Roender's FC-100 prototype and builder's thread - Page 3 - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 10th December 2011, 02:10 PM   #21
zdr is offline zdr  Belgium
diyAudio Member
 
zdr's Avatar
 
Join Date: Nov 2005
Location: Brussels
Here's something I devised for myself, since I suck at home-brew double sided boards: single sided variant, large MKT included, and still shorter pcb overall.
Attached Images
File Type: png fc100-ss4.png (108.1 KB, 930 views)
  Reply With Quote
Old 10th December 2011, 04:04 PM   #22
diyAudio Member
 
Rudi_Ratlos's Avatar
 
Join Date: Jun 2009
Location: Hamburg / Germany
ZDR: your layout looks good - and: you are much faster than I am!

I have two suggestions:

1) I think you are using caps with a diameter of max. 30 mm on the right side of the PCB.
Would you be able to change the caps to the EB35D - Eagle package?
Do we then need 4 capacitors or aren't 2 enough?

2) You talk about a "large MKT" capacitor that you have included. Can we settle on a specific capacitor that I will be able
to offer in this group-buy as well?
What about a Vishay MKT1813 - 1F / 100V? Its diameter is 8.5mm, its length is 19mm, so the pads shoud have a distance of about 25mm.

This is my current progress:

Click the image to open in full size.

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

I like when everything looks symmetrical.

Comments are welcome.

Best regards - Rudi

Last edited by Rudi_Ratlos; 10th December 2011 at 04:06 PM.
  Reply With Quote
Old 11th December 2011, 12:36 AM   #23
Luke is offline Luke  New Zealand
diyAudio Member
 
Join Date: Sep 2001
Location: Wellington NZ
Send a message via AIM to Luke
Hi Rudi, connectors for the power and input and maybe output would be nice. Im not sure about the need for 4 caps, 2 X 18KuF plus would do it I would have thought?
__________________
If you give a man a fish he will eat for a day. But if you teach a man to fish he will buy an ugly hat. And if you talk about fish to a starving man then you are a consultant. Dilbert
  Reply With Quote
Old 11th December 2011, 03:33 AM   #24
anatech is offline anatech  Canada
diyAudio Moderator
 
anatech's Avatar
 
Join Date: Jun 2004
Location: Georgetown, On
Hi Christoph, Rudi,
If you use the newer On-Semi output transistors, matching is not a problem. They arrive much tighter than matched pairs used to be bought from amplifier manufacturers. No worries there.

I also agree that one or a few people attempting to match sets for a group buy using the older parts would be totally exhausted. A far ir charge for doing this would have everyone making jigs to do it themselves.

To match the smaller (TO-220 and smaller) packages, there is a jig I developed a couple years ago that does the job amazingly well. It's cheap, simple and uses the basic characteristics of transistors to do it's thing. You supply the meter(s). Walt Jung looked it over, and aside from different component values he would rather use gave it his blessing. Personally, I decided to leave it as it was since it worked really very well. Bonus time, you can also use it to match J-Fet transistors. I'll post some of the info next. If anyone wishes to design a PCB for it, I'll be happy to help with some of the design details. Keep in mind that the NPN section was conceived, designed and constructed in one hour as I had an urgent need for it.

-Chris
__________________
"Just because you can, doesn't mean you should" my Wife
  Reply With Quote
Old 11th December 2011, 03:52 AM   #25
anatech is offline anatech  Canada
diyAudio Moderator
 
anatech's Avatar
 
Join Date: Jun 2004
Location: Georgetown, On
Default Match'in xistors

Okay, here is some info on it. I may need to add details tomorrow, but I'm tired tonight. Another thing you can do with this jig in addition to simply testing the balance on a pair of transistors is that you can also measure the beta at the test current divided by two if balanced. If you only insert one transistor, you no longer have a compensation for temperature, but at least you are still using a stable test current.

For some of you out there who prefer to match on Vbe only, no sweat, just measure it. However, this jig takes both Vbe and beta into account at the same time.

The most critical thing about this are the resistors. You need close tolerance resistors in matched pairs, but they are cheap. I recently bought some 0.1% resistors for a rebuild. The rest of the circuit is open for modification, nothing sacred here. It was a quick into service deal. In fact, I have worn out the sockets again. It gets a lot of use. You know how close the matches are? Emitter degeneration resistors in a diff pair also need to be matched or they will throw off the match. That is silly close, and it surprised me the first time I ran into this issue.

What would I like to change? The tail current generators.
What would I like to add as a feature? Current mirrored PNP section to NPN current

What would be really cool? Making a setup to match NPN and PNP complimentary pair matching.

One further note. This device will merrily match surface mount components. My adapter is a pain to use. The thing is, this can have adapters to plug in enabling vastly different packages to be assessed. Obviously you can not cut a matched pair package apart, but it would be nice to prove the part was still matched or not. Then there are really cool complimentary pair SM parts, plus packages that have common elements. An adapter board could bridge the NPN and PNP sections and make testing the match on anything.

Here we go...

-Chris
Attached Images
File Type: jpg N matcher (Medium).jpg (49.5 KB, 868 views)
File Type: jpg P matcher (Medium).jpg (46.6 KB, 813 views)
File Type: jpg Matcher bottom view (Medium).JPG (114.4 KB, 780 views)
File Type: jpg Jig Top (Medium).JPG (119.2 KB, 248 views)
File Type: jpg DSCN8190.JPG (968.1 KB, 239 views)
__________________
"Just because you can, doesn't mean you should" my Wife
  Reply With Quote
Old 11th December 2011, 04:15 AM   #26
anatech is offline anatech  Canada
diyAudio Moderator
 
anatech's Avatar
 
Join Date: Jun 2004
Location: Georgetown, On
Default Using the thing

Insert the transistors with the cases touching each other. I made some foam covers that slip over the pair to stabilize the temperature. This creates a common thermal environment for both transistors. They need to be at the same temperature, which only occurs with a match condition. Everything comes together when the transistors are balanced.

Give the parts time to settle down. You can pre-screen transistors by beta groups with the beta function on a DMM. Do your best to keep the temperature constant. Be prepared to see how far apart they really are once you begin balancing them in the jig. Also try to match one beta group up and down - you will find matches.

The sockets have four contacts per part. This way they will accommodate JEDEC, JIS and Pro-Electron pin-outs without having to mangle the leads. These sockets also carry the adapter PCBs, but I'm thinking that separate sockets should be used to prolong the life of the contacts. I used those snap-off terminal sockets you buy in strips. The current set terminals are headers as used on computer motherboards. That's also where the shorting terminals you can see in the pictures came from.

This jig is simply a differential (long tailed) pair with a current source for tail current. Since that is how the transistors you match are most commonly used, you are simply recreating the conditions you will use them in. The tail current is adjustable with the jumpers. The currents will be additive - sort of. It might be more accurate to say the current varies as the setting. That means you can check the balance at different current levels and change those on the fly. Just set the tail current close to what you intend to use in circuit and you will have happy, tight matched pairs.

When installing those freshly minted matched pairs, install them so the encapsulated parts touch. Add a touch of thermal compound between them and use heat shrink tubing to isolate them from the big, bad world. It also holds them together in contact with each other. For real low noise applications, there are brass clips that have a grounding lead that will hold the parts together. Adding silicon around the leads once they are in the PCB and foam (or whatever - plastic cap even) over the parts will eliminate air currents, and also keep the PCB sealed against contamination. If that is important, seal the foil side as well. You can also use copper sheet to form a shield below. That will be electrically quiet. Want to go nuts? Isolate the PCB inside an insulated space and use a regulated heater to maintain a near constant temperature. Drift? What drift?

Where would you find a setup like that I described above? Instrumentation (those real expensive instruments), RF equipment, Frequency standards, Voltage standards, on and on. An extremely quiet phono stage could be done this way if you're into pushing the envelop.

Oh, and don;t forget to install test points for your meter probes on the jig. Might be helpful. I used wire loops the first time (built fast, remember?), but you can buy commercial coloured test points, as well as probe jacks. I'm thinking the probe jack is the way to go, the small pin jacks, not banana sockets!

Edit:
I usually use +/- 10 VDC to run the thing. You can adjust the C-E voltage to whatever is convenient for you, the tail current generator is happy around 10 VDC as well. If you're setting up J-Fets, you might want to keep them in your intended application below about 7 VDC. They make less noise that way.

-Chris
__________________
"Just because you can, doesn't mean you should" my Wife

Last edited by anatech; 11th December 2011 at 04:18 AM. Reason: Added details I forgot
  Reply With Quote
Old 11th December 2011, 11:49 AM   #27
zdr is offline zdr  Belgium
diyAudio Member
 
zdr's Avatar
 
Join Date: Nov 2005
Location: Brussels
Quote:
Originally Posted by Rudi_Ratlos View Post
ZDR: your layout looks good - and: you are much faster than I am!

I have two suggestions:

1) I think you are using caps with a diameter of max. 30 mm on the right side of the PCB.
Would you be able to change the caps to the EB35D - Eagle package?
Do we then need 4 capacitors or aren't 2 enough?

2) You talk about a "large MKT" capacitor that you have included. Can we settle on a specific capacitor that I will be able
to offer in this group-buy as well?
What about a Vishay MKT1813 - 1F / 100V? Its diameter is 8.5mm, its length is 19mm, so the pads shoud have a distance of about 25mm.
I think we better stick to orginal circuit design, meaning 4 large caps. Experts might chime in on why it could be relevant, I can only guess one - lower total ESR. You can further reduce length by bringing output transistors closer together.

Vishay MKT 1839 is the correct 1uF one I think (unless I ordered wrong one)
Attached Images
File Type: png fc100-ss4.png (106.5 KB, 272 views)
  Reply With Quote
Old 11th December 2011, 01:23 PM   #28
diyAudio Member
 
Rudi_Ratlos's Avatar
 
Join Date: Jun 2009
Location: Hamburg / Germany
ZDR,

we do have a low ESR by wiring the capacitors C7 and C8 parallel to the big power cap C17 (C9 and C10 parallel to C13) and placing them
very close to the output-transistors. I do not think that 2 power caps per rail are really needed.

When I will offer this PCB, I would like and offer the "hard-to-get" components as well, like the NJL*-transistors, caps, ...
Call it "a small kit" then.

I do not have access to MKT 1839 / 1F.
I have access to MKT 1822 and MKT 1813 / 1F.

I thank you very much for sending me the EAGLE-schematic-files.
I seem to have taken an older version of Mihai's files that did not include the values of the components.

@ANATECH: thank you very much for your explanations how to match transistors.
I hope that it will rain and snow this afternoon.
If so I will have the time to solder a simple matching-circuit.

If not I will go to Hamburg's Christmas market.

Best regards - Rudi_Ratlos
Attached Images
File Type: jpg Weihnachtsmarkt-Hamburg.jpg (107.3 KB, 244 views)
  Reply With Quote
Old 12th December 2011, 04:14 AM   #29
Luke is offline Luke  New Zealand
diyAudio Member
 
Join Date: Sep 2001
Location: Wellington NZ
Send a message via AIM to Luke
Quote:
Originally Posted by zdr View Post
I think we better stick to orginal circuit design, meaning 4 large caps. Experts might chime in on why it could be relevant, I can only guess one - lower total ESR. You can further reduce length by bringing output transistors closer together.

Vishay MKT 1839 is the correct 1uF one I think (unless I ordered wrong one)
nice board, I quite like how you've made use of all the wasted space in the front end.
Four caps probably accommodates more users of the board. What are its dimensions?
__________________
If you give a man a fish he will eat for a day. But if you teach a man to fish he will buy an ugly hat. And if you talk about fish to a starving man then you are a consultant. Dilbert
  Reply With Quote
Old 12th December 2011, 08:17 AM   #30
zdr is offline zdr  Belgium
diyAudio Member
 
zdr's Avatar
 
Join Date: Nov 2005
Location: Brussels
Thanks. It's 27mm shorter than original one: 64x175mm.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



New To Site? Need Help?

All times are GMT. The time now is 02:22 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