That's funny. When I see those initials it reminds me
of a story my pop told me about a drummer named Gene Krupa...
My pop went to the same school, in Indiana, called St. Joseph's College.
Apparently the Prefect wasn't too thrilled to see a framed picture of Jesus
Christ over Gene's bed.
It was personalized and autographed, as follows:
To Gene -- From J. C.
You might want to watch my other CFA get designed.
(below 1) on Sunday , I will try to get the layout perfected.
Same VAS/super-pair as the K-C , the ND loses the caps !
Will the diamond be musical ? Electrically , regulated mirrors = no
Ic change with temp.(servo'ed resistors) , Q1-4 cancel , baxandall VAS cancels.
ZERO thermal drift - lesson learned from that evil "infidel".
Last CFA ... (hopefully the best).
PS - VFA "Wolverine" has the lowest THD. All these IPS's are <20ppm -
Could you hear these differences ??
OS
Attachments
I was just playing to see how good the servo handled mismatches.
you can mix 992/1845's with any other pair - even dis-similar pairs
will give PPM + 0V offset.
The K-C's system really holds up well with this 2 stage diamond
(better , even).
2sa/C 992/1845's are standard for Q1-4 .
With all 4 running at full rail - faster + zero thermal drift (If they are
all coupled).
A lower rail <50V could use Ksa/c 1015/1815.
Not a "picky circuit" (like the K-C) !!
OS
Here's what I want to start messing with. http://www.onsemi.com/pub_link/Collateral/BC846BPDW1T1-D.PDF Still need to find a worthy SMT replacement for the A1381/C3503. All this stuff is on sale at Newark right now. I'm stocking up.
For the spooky,wolverine, symasui you would want the same gender on the same die (s).
For this CFA , npn/pnp on the same die. actually ... all 4 input devices in
the same package for both amps (spook - K-ND).
Look below - all 4 coupled (circled) , feedback away from input , symmetric
returns to ground-OPS in for the 1W resistors.
If you look at the datasheet , you will see the same divergence between
P and N channel as you would see in separate devices.
for SMD , I would just mount 4 devices close and use thermal epoxy to
glue a small copper plate over them .... or just slap a big blob of epoxy
right over the whole deal. 😀
PS - sprint sucks for smd . wish I had something "fancier" . 🙄
OS
Attachments
Last edited:
Dual PNP and NPN are on sale too. They top out around 65V though.
Diptrace is free for non-profit and is great for SMT. It's completely different than Sprint but once you figure it out it's fast to work with and it will verify your connections for you if you enter the schematic properly.
Did-you tried to put the input filter in between the two stages ?
It will dump the diamond as well as making its FC insensible to source's impedance.
Too, the two paths for feedback is not optimal, omho, as the slightest mismatch in the resistances values increase distortion.
Just my two cents.
I hope so at least for now.
OS is working on the K-X? or something.
I got them from JW, he is the keeper of the Gerbers?
Then
Where do I find the additional boards?
I have to write down somewhere the other boards.
Does Anyone have a BOM? or a Partial BOM?
By name at Mouser? or soemthing for this stuff?
If I have to I"ll start putting together myself but I need
some kind of information.
Please....
Is it okay if I send the Gerbers to other people?
or do you just want one person to do it so it doesn't
have a problem down the road?
I got a pic of it and it looks good. I like the nice large pads
for through hole. Not sure how thick they are but by the looks
of them they shouldn't pop up nor pull off the trace when desoldering
to chance componant values, etc.
Cheers,
OS is working on the K-X? or something.
I got them from JW, he is the keeper of the Gerbers?
Then
Where do I find the additional boards?
I have to write down somewhere the other boards.
Does Anyone have a BOM? or a Partial BOM?
By name at Mouser? or soemthing for this stuff?
If I have to I"ll start putting together myself but I need
some kind of information.
Please....
Is it okay if I send the Gerbers to other people?
or do you just want one person to do it so it doesn't
have a problem down the road?
I got a pic of it and it looks good. I like the nice large pads
for through hole. Not sure how thick they are but by the looks
of them they shouldn't pop up nor pull off the trace when desoldering
to chance componant values, etc.
Cheers,
Last edited:
OStripper,
Your Kypton - ND looks like it may be your best yet. I will be paying close attention to this one. Thank you for sharing!
Your Kypton - ND looks like it may be your best yet. I will be paying close attention to this one. Thank you for sharing!
OS,
The conversation the other day about using two opamps in the servo system, how does that relate to your design? That was both Esperado and Bob Cordell saying that. I don't know enough to understand your servo circuit so I'm just asking. And now we have something better than the Krypton-C, the ND! Will it never end!
The conversation the other day about using two opamps in the servo system, how does that relate to your design? That was both Esperado and Bob Cordell saying that. I don't know enough to understand your servo circuit so I'm just asking. And now we have something better than the Krypton-C, the ND! Will it never end!
OS are you going to consider implementing the servo system with the BB op amps Bob Cordell described on the other thread? Thanks for all your hard and Great work don't stop now. It is a dream school for amp design and "Implementation then building"
The servo of OS modify the current balance between the two CSS instead of injecting a voltage in the feedback path.
Christophe,
So does that mean that he is controlling the dc offset in a different portion of the circuit rather than in the feedback node? Is that what I should understand here?
So does that mean that he is controlling the dc offset in a different portion of the circuit rather than in the feedback node? Is that what I should understand here?
Yes. Changing the current of the CCS modify the operating point of the input transistor, thence, the voltage on their collector. The benefit is that it is an other feedback path. I don't know witch is better, distortion side.