Jump to content

and now for something completely different part 3


kevin gilmore

Recommended Posts

Interested in building a single-ended CFA2 (or is that CFP2)? Can't find the schematic matching the cfp2gmt board files on the archive... I have the cfa2.pdf schematic, but it doesn't match the board file. Maybe I'm not looking for the right name here...

Also, what is the board size on these?

I'll have to admit I am having a hard time following this thread and all the loops and progressions from se to balanced to power amps... :)

  • Like 1
Link to comment
Share on other sites

39 minutes ago, Pars said:

single-ended CFA2 (or is that CFP2)?

File names are CFP2 which stands for Current Feedback Amp so sometimes referred to as CFA

CFP2 is SE                     CFP3 is balanced and Large

I have built dmt ver1.05, emt ver1.06, and had boards made and built fmt ver 1.07 which are 3.7" x 4"

they all work

gmt Ver1.09 is the newest

from memory there were small changes only between them

what part of schematic does not match?

its been a while but will try to help

Link to comment
Share on other sites

I have some CFP boards made for my own use - I plan to build a balanced CFA. I have some extras so PM me if you are interested.

These are 2mm and 2oz copper boards.They are about 101mm x 94mm. These are the gmt, version 1.09 boards.

EDIT: Here is an image of the board. I believe the Gerber file is cfp2gmt.  Kevin may have laid out a full balanced board which includes an unbalanced-to-balanced converter on the board that uses THAT340 but I am not certain.

I plan to build my balanced CFA using two pairs of boards as I believe that's what congo5 has done.  

cfp2gmt.jpg

Edited by mwl168
Link to comment
Share on other sites

Thanks Michael!

This has some differences from the BOM I was putting together and from trying to follow the thread.

No 15K resistors (guess they're 5K now)

20K resistor

22pf cap

I take it the 2x50K resistors in the schematic are now 1 100K?

Same with the 2x34K resistors are the single 68K?

I can't find the schematic mentioned in the thread (cfp2f.jpg), 2nd page of thread around April 1, 2015.

Edited by Pars
Link to comment
Share on other sites

This is the only schema I have and it does not match the board. I did not trace it but Q27 and Q28 are replaced with PZTA06 and PZTA56 on the board because they have higher dissipation ratings. I am in the process of building mine and just follow what Kevin always said to trust the board. I'll let you know if it goes boom!

EDIT: attached the schematic in PDF in case the image is difficult to read.

CFP.JPG

cfa2 (1).pdf

Edited by mwl168
Link to comment
Share on other sites

Yes, that is the same one that I have. R19 and R21 are 15k on it. The board has a pair of 5k on the silk, which is these resistors (I think).
The thread had a cfp2f schematic posted by Kevin on his northwestern site. I don't find this in the current archive.


Sent from my iPhone using Tapatalk

Link to comment
Share on other sites

Another CFP lives.

Started with a single-ended, two-board pilot, proceeded to testing a four-board, balanced setup. Have been listening to it for the past hour or so. Have to settle with 22VDC rails from a Sigma 22 owing to the limit of the transformer I have. Bias set to 150mA. Heatsinks barely get warm. Next is to build yet another GRLV for it with possibly 30VDC rails and higher bias current. 

Even in its somewhat compromised state, the speed, transparency and resolution of this thing is just incredible! 

 

 

CFP amp boards.JPG

CFP amp testing.JPG

2017_10_31-EOS 5D Mark II-5392.JPG

  • Like 8
Link to comment
Share on other sites

Cranked the bias current to 180mA. Broke out my LCD2 as I had a feeling that this amp will be a good match for it. 

People with LCD2 really ought to listen to it driven by this amp at least once just for the experience.

I am still in awe of the speed, clarity, resolution and control of this amp. 

I can understand now why Kevin said in his post that this should be the end-all of dynamic amps. 

This is my favorite of all my dynamic amps and I don't think it's entirely a new toy syndrome at play.

Edited by mwl168
  • Like 3
Link to comment
Share on other sites

Is there any benefit to doing any parts matching on this?
From looking at the schematic, maybe Q9/Q11, Q3/Q8, Q1/Q2/Q4/Q6, Q10/Q7, and Q27/Q28 look like potential candidates?
I know mwl168 didn't do any matching on his with no ill effects.


Sent from my iPhone using Tapatalk

Link to comment
Share on other sites

Without matching any parts other than the LEDs which I had matched for other projects, and no servo engaged, all 4 boards came up from cold start with output offset around 0.06VDC. This is without shorting the input and just rely on the 20K resistor to ground on the board. The offset does not seem to be affected by the adjustment of output bias.

Once I inserted the servo OP, the offset is basically 0 V from cold start. I am using OP07 I have on hand but need to get OPs that can take higher supply rails.

Did I say this amp is good?

Edited by mwl168
Link to comment
Share on other sites

55 minutes ago, kevin gilmore said:

quite a bit more power and heat and bigger power supply etc

Yes. I thought the SuSy Dynalo was powerful until I built the balanced CFP. This thing got enough power to blow my low-efficiency LCD2 to Kingdom Come.

And to Kevin's other point, after some measurement and calculation and verified with Kevin, I just ordered a 100VA trafo for my CFP amp.

Link to comment
Share on other sites

Nice! What’s the max current through the Sigma?

Nevermind, ten seconds of research later I see it doesn’t have a current limiter :)

Are you planning for dual GRLVs when running it at 30V, or is one enough currentwise?

Edited by MLA
Link to comment
Share on other sites

I am running 200mA bias current and 22 VDC rails now using a 200VA trafo with 22v secondaries supplying 4 boards and the Sigma 22 heatsinks are quite warm to the touch at 45 C. The Sigma 22's dropout voltage appears to be about 8-9 VDC which is consistent with the AMB documentation. 

If I had known how good this amp is I would have made it a full-blown, dual-mono, two chassis build with PSU in a separate chassis so I don't compromise its performance. Unfortunately the chassis I picked for it has barely enough space for a single GRLV so that's what I will do. 

Waiting for parts to finish the PSU and I will report back if a single GRLV is adequate to supply 4 CFP2 boards with 200mA bias current. In it's current setting, the current draw on each rail is about 850mA 4 boards combined.  

Here is a photo to show the chassis I am using.

 

 

2017_11_04-EOS 5D Mark II-5394.JPG

  • Like 7
Link to comment
Share on other sites

I did an empirical exercise to ascertain the current draw of the CFP amp, using a 1 ohm resistor in series between the Sigma22 and the amp boards. (My Fluke meters seem to have a hard time measuring resistance less than 1 ohm. Actually I don't even know if I can totally trust the 1 ohm measured resistance is solid.)

Anyway, here it goes; with 200mA bias current, 22VDC rails and trusting the 1 ohm in series resistors are truly 1 ohm;

For 4 boards combined:

cold start: 1.1 A

+ 15 min: 988mA

+ 30 min: 970mA

+ 60 min: 963mA

The + rail and - rail draw the same current.

So, I think a trafo that can supply 1.5A is better suited for a balanced CFP amp.  I ordered a 100VA/28v from Antek to use for my amp.

Edited by mwl168
  • Like 1
Link to comment
Share on other sites

14 hours ago, mwl168 said:

The + rail and - rail draw the same current.

So around 2A in total (pos and neg rail combined)? If I remember correctly, Dynahi pulls about 2.4A in total (600mA per rail per board), so you'll be close to Dynahi levels with your next step

Btw, can GRLV deliver that much current? I thought it maxed out at 1.5A, but maybe that was per rail, not total?

Link to comment
Share on other sites

On 11/4/2017 at 9:28 PM, mwl168 said:

Yes. I thought the SuSy Dynalo was powerful until I built the balanced CFP. This thing got enough power to blow my low-efficiency LCD2 to Kingdom Come.

And to Kevin's other point, after some measurement and calculation and verified with Kevin, I just ordered a 100VA trafo for my CFP amp.

How would you describe the sound?

Link to comment
Share on other sites

46 minutes ago, MLA said:

So around 2A in total (pos and neg rail combined)? If I remember correctly, Dynahi pulls about 2.4A in total (600mA per rail per board), so you'll be close to Dynahi levels with your next step

Btw, can GRLV deliver that much current? I thought it maxed out at 1.5A, but maybe that was per rail, not total?

one might do one/two unregluated PSU's, and feed it to each channel through some pairs of C5200/A1943 being controlled by one GRLV, like Kevin is doing in the Über

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.