Jump to content

and now for something completely different part 3


kevin gilmore
 Share

Recommended Posts

Hi Pars,

the latest revision v1.2 THD boards say BC546B and BC556B on them.

Regarding to the ZF and feedback resistor change it is for both around 2x, so 50k instead of 25k and 4.2k instead or 1.9k, basically what Paradoxper used. Important is that both modes match in volume - you may scope with 1kHz and match the 4.2k so that you get same volume in ZF-mode as in feedbak mode. This being said, I am still using 25k for feedback and 1.9k for ZF and only using the amp in ZF-mode.

As well as using the B-version sands from onsemi.

Link to comment
Share on other sites

1 hour ago, audiostar said:

Hi Pars,

the latest revision v1.2 THD boards say BC546B and BC556B on them.

Regarding to the ZF and feedback resistor change it is for both around 2x, so 50k instead of 25k and 4.2k instead or 1.9k, basically what Paradoxper used. Important is that both modes match in volume - you may scope with 1kHz and match the 4.2k so that you get same volume in ZF-mode as in feedbak mode. This being said, I am still using 25k for feedback and 1.9k for ZF and only using the amp in ZF-mode.

As well as using the B-version sands from onsemi.

As I recall, the one built by paradoxper that I fixed for tkam had a bit of difference in volume (not measured) between ZF and SS modes. Close, but not exact. It was close enough that it wouldn't have bothered me. I'll order B on-semi sand.

  • Like 1
Link to comment
Share on other sites

6 hours ago, Pars said:

As I recall, the one built by paradoxper that I fixed for tkam had a bit of difference in volume (not measured) between ZF and SS modes. Close, but not exact. It was close enough that it wouldn't have bothered me. I'll order B on-semi sand.

This is why I use the original values from Kevin. 

Link to comment
Share on other sites

Hi, Guys, 

I'm working on my CFA2 project.

I've got a matching MJF15030/15031 question : How nice is the matching?

I had a Peak DCA-75 to match these two BJTs, the result is horrible. The hfe values of One bjt are most from 280-290, the other are from 70 to 120. It is seem not nice enough, right? 

So, how nice is the close enough matching of these two BJT?  May I need to buy more bjt to get nice matching?

 

Link to comment
Share on other sites

For many device series types, the N and P are not going to match well. If you can get more devices at a reasonable price, that is always best. Where are you purchasing these devices from? A known supplier? I guess my point is, are they genuine devices? A lot of fakes out there (ebay, aliexpress, dalbani, etc.)

Link to comment
Share on other sites

6 hours ago, Pars said:

For many device series types, the N and P are not going to match well. If you can get more devices at a reasonable price, that is always best. Where are you purchasing these devices from? A known supplier? I guess my point is, are they genuine devices? A lot of fakes out there (ebay, aliexpress, dalbani, etc.)

I got this DCA75 Pro from digikey. I think it can not be fake one. So, How close the two hfe values of these 15030/15031 are? Can you give me some values as a reference?

Link to comment
Share on other sites

  • 4 weeks later...
  • 2 weeks later...
On 10/27/2021 at 4:23 AM, kevin gilmore said:

bc546/bc556 starting to get very hard to get even though onsemi says they are going to make more, so this which needs massive amounts of checking.

 

cpa3smt.jpg

Please post the gerber file

Many thanks

Link to comment
Share on other sites

Late to the party, but here goes. Lured a nice small case off of MLA for a cheap penny, wanted a compact CFA (SE).

Took some time for my slow mind to figure out how to best use the space in three dimensions, but I found a solution that works fine so far, if you're planning something similar. No interfering leads or unwanted nearness of critical components, and bias can be set without problem. In reality, I'm surprised it turned out quite airy and spacious, considering the limitations.

IMG_1227.thumb.jpg.b362fceade51a24ebba4235686dae337.jpg

IMG_1212.thumb.jpg.f99e92043f0277e0581ba647903fd4e4.jpg

Note the Müller Rhombus transformer device ("Konzept Raute"). This romboid plate interacts wit the circular toroid shape and the electromagnetic waves are forced through the four holes in an endless loop in this electronically confined space, and thus traps all hum in an existant/nonexistant void. This amp is noise and hum free.

I'm particulary happy with the volume control, a 24-position Swiss Elma switch that's been lying in a drawer for 35 years. 15K (why not?), shunt coupled and making as little contribution to the signal as possible. I think this plays a part in the clean, revealing sound of the CFA design.

IMG_1214.thumb.jpg.2965d147d283393f083a11e3afd89966.jpg

IMG_1219.thumb.jpg.5b0fe618f21ec27da33194af80da7249.jpg

The odd resistor out is in the position I mostly listen to, a hand selected Syldavian military plutonium component (0,01%) from 1953, made in a numbered series of three of each value (I own the third one as well) and these days sold on ebay in Hong Kong for not less than $1200, if available at all (only one for sale during the last nine years).

Set the bias at first to 150mA, it sank to 135 after some time, which was to be expected. Decided too raise it (why not?) to 200mA. Put the lid on and after three hours it was stable at 175mA after a last fine-tuning. Gets warm, but not too hot.

Like all successful, completed builds it sounds wonderful , but if that lasts only time will tell. If so, I may be tempted to follow up with a CFA3.

IMG_1214.thumb.jpg.2965d147d283393f083a11e3afd89966.jpg

IMG_1208.thumb.jpg.5739712d863f1fb596447893d0b94e15.jpg

Edited by mdr30
  • Like 8
Link to comment
Share on other sites

so i have got a number of contacts, most with felista built amps, definitely with fake toshiba transistors that blow up.

Now another person with toshiba parts with the amp oscillating. Other changes were made, so who knows, and no idea if the transistors are real or fakes. being discontinued more than 10 years ago, everything on ebay has to be fake.

so word of warning to people building them and trying to use toshiba (fake) parts. no way to know if they are fake unless you use a transistor tester. biggest tell is that they test barely at 25-30 volts instead of 120

  • Thanks 1
Link to comment
Share on other sites

so split versions of the cfa3 board.

stack the boards together and put on the heatsink (minimum heatsink vertical dimension 83mm)

stack the boards together, use angle bracket mount horizontally for low profile chassis

etc

outputbuffer.jpg

preamp.jpg

 

cfp3smt2splitamp - CADCAM.ZIP cfp3smt2splitpre - CADCAM.ZIP

Edited by kevin gilmore
  • Like 3
  • Thanks 1
Link to comment
Share on other sites

11 hours ago, kevin gilmore said:

so split versions of the cfa3 board.

stack the boards together and put on the heatsink (minimum heatsink vertical dimension 83mm)

stack the boards together, use angle bracket mount horizontally for low profile chassis

Wow, thanks Kevin - thats pretty!

I wish I have waited for this but I already finished the regular cfa3 and casing it right now. 

Add a board for input selection in the back, the protector3 and a board holding the pot and phones connectors in the front and this makes up for an awesome amp. 

Could not resist, ordered boards...

Edited by audiostar
  • Like 1
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.

 Share

×
×
  • Create New...

Important Information

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