MPP

Joachim Gerhard · 2012-10-05 11:53 am

Alfred, i have read your instructions. I like it a lot.
I will read the offset part again and may do some additions i talked about.
Also on the matching of the input devices.

MiiB · 2012-10-05 11:55 am

The buffer off set can be adjusted with the resistor. with the SK170 and 5 ohm it is app 60-70 mV.. with 11 ohms it practically Nill... (8,5 mA through the Jfet)

RCruz · 2012-10-05 8:37 pm

MiiB said:
The buffer off set can be adjusted with the resistor. with the SK170 and 5 ohm it is app 60-70 mV.. with 11 ohms it practically Nill... (8,5 mA through the Jfet)

Thank you Miib... precious info

Algar_emi · 2012-10-05 10:36 pm

Hi. Since the official R3 PCB was no longer available, I did my own

I also did a pcb for the pre-reg according to the included schematic. I should have them built by the end of the day.
The latest version of the parts list match these PCB.

Algar_emi · 2012-10-05 10:45 pm

Question about the BC327/337 HFE matching. Is it necessary to match all the transistor one one channel pcb, or just the Gain Cell ones (Q2,25,26,29 and Q1,30,31,32). And finally, I guest it is necessary to match both channel together as well to get the same gain?

Thanks.

Stefanoo · 2012-10-05 10:55 pm

yep the input devices.
also it would be good to match the current mirror (not the one on the servo) for best performance.

Bksabath · 2012-10-05 11:09 pm

Algar_emi said:
Question about the BC327/337 HFE matching. Is it necessary to match all the transistor one one channel pcb, or just the Gain Cell ones (Q2,25,26,29 and Q1,30,31,32). And finally, I guest it is necessary to match both channel together as well to get the same gain?

Thanks.

Hesener Put this mach beter way in the guide.
http://www.diyaudio.com/forums/analogue-source/218625-paradise-builders.html#post3189930

Se post 13

Match 10% or better pick closest matched and fit in groups as in circuit
Imput Current mirrors and soon

Joachim Gerhard · 2012-10-05 11:17 pm

Cool, we have a PCB for the prereg.

Algar_emi · 2012-10-05 11:23 pm

Here the completed PCB. Do we need a GB for the Pre-Reg PCB

Joachim Gerhard · 2012-10-06 12:40 am

Well, i was designing a prereg too that could lead to a group buy. That is obsolete now.
I would like to try your PCB and add chokes and output caps eventually.

Holger Barske · 2012-10-06 1:49 am

Algar_emi said:
Hi. Since the official R3 PCB was no longer available, I did my own

This looks like transformer - choke - rectifier - cap,

Guys. This isn't what we talked about. We want transformer - rectifier - cap - choke - cap.

Algar_emi · 2012-10-06 3:09 am

Joachim you have mail

Joachim Gerhard · 2012-10-06 3:22 am

Holger, Algar designed the pre regulator for the pre regulator so to say. I posted in 7490 that i will add chokes and caps.

Joachim Gerhard · 2012-10-06 3:25 am

I think the use of the common mode chokes is correct in this design but for ripple rejection we need the bigger sized stuff from Hammond or Lundahl. Those can be put behind Algars
PCB or could be integrated in a bigger PCB.

Bksabath · 2012-10-06 9:09 am

Holger Barske said:
This looks like transformer - choke - rectifier - cap,

Guys. This isn't what we talked about. We want transformer - rectifier - cap - choke - cap.

Wasit not traffo - common mode coke - rectifier - cap - coke - cap ?

kamis · 2012-10-06 9:16 am

Bksabath said:
Wasit not traffo - common mode coke - rectifier - cap - coke - cap ?

Why not use John Curl topology? He attached it in another thread.

Joachim Gerhard · 2012-10-06 12:32 pm

The common mode chokes suppress only HF/RF. For ripple rejection and smoothing we need big coils that work in the audio range.

MiiB · 2012-10-06 1:30 pm

To me theres no need to target ripple rejection.. HF is the primary.....

The Paradise has extreme ripple rejection.... all is effective shielded by high impedance current-sources/mirrors. poison is the Hf, that can bleed through stray capacitance.
Curls schematic seems rather fine to me or at least a good start. ut no need for the LM's as the Shunt is already onboard..:9

Joachim Gerhard · 2012-10-06 1:35 pm

I agree, there is no need for extreme pre regulation but it may be fun. Some may hear a difference, some not, one never knows.

Bksabath · 2012-10-07 12:40 am

kamis said:
Why not use John Curl topology? He attached it in another thread.

Sorry that to me is straight LM317 and a common mode coke classic straight out of book

Why do we have to put a name on it?

In other words....

Besides one may need to step back and see all the work FDW did on the suply before mentioning LM317

MPP

Member

Member

Member

Member

Attachments

Member

Member

Banned

Member

Member

Attachments

Member

Member

Member

Member

Member

Banned

Member

Attachments

Member

Member

Member

Banned