• These commercial threads are for private transactions. diyAudio.com provides these forums for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members, use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

Amplifier Modules and PCBs For Sale

Status
Not open for further replies.
Jump to Latest
D

Deleted member 148505

I actually cannot find decent IRFB4227 class-d modules on aliexpress (at least with my standard), so I designed one myself.
Gain should be around 29dB. (2Vrms in, 56Vrms out at 4Ω/8Ω). Target >= 80dB SINAD at 5W 4Ω. Max voltage +/-85V.
Used switching regulators for housekeeping supplies to avoid excess heat. Onboard low noise opamp buffer since I've added PFFB components for experimentation. Price will be around 80USD.

1697187934940.png
 
Last edited by a moderator:
J

JIM82

Member
Joined 2010
Last edited:
D

Deleted member 148505

JIM82 said:
What value of resitors for an unity gain input stage ?
Click to expand...
Change the resistors from 3.3K to 10K 0805 for unity gain input stage.

Feedback resistors and caps are under the opamp socket so they are difficult to replace. (You can change to 4.7K input resistor and 4.7K || 680p feedback without stability problems. But the current config is the most compatible with different opamps for a socketed configuration, especially if you use the boutique / discrete opamps)

It's easier to replace the opamp socket if you destroy / split the socket in 4pcs using side cutting pliers and desolder each section (2pins) simultaneously.


JIM82 said:
If Bypass, how to do it and what input impedance value?
Click to expand...
I suggest not to do a bypass, because at this level of performance, an onboard buffer is required, wiring the buffer outside the feedback loop may actually worsen the SINAD due to long wiring.
 

Attachments

  • 1697432346536.png
    1697432346536.png
    15.2 KB · Views: 75
D

Deleted member 148505

Changing r_in (RP1...RP4) to 6.99K will make the performance of PFFB to 105dB to 106dB SINAD at 1kHz 5W 4R.
Although you have to increase the opamp supply as well (requires changing another 4pcs 0603 SMT on the supply section)

But... the present 5.1K values do sound the same. (6.99K r_in = 106dB SINAD vs 5.1K r_in = 103dB SINAD, just 3dB difference)
So, you might understand my design decisions here for this module... It's a balancing tradeoffs act.

****

Deepening the PFFB is just changing the ratio of r_fb and r_in, maintaining stability is the challenge.

1697439546717.png
 
Last edited by a moderator:
J

JIM82

Member
Joined 2010
JLesterP said:
I suggest not to do a bypass, because at this level of performance, an onboard buffer is required, wiring the buffer outside the feedback loop may actually worsen the SINAD due to long wiring.
Click to expand...
Thanks for all the answers.

I was thinking of possibly directly attacking the power stage with PFFB. I use a motu 24ao symmetrical sound card which already has its own output AOP compatible with a 1kOhm load.

I did this on a basic tpa3255 based amp (but without PFFB).

You think it's better to keep the buffer ?
 
D

Deleted member 148505

JIM82 said:
I was thinking of possibly directly attacking the power stage with PFFB. I use a motu 24ao symmetrical sound card which already has its own output AOP compatible with a 1kOhm load.
Click to expand...
This would work in a non-PFFB board. In a PFFB setup, the moment you disconnect the MOTU, the module will go into protection mode.
There are onboard jumpers under the PCB to disable the PFFB. Some D100 users prefer the D100 with the PFFB disabled.

JIM82 said:
You think it's better to keep the buffer ?
Click to expand...
If you have followed this thread, I've created a gain buffer module for this purpose.
Although the last time I tried it is that the r_in is on the gain buffer module, hence the long wiring affected the SINAD performance.
I will try again with the r_in located on the Sylph-D100 board. Though it will surely lose 1dB or more SINAD if the opamp buffer is away on the PFFB network. So still the recommendation is to retain the onboard buffer.
 
D

Deleted member 148505

JIM82 said:
I don't need any additional external gain
Click to expand...
If the opamp section is unity gain, the total gain of D100 will only be 2.6dB (1.35x voltage ratio). So, you need a preamp that is capable of outputting 15Vrms to drive the amp into clipping.

JIM82 said:
So, I'm going to modify the gain with two 4.7 kOhm resistors (unity gain buffer).
Click to expand...
Nice, with 4.7kOhm resistors I think the buffer won't be a performance bottleneck especially if you use OPA1656.
 
D

Deleted member 148505

JIM82 said:
Hi,

Here is my assembly with x3 D100 modules.
https://www.homecinema-fr.com/forum...ymetrique-faible-bruit-termine-t30127894.html

How to reduce the gain by 6dB?
Click to expand...
1697616391606.png

Speechless on your build quality.

Regarding DC offset that you've encountered, the polarity of the capacitors has been overlooked on assembly. On non-PFFB mode, the input coupling cap's negative side is on the opamp output (ground referenced due to opamp split supply then the TPA325x input is referenced to AVDD/2).

On PFFB mode, the polarity changes (feedback referenced to PVDD/2 on each out_x).

So the default input coupling cap's polarity assembly should have been on default PFFB mode, we will do it going forward.
 
Last edited by a moderator:
D

Deleted member 148505

JLesterP said:
I actually cannot find decent IRFB4227 class-d modules on aliexpress (at least with my standard), so I designed one myself.
Gain should be around 29dB. (2Vrms in, 56Vrms out at 4Ω/8Ω). Target >= 80dB SINAD at 5W 4Ω. Max voltage +/-85V.
Used switching regulators for housekeeping supplies to avoid excess heat. Onboard low noise opamp buffer since I've added PFFB components for experimentation. Price will be around 80USD.

View attachment 1223032
Click to expand...
HV DC-DC converter got out of stock. 🤦‍♂️
Replacing with LM5164 converter IC.
Used 2W isolated DC-DC 12V to +/-12V for the analog supplies.
800W Peak into 4 ohms

1697963028881.png
 
D

Deleted member 148505

SylphAudio Preamp has taken its shape. :)
  • No input coupling caps on the signal path.
  • 5x OPA1612, 10dB gain, can be reduced by 10K volume potentiometer.
  • Relay switching for SE/Diff input selection.
  • Simultaneous SE and Diff output (had to look at my DO100 DAC on how it's being implemented lol).
  • 11.5Vrms max for SE, 23Vrms for Diff output
  • Single 24V to 60V DC supply input.
  • Expected THD+N: -120dB at 1kHz, 0dB gain.
1698144018556.png
 
  • Like
Reactions: 5 users
D

Deleted member 148505

JLesterP said:
10dB gain, can be reduced by 10K volume potentiometer.
Click to expand...
0dB gain is the default. I added empty 0805 SMT pads (circled in yellow) so that the gain can be easily increased.
Frontend buffers are non-inverting for high impedance input.
FR flat at 20Hz to 20kHz, -3dB is at 96kHz,

Will test the prototype in a couple of weeks (Victor's osc + Cosmos APU + MOTU UL MK5 ADC)

1698506226304.png
 
  • Like
Reactions: 1 user
Status
Not open for further replies.
Top Bottom