JoaMat Posted December 3, 2025 Report Posted December 3, 2025 Perhaps you are right, but that’s nothing I can verify. It’s correct that pos input is dancing. Now I think that the neg input is a perfect dancing partner which results in a working servo. With the servo the amplifier behaves similar to the ones without servo except that the bias doesn’t drift. And cross talk? Not a problem - I’m married and used to cross talk.
jokerman777 Posted December 3, 2025 Report Posted December 3, 2025 (edited) my bad could've just read to pages ago more carefully where the discussion on the matter already happened... I did a quick sim on Joamat's "new bias servo", 5Vpp on one side into 32ohm load, ofc not the same amp circuit, not opa445, and not assuming sim == real life behavior, but probably enough to convince me it's not gonna be free lunch. Below is before and after (also tried 300ohm load and difference is much smaller though). Now there is also KG's optocoupler version but between adding another 8 x opa445 and more and just doing thermal coupling right in the first place, well... 😃 10 hours ago, kevin gilmore said: i looked at the ltspice file for a bit. best guess is that there is something wrong with the jfet model that causes issues. replace the fets with bipolar and see if it works right. the reason all the darlington transistors are on the heatsink is so they can thermally track. effects thd. another thing that really effects the thd is the power supply rails. anything under +/-20v is too low and causes much higher distortion. +/-30v is the sweet spot. and once you do that, you end up with an amplifier that is a lot more than 5 watts. in your second schematic, you have taken out the low impedance load on the output of the current mirrors. lots more voltage gain resulting in required high amounts of feedback. definitely not the original design goals. if you want lots of open loop gain, probably best to pick a different design. as far as the bias servo, would have to see thd with and without it. And many thanks for the advice and detailed explanation! Yeah I'm aware of the generous amount of feedback used here, much like the original CFA2 right? where CFA3 is changed to open loop design with optional feedback config. I didn't "choose" to go with this way on purpose and haven't built/listened to enough stuff to form a particular preference on the camps between no feedback/some feedback/a lot of feedback (which kinda has a bad reputation). I'm pretty much just randomly exploring options now, for possibly a smaller scale project than CFA3 for my need. Might still jump back on doing the original if I got enough messing with spice Edited December 3, 2025 by jokerman777
jokerman777 Posted December 7, 2025 Report Posted December 7, 2025 Too far down the road of downgrading? loading second stage down with resistors, loop gain went from 65dB to 25dB, thd will increase by more than an order of magnitude. No idea how much would be the sweet spot subjectively but can possibly be further tweaked. FET output, linearity not as good as triple BJT, just bc still try to cut corner on size... A quick search suggest lateral FET pretty much went extinct, for vertical ones should probably change vbe multiplier to something TO-126 and also put on heatsink? 1
MASantos Posted February 4 Report Posted February 4 (edited) On 12/2/2025 at 12:35 PM, JoaMat said: I’ve a smd version of CFA3. Only output transistors on heat sink. Bias is servo controlled. Works all right. Output stage with servo looks like this… Reviving this thread a bit. DIY itch is getting stronger. I would love to try and make a smd cfa3 with digital attenuator and protection circuit on 3 board layout. One board per heatink with the amplification circuit, and a central board with GRLV, input and output attenuator and protection circuit. Sort of what Justin did with his cfa3. Do you have a layout for your smd cfa3, and would you be willing to share it? I have played a bit with schematics editing and board layout, my idea would be to add the individual building blocks to the main board and try to make it fit a nice case such as the mini dissipante from modushop. Edited February 4 by MASantos
JoaMat Posted February 5 Report Posted February 5 On 2/4/2026 at 12:27 PM, MASantos said: Do you have a layout for your smd cfa3, and would you be willing to share it? Attached ZIP-file with gerbers, shematics and some 3D views. CFA3smd.zip I’ve used both gerbers when I milled the boards (kitchen made) and that is about three years ago. Board size 160mm x 74mm to fit on Modushop 200mm x 80mm heat sinks. Take a look at the files and see if/how the boards could be a part of your project. I think one should make a few changes to attached gerbers before considering sending them to a fab house. 1
MASantos Posted February 9 Report Posted February 9 Thanks! I'll take a look at it. Regarding the smd transistors and input jfets, what was your reasoning for these choices?
JoaMat Posted February 10 Report Posted February 10 (edited) No particular reason behind the jfets, aside I like the to-71 package and that it’s fairly easy to build the layout around them SMD transistors because they are small. I had to decrease current at some locations to reduce heat. I see this amplifier merely as an experiment. I wanted to test if it was possible to make a small board. The bias servo is probably needed while moving from heatsinked MJExxx to small smd transistors. I believe the bias servo previously posted by Kevin, with dual op amps, is preferred. So, another experimental board… Edited February 15 by JoaMat 1
bbest Posted February 21 Report Posted February 21 Hello all! There is interesting amplifier Headtrip 3 level 2 https://www.wellsaudio.com/headtrip . Who heard it, says that it is better then CFA3. How you think - does it make sence to make clone of this amplifier or may be it is already exists?
kevin gilmore Posted February 22 Author Report Posted February 22 (edited) you mean this grossly overpriced piece of antique design. unbalanced and lots of feedback. clone boards were available somewhere at diy audio. chassis filled with many audiofool components. like bybee's and fancy caps. if 50 watts into 8 ohms is what you really want then build my pioneer a09 clone. lots better in many ways. i could publish pictures of the zahl boards, or the mass kobo. both of which are also grossly overpriced pieces of garbage. Edited February 22 by kevin gilmore 2
bbest Posted February 22 Report Posted February 22 (edited) Kevin thank you for your opinion 👍 What is the latest boards or schema of CFA3, cfp3largeT_1.35? Edited February 22 by bbest
kevin gilmore Posted February 22 Author Report Posted February 22 (edited) many different versions. a couple of different sizes. i don't pay attention to such things. edit: here is list of boards by date 04/11/2015 05:43 AM 103,192 cfp2c.zip 04/11/2015 05:44 AM 104,260 cfp2cm.zip 05/29/2015 08:20 AM 462,573 cfp3rs.zip 08/09/2015 02:25 PM 156,813 cfp2dmt.zip 09/06/2015 07:45 AM 158,031 cfp2emt.zip 09/20/2015 04:37 PM 402,743 cfp3rss.zip 02/07/2016 11:11 AM 159,746 cfp2fmt.zip 02/28/2016 06:04 PM 159,406 cfp2gmt.zip 11/06/2017 07:26 PM 55,189 cfp2gmt - CADCAM.ZIP 11/30/2017 09:07 AM 66,091 cfp2hmt.zip 05/24/2020 11:29 AM 131,243 cfp3rssfixedss - CADCAM.ZIP 10/31/2021 07:41 AM 141,717 cfp3rssfixedsssw - CADCAM.ZIP 02/02/2022 06:19 AM 126,329 cfp3smt2 - CADCAM.ZIP 02/07/2022 06:54 AM 78,922 cfp3smt2splitpre - CADCAM.ZIP 02/07/2022 06:59 AM 63,188 cfp3smt2splitamp - CADCAM.ZIP 01/11/2023 08:18 AM 43,551 cfp3largeT.ZIP 08/05/2024 01:40 PM 125,671 cfp3smt2splitampdual2024-2-mosfetdualps - CADCAM.ZIP 12/28/2014 11:06 AM 77,647 cfa2.PDF 12/01/2017 08:48 AM 55,404 cfa2cmirror.PDF 05/24/2020 06:47 AM 53,387 cfa3production.PDF 07/23/2020 04:34 AM 69,492 cfa3productionss.PDF Edited February 22 by kevin gilmore 1
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in
Sign In Now