diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Solid State (http://www.diyaudio.com/forums/solid-state/)
-   -   Roender's FC-100 prototype and builder's thread (http://www.diyaudio.com/forums/solid-state/202047-roenders-fc-100-prototype-builders-thread.html)

Rudi_Ratlos 7th December 2011 05:30 PM

Roender's FC-100 prototype and builder's thread
 
Gentlemen,

thank you very much for your interest.

In this group-buy my friend METAL and I will offer you PCBs for Mihai Roender's very fabulous RMI FC-100 amplifier.

The design of his amp, technical questions and answers, is covered in this thread:

http://www.diyaudio.com/forums/solid...amplifier.html

METAL and I - after building Carlos Mergulhaos' DX Blame ES/ST and Michael Bittner's SYMASYM -
have been thinking a long time:
"Which AMP will be our next DIY - project?
Which AMP can play distinctly and visibly, can offer the whole stage, where the sound is not stuck to a pair of speakers?"

We raked up the AMPs (the schematics) of this forum and finally settled on Mihai's RMI FC-100.

Mihai was friendly enough to publish the EAGLE files for the AMP's - and frontend PSU's - PCBs.

METAL and I will take Mihai's EAGLE files and do some very slight, nearly invisible modifications to his layout, to his way to layout the tracks,
but we will not change anything concerning the schematics and used components of his AMP.

Mihai's RMI FC-100 AMP PCBs have 2 layers: top and bottom.

This is Mihai's original top layer:

http://www.abload.de/thumb/original_top_layerr8js8.png

http://www.abload.de/img/original_top_layer73k4r.png

This is the top layer of "my" layout - my "EAGLE handwriting":

http://www.abload.de/thumb/top_layerkdkty.png

http://www.abload.de/img/top_layerkdkty.png

The differences may be "peanuts in your eyes".
But I (!) like the tracks being "round and smooth", I do not like small transistors having big, oval pads, I do not like tracks having different width, I do not like ...

I have just begun to rework Mihai's layout: it will take some time, until I am done.

When METAL and I have completed our job, we will offer you:

- the RMI FC-100 AMP PCBs
- the frontend regulated PSU
- if possible I will try and offer you the output NJL0302/NJL0281 transistors

The quality of the PCBs that you can expect, will be like this:

http://www.abload.de/thumb/to-3_symasym1hkhqvwjto.jpg

http://www.abload.de/img/to-3_symasym1hkhqvwjto.jpg

The best quality that you can imagine!

METAL has created a formsheet:

https://docs.google.com/spreadsheet/...xVUQ2clE#gid=0

in which you can sign-in, if you are interested.

I expect the price for a single AMP- or PSU - PCB to be lower than 10.

Maybe (I hope so) some of the HighEnders of this forum (for example Andrew T., whose recommendations I appreciate very much) or somebody,
who already built this outstanding AMP, or even Mihai himself, will give METAL and me some hints, how to even improve the layout / schematics.

Give METAL and me some time to do our part.
I will come back to you soon.

Best regards - Rudi_Ratlos

Rudi_Ratlos 7th December 2011 06:42 PM

Gentlemen,

output NPN- and PNP- transistors cannot be matched!

The NPN-NJL0281D - and PNP-NJL0302D - transistors themselves are usually matched by the fabrication process within 10%!

What is more important than to match the output-transistors is to match the transistors of the input differential amplifier: the 2SK170.

There are 2 types of 2SK170: 2SK170BL and 2SK170GR.

Which type does Mihai recommend?

What about matching the BC550C and BC560C? I will read through Mihai's thread again!

Is anybody of you able to provide some matched pairs of 2SK170, BC550C and BC560C?


Best regards - Rudi

ernesternest 7th December 2011 06:53 PM

Outputtransistors
 
Rudi, Mihai strongly recommended matching of the outputtransistors. Some time ago I bought some of them and saw big differences......

Matching 550 etc is not a big deal, needs just a DMM and a few of these cheapies.

SK170: You can buy them matched for not too much money from a nice guy which is a diyaudio member (forgot his diyname) and has an ebay shop for such things. I've bought some parts from him and was always happy with them.

Regards Ernst

Rudi_Ratlos 7th December 2011 06:56 PM

Ernst,

thank you for this hint.
I will read through Mihai's thread again!

Rudi

anatech 7th December 2011 07:49 PM

Hi Rudi,
I agree with Ernst also, but I'll add more.

Try to match the complementary transistors, like drivers. Match any current mirror transistors and of course the diff pair.

If you use the On Semi transistors, their parts match more closely from both each other and NPN to PNP as well. The very best I ever saw were the MJW0281A and MJW0302A. Each measured very close to a beta of 100, both NPN and PNP. Outstanding. Using matched outputs can reduce THD before distortion by a factor of 10 (yes, 10X) before feedback is applied. This is noted in some On Semi data sheets. Another happy fact is that your required bias current also becomes much lower for minimum distortion measurements, even at 10 KHz and higher.

Now to measure transistor beta, you need a jig, not just a meter. Beta is also temperature sensitive as well as current sensitive. So unless you can control the temperature and test current, you need to build something that will run a bunch of transistors up to a temperature and measure them all at the same current. I do this sometimes when experimenting and to confirm measurements from other methods.

-Chris

metal 8th December 2011 10:19 AM

I think that Rudi will only provide unmatched output pairs. It is difficult for him to buy a large quantity and do the matching, he has not the required tools for this job. I and Rudi will discuss it and come back to you soon.

Rudi_Ratlos 8th December 2011 12:21 PM

I drove to Hamburg this morning, bought a couple of NJL0302 and NJL0281 transistors and measured the hFe of the transistors
with my old Voltcraft DMM.

http://www.abload.de/thumb/joe006xfomz.jpg

http://www.abload.de/img/joe006xfomz.jpg

Here is the result:

http://www.abload.de/thumb/joe008yvrm6.jpg

http://www.abload.de/img/joe008yvrm6.jpg

I am sorry to tell you that I could not measure the hFe of the NJL0281 transistors, since the NPN - hFe - section of my DMM did not work properly.

The result of the PNP's hFe measurement is quite promising.
All values are within 5% deviation of a "standard hFe" of 95.

In my eyes selecting proper pairs of output transistors will not be a NOGO for this DIY project.

Best regards - Rudi_Ratlos

Rudi_Ratlos 8th December 2011 12:36 PM

METAL is right.

I will not offer matched pairs of output-transistors in this group-buy.
Matching - hFe and Vbe - of transistors is very time-consuming!

But I can - if you agree - order several hundreds of them from Mouser or DIGIKEY or ..., from any distributor, who has them on stock,
and ship them together with your PCB order.

Best regards - Rudi_Ratlos

AndrewT 8th December 2011 02:46 PM

Rudi,
I can't expand your pics/links.

You persist in using a remote server.

cocolino 8th December 2011 07:02 PM

Quote:

Originally Posted by Rudi_Ratlos (Post 2811691)
METAL is right.

I will not offer matched pairs of output-transistors in this group-buy.
Matching - hFe and Vbe - of transistors is very time-consuming!

But I can - if you agree - order several hundreds of them from Mouser or DIGIKEY or ..., from any distributor, who has them on stock,
and ship them together with your PCB order.

Best regards - Rudi_Ratlos

How tight have these NJL output transistors to be matched for the Roender RMI FC-100 amp?
Anyone out there with experience with this parts who can advise a ballpark number how many would be needed to get matched sets for let`s say 10 amps (I reckon building more amps need comparatively "less more" devices than building only one pair)?


A question regarding the front-end power-supply PCB:
I believe the V-Reg circuit on the front-end Power-Suppy PCB would be beefy enough to feed several amp front-end circuits (boards).
Is this correct?
If so and to do so, will there be solder-pads for multiple output connectors on the front-end PS-PCB?



Quote:

Originally Posted by AndrewT (Post 2811822)
Rudi,
I can't expand your pics/links.

You persist in using a remote server.

links do work fine for me


All times are GMT. The time now is 09:59 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2