Transistor matching methods. - 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 22nd December 2002, 02:38 AM   #1
aborza is offline aborza  United States
diyAudio Member
 
Join Date: Nov 2002
Location: Sunny Florida
Question Transistor matching methods.

What method/s (circuit/s?) do you use to match differential transistors?

How about complimentary transistors?

I mean other than a handy RS meter with hfe option or a curve tracer which is not usually found at your local RS store?

Ideally transistors should be matched over the expected operating currents, not at one point only. So, how are you matchers doing it?

Any advice (and circuits) will be appreciated.
  Reply With Quote
Old 22nd December 2002, 02:50 AM   #2
diyAudio Member
 
chris ma's Avatar
 
Join Date: Sep 2001
Location: Pickering, Canada
I did not really match my Mj15003 with my JLH for ESL. I have 4 MJ15003 per channel, ideally they should all have equal share of the bias current. When making the chassis I kept in mind that I have easy access to measure and replace them. Apparently the best way to check them is in circuit. I have them running for an hour to let them stablized and check them. I keep swoping them until each of them have the same current each, in my case 2A each. But if you have too many transistors to match then my method is not good for you.
Regards,
Chris
  Reply With Quote
Old 22nd December 2002, 02:58 AM   #3
diyAudio Member
 
Equilibrium's Avatar
 
Join Date: Oct 2002
Location: Mad Dog, Texas
Default Mr. Elliott is just full of tricks

give this here link a looksee
__________________
If we could just get everyone to close his or her eyes and visualize world peace for an hour, imagine how serene and quiet it would be until the looting started.
  Reply With Quote
Old 22nd December 2002, 07:02 AM   #4
Electrons are yellow and more is better!
diyAudio Member
 
peranders's Avatar
 
Join Date: Apr 2002
Location: Göteborg, Sweden
Blog Entries: 4
Default Re: Transistor matching methods.

Quote:
Originally posted by aborza
Any advice (and circuits) will be appreciated.
The best thing is to use matched pairs. Check here.

These BC847BS are really cheap.

If you want to match single transistors do like this:

Make a diff stage (or the one you are going to use) with same currents and the voltage across the transistors. If the bases are connected with some resistance try to do the same in your test setup in order to get working conditions as much like the real stage as possible.

Connect a voltmeter across the collectors so you can measure the unbalance.

Keep the bases grounded (if you use +- voltage) or at the same potential (if you use signle supply voltage).

Let one of the transistors be there and change only one of them. Write down the offset for each transistor and place the transistor where you have written the numbers.

Then when you have a couple of transistors which creates the same offset, they will probely work good togheter.
__________________
/Per-Anders (my first name) or P-A as my friends call me
Tube Buffered Gainclone in work |Thread ||
  Reply With Quote
Old 22nd December 2002, 05:12 PM   #5
aborza is offline aborza  United States
diyAudio Member
 
Join Date: Nov 2002
Location: Sunny Florida
Thanks all,

Peranders really got me thinking. So I came up with the following circuit.

It is not "exactly" like the circuit the transistors will be used in because Q2 in the operating circuit will not have a collector resistor and the match would be impossible. So I placed a 680R in the collector of Q2. It gives a place to take a measurement from and also balances the system. The simulator gives it a go!

R1,2,3,4,6 & 7 should be at least 1% types.

Notice the two meters.

Both meters will see 0V with a perfect match (unlikely of course). But Meter A between nodes 1 and 4 will be comparing points at about 31V and Meter B between nodes 3 and 6 will be comparing points at about .7V. Any comments as to which will result in better accuracy re the match?

Also the 16k can be varied to see if the transistors go off in different directions with different currents.

I do not know how close a match this circuit will show, but it should be better than nothing.
Attached Images
File Type: gif da match.gif (4.1 KB, 1035 views)
  Reply With Quote
Old 22nd December 2002, 05:18 PM   #6
aborza is offline aborza  United States
diyAudio Member
 
Join Date: Nov 2002
Location: Sunny Florida
In regards to complimentary matching, I rustled around and found this circuit presented by Richard N. Marsh in AA during the 80's.

I think all the resistors in this circuit should be 1% or better for proper matching.

I hope you find it useful.
Attached Images
File Type: gif marsh compl match cir.gif (5.1 KB, 1027 views)
  Reply With Quote
Old 22nd December 2002, 06:23 PM   #7
Electrons are yellow and more is better!
diyAudio Member
 
peranders's Avatar
 
Join Date: Apr 2002
Location: Göteborg, Sweden
Blog Entries: 4
Quote:
Originally posted by aborza
Thanks all,

Peranders really got me thinking. So I came up with the following circuit.
This was what I ment but I think it's sufficient with only measuring the collector voltage. This is a good approximation because you can't match every parameter.

May I ask what is the purpose of this matching? Don't forget what feedback will do.

I you have the opportunity test the BC847BS, good!
__________________
/Per-Anders (my first name) or P-A as my friends call me
Tube Buffered Gainclone in work |Thread ||
  Reply With Quote
Old 22nd December 2002, 07:44 PM   #8
aborza is offline aborza  United States
diyAudio Member
 
Join Date: Nov 2002
Location: Sunny Florida
perancers,

Thanks for the reply,

I was not suggesting multiple tests. I was asking which point of measurement (A or B NOT A and B) would result in a more discriminating and accurate matching process.

As for the matched transistors (BC847BS), no luck. I was able to find dual BC847 surface-mount transistors for sale but not matched pairs.

As for the reason for matching:

1. It seems that, especially in the front end of an amplifier where error correction is focused, well matched devices will have less to deal with if they are not forced to correct for inherent distortion and offset that would otherwise be caused by poorly matched devices.

2. This is diy. The transistors are cheap and matching is like chicken soup to a cold. I couldn't hurt.

3. There is opinion that matching helps sonics and opposing opinion that feedback can cure all ills. By giving a diy'er the opportunity to try both, he can choose for himself.

4. Matching through-hole components gives the beginner options he would not otherwise have given poor soldering skills and the difficulties of handling surface-mount devices.

Ultimately I wish to post a site with some of these options for the beginner along with a versatile PCB to incorporate a variety of other options. It is a re-learning experience for me since I have been away from hands-on electronics for many years and I hope that in getting some of my questions answered, others will have questions answered as well.

The beginning of the quest can be found at: http://www.diyaudio.com/forums/showt...0247#post90247

And I need all the help I can get.

BYW revised schematics are comming soon.
  Reply With Quote
Old 22nd December 2002, 08:24 PM   #9
Electrons are yellow and more is better!
diyAudio Member
 
peranders's Avatar
 
Join Date: Apr 2002
Location: Göteborg, Sweden
Blog Entries: 4
Quote:
Originally posted by aborza
A: I was not suggesting multiple tests. I was asking which point of measurement (A or B NOT A and B) would result in a more discriminating and accurate matching process.

B: As for the matched transistors (BC847BS), no luck. I was able to find dual BC847 surface-mount transistors for sale but not matched pairs.

As for the reason for matching:

1. It seems that, especially in the front end of an amplifier where error correction is focused, well matched devices will have less to deal with if they are not forced to correct for inherent distortion and offset that would otherwise be caused by poorly matched devices.

2. This is diy. The transistors are cheap and matching is like chicken soup to a cold. I couldn't hurt.

3. There is opinion that matching helps sonics and opposing opinion that feedback can cure all ills. By giving a diy'er the opportunity to try both, he can choose for himself.

4. Matching through-hole components gives the beginner options he would not otherwise have given poor soldering skills and the difficulties of handling surface-mount devices.
A: You are always interested in the collector currents, you shuold measure therefore the currents (or voltages over the collector resitors)

B: These BC8x7BS is monolithic pairs and have descent matching and is very cheap, only 20 cents!... but not suitable for newbeginners unless they are very handy. 0.025" between the pins isn't for everyone.

1-3: The whole idea of a differantial input stage is that the input transistors are excactly alike. If they are, the offset will be small, offset drift with temp will be small, the power supply rejection will be better and also the distortion will be lower.

The feedback can't cure or cancel defects in the first stage. Therefore it's important to make this stage as good as possible. Low noise is also often important.
__________________
/Per-Anders (my first name) or P-A as my friends call me
Tube Buffered Gainclone in work |Thread ||
  Reply With Quote
Old 23rd December 2002, 04:51 AM   #10
aborza is offline aborza  United States
diyAudio Member
 
Join Date: Nov 2002
Location: Sunny Florida
Unhappy Correction

Sorry,

The prior schematic for the complimentary match circuit was wrong.

Here is the corrected schematic. But I am not sure this circuit will work properly either.

Note that the meter is suggested to read uA and is not in series. It would seem the measure should be in V's not A's. Also I put the circuit into simulation and it did not seem to work out sensibly.

Can someone check this out?
Attached Images
File Type: gif marsh compl match cir.gif (5.9 KB, 929 views)
  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


Similar Threads
Thread Thread Starter Forum Replies Last Post
Transistor matching graham-r Parts 1 6th May 2009 07:21 PM
transistor matching derekyu Solid State 38 26th May 2007 07:57 AM
Transistor linearization methods Bricolo Solid State 33 26th October 2006 09:20 AM
transistor matching weinstro Parts 28 6th July 2003 02:52 PM


New To Site? Need Help?

All times are GMT. The time now is 04:38 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