Jump to content

Recommended Posts

Posted

Great! Having some boards made; have some current drive friendly speakers as well a 3 way active system that both will be great matches I think. Thanks!!!

Posted (edited)

A question on the CFA3; I just built one using 50k feedback resistors for SS and 4.2k ZF resistors, also excluding the C9/C10 caps as per mirkos final edit earlier in this thread.

image.thumb.png.1da87becad5a6e50fd7da7ff06d5158a.png

Builder error: was using the wrong LEDs (too much forward current I believe). Swapped those and now it's all well balanced with 10k at R1.

Edited by MLA
  • 1 month later...
Posted (edited)

Just finish soldering my split boards.

There are some problems happened dual to my bad SMDs solder technique. 
IF non-wetting happens / flux not clean-up between pins,
no bias from 1R resistors and all output transistors not heating up;
and very high DC offset can be measured from output to GND (~10-27VDC).
I resoldered all the SMDs to fix this.

After that I powerup my spirtboards, an 1R resistor heats up quickly and smoke comes out.
One of the output transistors(Q17) shorted, replace it and no smoke again.

Finally I test the split boards with a pair of speakers and HD600,
both speakers and headphone sounds very well.

But when I test the boards with my DAC and preamp which have ~100ohms and ~1ohms output impedance respectively,
a 10k pot can only adjust volume a bit , only a little volume change from 10kohms to 0ohm;
sound still comes out at 10kohms.
A 50k pot could fit but some previous works in this post use 10k pot.
Did I got some mistake with my work?

Edit: A 10k pot can adjust volume from no to full with proper wiring.

CFA3@0.3x.jpg

Edited by EL_Ken
Posted

Haven't tested these particular boards but my CFA3 runs fine and fully adjustable from no to full volume with a 10K pot, so I would say something's off...

Posted

If you are using THAT340s for the input devices (BJT), impedance issues may be causing a problem. The Pass SuperSymmetry results in a low input impedance, which can cause problems for certain devices. One way around this is to use JFETs on the input instead, as these greatly raise the input impedance. Something like the Toshiba 2SK170/2SJ79 or the equivalent Linear Technologies singles or duals. I'm not sure what device package these boards are set up for. If it is surface mount, that would make this more difficult.

Posted (edited)

you may be the first person to test these boards. but if you have music, highly likely that everything works right.

if the dac has a significant dc offset, that may cause trouble.

maybe wiring of the pot is wrong?

edit:  10k pot highly recommended with the that340 chips. 50k is not going to perform well.

Edited by kevin gilmore
Posted

I have my amp with split boards finished but undecided between a 4CP-2511 and the digital attenuator, so it still sits on a shelf without a pot. While bench testing the boards prior casing it up, I didn't notice anything weird though. I am using the Toshiba 2SK389-BL & 2SJ109-BL JFETs on DIP adapters. I'll order a 4CP-2511 and give it a final bump.

 

Posted

Sorry for making every one confuses,
After rewiring the pot, problems gone.
It causes by incorrect grounding.
A 10k pot can adjust volume from no to full with proper wiring.

Now using that340, will try swapping input transistors later.

Had music continuously with DAC-preamp-splitboards for 10+hours with no problems,
Testing with DAC-10kpot-splitboards setup, works fine too.

Thanks for the help.

Posted

All GND pins and pin1 of XLR cables are shorted to each others.


Screened cables - POT - silicone wire
Some silicone wire from local store, no much interference issues.
Only a bit white noise from headphones when pot turns over 3 o'clock.

pot (2).jpg

Posted

I see.

I tie pin1 to to the case immediately via the XLR connector (and not to ground) and use the metal case as shield. I use twisted wires for the signal inside the case. Circuit ground connects on a single place via a circuit breaker (at least 10ohm resistor) to the chassis. Gives me no hum whatsoever. 

 

Posted

So, if all goes well, I'm picking up a pair of Quad 2805s next weekend, which means I have to build an appropriate amp for them :)

The Uberamp I have boards for and that's probably what I'll build, but I'm wondering if a cfp3Large (with the Sanken output boards) would be a good fit also? Or is that a less good match?  

Posted

Output transistors are MJE15032G/15033G, extra insulation is needed.
Size of the heatsink is 200x62x40mm, ~50°C with 120ma bias.
It could handle more heat but I need larger heatsink for GRLV.
The nichicon kz cap (Ø:12mm) is too big so one has to lying down.

Edit: one more pic added

230309-cfa_P1.jpg

230309-cfa_P2.jpg

230309-cfa_P3.jpg

230309-cfa_P4.jpg

230309-cfa_P5.jpg

  • Like 7
  • 4 weeks later...
Posted

Just finished a v1.21 board SuSy version. Volume controll is in the DAC, left that out, just a power HPamp. Running on 200mA, only balanced in- and output. THAT340 (could not find anything else in Europe), Nichicon Muse caps, SPC cables, dual v0.44 GRLV with 100VA/24V. Modushop 2U Dissipante 300 and 2U SlimLine 280.

Thanks go to Mirko for the remote support and to Dr. Gilmore for the incredible design!!

CFA3 outside.jpg

CFA3 amp inside.jpg

  • Like 6

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

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