adjustments

...some 'minor' adjustments:
as BC´s were running to their limits, swapped with BD140,
added a 4th pair of Powerfets, downscaled the bias by adding 47k
in the currentsink, changed resistors in the sink from 1k5 to 1k6
and 1k to 910R to reduce turn-on transients, added standby-
relays, (not drawn) doubled powersupply to 2x 500VA, 2x18V

...have to think about buying stocks of some aluminium company...

:D Uli
 

Attachments

  • xa1_v3.zip
    31.4 KB · Views: 1,414
Success!!!

Thanks to all those who offered encouragement along the way.

I have now tested the unit in a workbench and it worked flawlessly - it is too early to give distortion figures as this will be a function of build quality etc. The design is essentially that of page 1 (with minor resistor changes) + a VBE multiplier at the output instead of the simple resistor (which should work as well, but which does not provide negative temperature coefficient).

The only real trouble (apart from time and mechanical work) was:
1. Understanding circuit and it's inherent trickyness in adjustment
2. Having adequate heatsinking on the input stage, particularly bottom current sources. Naturally version 3 will have much more significant heatsinking, and probably higher currents as well. I am now at 30mA for each of the two legs.
3. Next version will have autobiasing of all current sources even though current setup appears very stable.

I have aquired a truly excellent case and plan to run this unit in 2 modes: 100W with 10W Class A switchable to 100W with 50W Class A (yes, I know this requires huge heatsinking, but it is intended for winter use.) For various reasons, the unit is presently set up for 600W output so I will have to lower the gain. Even without my planned emitter follower, the driver stage did not have any trouble driving a bank of 16 high capacitance IRF TO3P devices cleanly and without loss of amplitude at 110V peak to peak to 100KHz which was my highest frequency generator.

I am considering switched mode power which would make a lower powered Class A device easily attainable through switching the rails.

Petter
 
Re: Success!!!

Petter said:
Thanks to all those who offered encouragement along the way.
....
2. Having adequate heatsinking on the input stage, particularly bottom current sources. Naturally version 3 will have much more significant heatsinking, and probably higher currents as well. I am now at 30mA for each of the two legs.
.....
Petter
So you are already
"on the way"
to next version .....


This is also typical for mankind:
will we ever be content and happy
with the good things we have
lay back
and enjoy

The rich ones want to get more money, and more ...

This is UN-Natural
Ever seen a Cat or Dog, after it has gotten
some food to eat

That is Natural Way to be.
Laid back and content
Enjoying, just the fact you are alive

When you are dead - it will be too late
to be happy & content
with your life

/halojoy 2003

halo :cool: relax now, dear petter - put on some of best lovesongs
take a rest
:) In Good Music :)
 
uli:hi

why original aleph ccs do not need this bootstraped supply circuit, but yours need this?
can you explain this detailed. if i design your ax100 , i will leave the transistor in aleph ccs alone , do not attach the mosfet(irfp9610) 's Drain . what is the advantage of your design?
 
haha

hi uli.

how stupid i am! yes , both circuit about bootstrap supply is identical. i read them too recklessly. thanks uli.
do you understand my second question about location of the transisitor in Aleph CCS.? maybe my language is too hard to understand. let me explain it..... maybe more explanation more
obscurity, haha.
i mean, in normal Aleph , the transistor of aleph ccs exists lonely,
do not attach any driver mosfet , it is only a part of CCS. but in your circuit , the transistor connects 1K attached by the drain of mosfet( irfp9620). i just think this transistor needn't be relational to irfp9620. why do you design such .? what advantage?
 
The one and only
Joined 2001
Paid Member
I believe that crowd moved on to Aleph X, which combines the
X circuit with the Class A Aleph output stage. You will find it
"permanently" at the top of this forum. From there, many have
become excited about Gain Clones, which has its own forum.

:cool:

pass: / still likes the way Halo signed off.
 
I confess to being less than excited by chip amps, although I've got a couple of 3886s (or whatever the part number is...) lying around. I've always thought that Petter's investigations here were of interest. I have a couple of variations on the X circuit that seem fun to me, but then I'm easily amused.
It certainly wasn't my intention to put a damper on this thread by starting the Aleph-X thing.
So...just to see if anyone is listening...why not try for a class D X amplifier? The "X" aspect might be of use in removing residual nasties. Who knows, it might just be the ultimate subwoofer amp.
How's that for a blatant attempt to prime the pump?

Grey
 
:(

Nelson Pass said:
I believe that crowd moved on to Aleph X, which combines the
X circuit with the Class A Aleph output stage. You will find it
"permanently" at the top of this forum. From there, many have
become excited about Gain Clones, which has its own forum.

:cool:

pass: / still likes the way Halo signed off.


I have to admit, the first amp that I built was of the GC variety, but my first and still most poignant draw to this forum has been the Pass section. I have a stack of heatsinks, Dale's Aleph 2 boards and the AX boards. The problem I have is dedication. The GC was too easy and therefore jumped up the list. I learned from it and hope to learn from the P3A that I am about to build (the first boards I ever bought).

As punctuation to this otherwise non-contributing response, I went to DIY Atlanta and heard a DIY Aleph 5. For the first time in my life, I heard something that was beyond real. To date, my goal is to complete the Aleph 2's I decided to try. Either I'm too chicken or too much the perfectionist to try it. If only courage came at 6:00, instead of at bed time. . .

Sandy. /Who misses the time when the activity on the Pass forum was the biggest on the site. . .