An idea of the layout ....
I want this for ME ! So (full tilt) "anal" we shall go ... 😀
(below 1)
EVERY return goes back to the star individually. I decided to go
with either 10V or 16v polarized 8-10mm blocker caps.
The "NFB" will be just one big pad.
Added protection diodes to the blockers , as well (D7/8 - below 2).
Most of the D's (diodes) correspond to the associated capacitor.
Easy BOM - easy PCB.
These are the final CFA-X-H designs ... it shall be done😀 .
OS
I want this for ME ! So (full tilt) "anal" we shall go ... 😀
(below 1)
EVERY return goes back to the star individually. I decided to go
with either 10V or 16v polarized 8-10mm blocker caps.
The "NFB" will be just one big pad.
Added protection diodes to the blockers , as well (D7/8 - below 2).
Most of the D's (diodes) correspond to the associated capacitor.
Easy BOM - easy PCB.
These are the final CFA-X-H designs ... it shall be done😀 .
OS
Attachments
OS,
Are you saying that these would be the best transistors that you would use that are in current production or that they will just do the job? If you had your druthers of any current output put transistors and drivers these would be what you'd choose? If they are then thank you for that. I'll start putting my list together to order the parts. On things like diodes you wouldn't use anything special such as Schottky types ever for anything outside a power supply? This is a bit different than building my Heathkit digital clock and other things were I only had to put the parts in the right places in the correct direction!
Are you saying that these would be the best transistors that you would use that are in current production or that they will just do the job? If you had your druthers of any current output put transistors and drivers these would be what you'd choose? If they are then thank you for that. I'll start putting my list together to order the parts. On things like diodes you wouldn't use anything special such as Schottky types ever for anything outside a power supply? This is a bit different than building my Heathkit digital clock and other things were I only had to put the parts in the right places in the correct direction!
I'm sure since their role is as a cascode stage other devices are going to be suitable. Although they are brilliant devices I try to avoid specifying anything that folks have to hunt around for.
The only "upgrade" I can think of is the IP pair.
Here we want to make for more "bombproof" while retaining
performance.
For this IPS (cfa-x-h) , each device of the IP pair "sees" rail + .6v.
For 70-80V rails ... 100Vceo + would do.
2sd667 /2sb647 or 2sc3228/2sa1275 .... any larger to-92 (L) with
800mw -1W specs , 100-120V , < 20 Cob (capacitance).
The 992/1845(as IP )would actually be alright with a small heatsink. The
above upgrade is to de-rate expected wattage/junction temps.
All the other 992/1845's could be substituted with (anything) !!!
They all run < 12V and at microwatts.
The to-126's for the VAS could also be ksa1220/sc2690 (A)'s
"a" is 160V device's 60-70V rails could be achieved.
The 2 CCS trimmers will allow for just about any device combo on this
IPS 😎 .
PS - mouser has all the 992/1845 + 1381 "E"'s in stock.
only "D" for the 3503 ... but they are cascodes.
Better bet is the KSA1220AYS / KSC2690AYS (the 1220 was shipped on
4/30 from fairchild) these are high voltage /gain. Excellent devices -
made for audio ... 🙂
Edit - one could even use drivers ! (toshiba 2sc4793/sa1837) for VAS devices.
Also , what performance will you lose ? Higher Cob seems to be the only limiting factor
when you approach the "magical 1KV/us slew rate ...
OS
Last edited:
I'm finding that I really like the KSA1220AY and KSC2690AY devices. They don't have the ultra-low Cob that everyone clamors for but in every other respect a fine device.
They are nice , except we will be limited to 75V rails and still have 20%
derating.
I have my trafo's - they are 45-0-45 400VA's ... @ 67V no-load rails.
The new IPS ... is coming along. I'm really trying for the best performance/
aesthetics balance . After seeing Still4given's OPS's - I want a "match". 🙂
OS
Attachments
Please do share the Sprint file when you are satisfied, I want to make a match to the OPS re-draw and output some gerber files for manufacture. I think this will be the first to see production.
I used your rounded holes and such on this one.
Another thing I added is standard miller compensation. THD is lower still,
but the products (H2/H3) are different.
Pads are there to do it both ways (shunt or miller).
Unbelievable <5ppm 4R high output (below)
shunt = 8-12ppm ... miller= 4-6ppm ... used many different models.
- - same output.
This is a really good circuit. 🙂😎😎
OS
Attachments
Thank you ... those are EXCELLENT !! 😱
2.5pF Cob. Still , close to the 1220/2690 at 150vceo. 75 v rails
are the top limit.
OS
By all means do your layout in your usual style, no need to bend to my tastes.
I have a certain 'anal-ness' myself and will almost certainly want to do a re-draw anyway to suit my silk screen style and PTH, etc.. I will also take advantage of the top to make for no jumpers. I have given up on toner transfer and will almost always order from a board house for anything worth while from now on.
Soldier on my maniacal DIYA friend!
Yeah, I looked into doing photo sensitive etching but the boards cost more than finished boards from the board house.
Can't wait to actually here a completed input and output section. I am in for 4 of these input boards whenever someone makes them. I'll need to get two more of the output sections so I have two pairs. I'm getting excited, this should be something nice.
You won't have much to do. I made for jumper-less.
Re-lable away with your "style" 😀 ..
-Both trimmers "trim" the same way ,
-ALL returns are starred.
-Electro's are in the same direction.
-diode protected blockers and 2 compensation methods.
The best , most "anal layout" to date . Should facilitate close to
that 1 KV/us slewmaster threshold 😱 😀 . (with a few tweaks
and select components).
OS
Attachments
That is really neat!-Both trimmers "trim" the same way ,
-ALL returns are starred.
-Electro's are in the same direction.
-diode protected blockers and 2 compensation methods.
The best , most "anal layout" to date . Should facilitate close to
that 1 KV/us slewmaster threshold 😱 😀 . (with a few tweaks
and select components).
OS
I like that compact board.
But I have a question: What does it look like with the outputs on it?