Slewmaster - CFA vs. VFA "Rumble"

The final VAS current on jason's board has 3 variables.

1- the VAS Re's

2- The resistors from base to rail on these devices, (R5/6) in the post.

3- the IP current. Jason's 500R VR1/3 need to go a little higher to get
<1.7ma at the IP pair. 1K trimmers would give you a range of .5ma -
(infinity - too much).
PS- I use a fixed resistor in series with the trimmer for safety and a limited
range of 3- 7ma VAS ! (470 -670R at CCS)

You should be able to get ANY VAS current with these changes.

OS

Hmmm, not sure what I did wrong then. I tried adjusting both pots to their full extent and the lowest I could get was 1.2V across the 150R resistor. If I'm doing the math right that's 8mA. what is the problem with using the 47R/390R combination?

Thanks, Terry
 
Last edited:
I doubt there is anything wrong with the values I suggested, they just aren't the same as those suggest by OS. I think he was trying to get the VSSA patchwork IPS to be closer in configuration the the official CFA-X (v1.2). I think OS is suggesting to run a little less current in the input pair and degenerate the VAS a little more. I can't say how much of a difference those changes would make as far as actual performance goes.
 
Having the greatest range for the CCS's is the major variable.
Even with 47/390 or 150/820 , or any other combo .... being able to
have an extended range of .5 - >2ma would allow you to "dial in" any final VAS
current.

PS - 500R on my "CFA-X" is the max Ic for the IP pair (2.0ma).
@ 680R is needed to get my desired 1.6ma Ic.

There is also no harm in running the VAS "hotter" at 7-8ma ... as long
as you still have enough (trimmer) range on the OPS to properly bias it.

OS
 
Last edited:
I had it set up through my A/B setup so I had it running against my Dx Super A. This amp seems to have a little too much sizzle and high mids compared to the Super A. It may just need some more time to break in. It might benefit from a little higher bias too. I have it set at 25mV across a pair of emitter resistors. That is about all my heatsinks can handle. I'll play it a lot tomorrow and see if it calms down a little.

Blessings, Terry
 
To much "sizzle". You might have a little "overshoot".

Those different Re's (47/390 vs. 150/680-820) will trim loop gain.

The main compensation caps .... (see Thimios's variable capacitors ?) ,
can also "trim" the "sizzle" to a desired optimum.

What is actually being changed with the above scenario's is the
"sharpness" of the square wave ... from rounded edges to a "hard"
sharp "cliff". :D Simulated high order harmonics change drastically
as you change these values.

The final "sound" can be anything that pleases you ...with proper
"tweaking". I'm sure this process is also true at the OEM R&D labs!

OS
 
To much "sizzle". You might have a little "overshoot".

Those different Re's (47/390 vs. 150/680-820) will trim loop gain.

The main compensation caps .... (see Thimios's variable capacitors ?) ,
can also "trim" the "sizzle" to a desired optimum.

What is actually being changed with the above scenario's is the
"sharpness" of the square wave ... from rounded edges to a "hard"
sharp "cliff". :D Simulated high order harmonics change drastically
as you change these values.


The final "sound" can be anything that pleases you ...with proper
"tweaking". I'm sure this process is also true at the OEM R&D labs!

OS
Will be useful if we know what frequencies these must test/trim.
Thimios.
 
Last edited:
CFA-XH (v1.3) Redraw Progress

A quick snapshot of where I'm at with the CFA-XH IPS. I have not yet decided how to connect the star to the input ground lift components, it is currently absent. Some minor pushing and pulling, but leaving more room in some areas makes my art 3"x3.5", so a little bigger than OS's.
 

Attachments

  • CFA-XH.JPG
    CFA-XH.JPG
    565 KB · Views: 578
Will be useful if we know what frequencies these must test/trim.
Thimios.


(below)

A generalized plot of what happens.

3 factors will affect harmonics.

-OPS bias
-Compensation capacitor value
-resistor loading of VAS

Last 2 also effect slew rate.
This effect might be interpreted as "sizzle" :D .
This is also (one of the factors) what makes tube amps/SS amps different ,
or what separates a good amp from a outstanding amp. :)
OS
 

Attachments

  • CFA-X-FFT.jpg
    CFA-X-FFT.jpg
    41.3 KB · Views: 530
Where are the schematics for the 1.3 versions or did I miss something? The first post of the thread links to v1.2 for the cxa and wolverine.

I looked , 1.2 -1.3 are the same. I might of tweaked a resistor value
or two. PCB is identical.

(schema below) :)
Edit - I did update the PCB (either will perform identical).
below 2/3 - allows diode or active transistor saturation clamp.
If you are using 1.2 - It is still SOTA !!
OS
 

Attachments

  • wolverine v1.3.jpg
    wolverine v1.3.jpg
    122.3 KB · Views: 614
  • wolverine-C-V1.2.zip
    44.3 KB · Views: 181
  • wolverine v1.3pcb.jpg
    wolverine v1.3pcb.jpg
    181.8 KB · Views: 590
Last edited:
OK, I hooked up the scope and am attaching some pictures. These are all square waves with an 300W, 8ohm dummy load attached. Please take a look and give me some advice as to what, if anything should be changed.

Thanks, Terry
 

Attachments

  • 100k.jpg
    100k.jpg
    194.7 KB · Views: 136
  • 50k.jpg
    50k.jpg
    203.5 KB · Views: 140
  • 20k.jpg
    20k.jpg
    213.3 KB · Views: 146
  • 10k.jpg
    10k.jpg
    215.9 KB · Views: 136
  • 1k.jpg
    1k.jpg
    217.6 KB · Views: 404
  • 500hz.jpg
    500hz.jpg
    214.2 KB · Views: 414
  • 100hz.jpg
    100hz.jpg
    241.5 KB · Views: 522
What is the output level? Are you swinging through from rail to rail or are we looking at a lower voltage like 10Vp-p? Might make a difference in what the net result is, but it looks like some speed is lacking. What are your current values for R5, R6 & R13, R14 and COMP1 and COMP2? Also, I specified a lower cutoff frequency for my input filter at about 220kHz so you could temporarily pull C2 and see what you get.
 
Driver heatsinks ....

Going through my E-waste pile today , I noticed most common
PC PS's (250-300w units) have the perfect 75mm extruded units for
the "slewmonster" (below).

Out of 15 SMPS's , 12 had 75-80mm (5-7C/w estimated) long
"overkill" extrusions. Could be lower than this.

A 300W SMPS at 88-92% efficiency would have 30w waste (heat).
Even as it is fan cooled - this is far more than the slew driver dissipation.

Being so common (PC SMPS's) this would be perfect (and free)
to apply to a build. :)

PS - with one of these , Driver Re @ 68 (or even 47-56)- no issue. A flat plate
is over 15C/W - 3 X the cooling for nothing !

OS
 

Attachments

  • PC-smps-HS.jpg
    PC-smps-HS.jpg
    140.9 KB · Views: 274
Last edited:
I checked again. Pretty healthy output in the pics. The amp starts clipping at about 41Vac into 8ohm with 1K input. The pics above represent about 30V output at 1K Of course at 100K it is probably 1.5Vac. I checked again with each channel and they both look the same. 100k on the square wave looks almost the same as the sinewave. No idea why 100k matters since I doubt I can even hear 20K anymore.