Author |
Message |
Sebo
Joined: Apr 27, 2007 Posts: 564 Location: Argentina
|
Posted: Thu May 24, 2007 10:19 am Post subject:
How to match transistors? |
|
|
Hi:
I want to know if I can match transistor using the Hfe meter of my cheap
multimeter.
If not, can anybody tell me a easy way to do it.
I need to match about 6 transistor for a transistor ladder filter, and I think
it will be useful knowledge for future projects.
Thanks.
Sebo _________________ Sebo
---------------------------------------
My Music:
https://www.facebook.com/cosaquitos/ |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Thu May 24, 2007 1:04 pm Post subject:
|
|
|
Did try some googling on this but can't really find what characteristic trannies should be matched on for ladder filters. I'll try to write down what I think to know, that will be a bit general though
A multimeter doing Hfe measurement will do so by applying a fixed base current and measuring the resulting collector current. Usually though matching should be checked over a wide range of signal levels. Otoh for filters this is usually considered less important than for oscillators (unless you want to build tracking oscillating filters, but then a ladder filter seems less ideal than let's say a state variable filter).
What can be said is that Hfe matching probably will be a good first order approximation for matching, in that when the Hfe-s differ wildly other parameter will not match good either.
So I guess you can get away with the Hfe measurement on your meter. But I'm not really very sure, as I don't know how important the Vbe vs Ib characteristic is for a ladder filter. to measure that relation Hfe measurement is not directly useful as I think this relation varies with the size of the base emiter junction - I read that somewhere - and that would depend on manufacturing precision ...
Well, maybe someone else has more on this. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
Mooger5
Joined: May 02, 2007 Posts: 199 Location: Portugal
Audio files: 8
|
Posted: Thu May 24, 2007 1:18 pm Post subject:
|
|
|
There are instructions for transistor matching on page 9-3 of the Minimoog service manual. |
|
Back to top
|
|
|
Funky40
Joined: Sep 24, 2005 Posts: 875 Location: Swiss
Audio files: 1
G2 patch files: 5
|
|
Back to top
|
|
|
Mooger5
Joined: May 02, 2007 Posts: 199 Location: Portugal
Audio files: 8
|
Posted: Thu May 24, 2007 1:26 pm Post subject:
|
|
|
Yes, the second circuit is taken from the Minimoog manual. I was going to post the page but couldn´t edit my post. |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Thu May 24, 2007 1:39 pm Post subject:
|
|
|
Ah good practical info
That's matching on Vbe vs Ic so on the Vbe vs Ib and Hfe at the same time and so basically for expo converters. Does the Moog manual say this should be used for the ladder filter as well ?
These measurements will be temperature dependent, so maybe pay attention to the notes given by Moog about that. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
Mooger5
Joined: May 02, 2007 Posts: 199 Location: Portugal
Audio files: 8
|
Posted: Thu May 24, 2007 2:02 pm Post subject:
|
|
|
Blue Hell wrote: |
Does the Moog manual say this should be used for the ladder filter as well ?
|
"Q26 & Q28 matched at +/-3mV at 20mA" - Q26 is a PNP controlling the cutoff range and is connected to Q28, the NPN at the bottom of the ladder.
AFAIK, they started using CA3046 instead of matched transistors on the Micromoog, for economy reasons. Last edited by Mooger5 on Thu May 24, 2007 2:14 pm; edited 1 time in total |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Thu May 24, 2007 2:10 pm Post subject:
|
|
|
Thanks ! _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
Mooger5
Joined: May 02, 2007 Posts: 199 Location: Portugal
Audio files: 8
|
|
Back to top
|
|
|
Sebo
Joined: Apr 27, 2007 Posts: 564 Location: Argentina
|
Posted: Thu May 24, 2007 6:07 pm Post subject:
|
|
|
Thanks all for the replies !!!!
I gonna try the Moog circuit, seems simple. I hope my multimeter
have enough resolution to do it. _________________ Sebo
---------------------------------------
My Music:
https://www.facebook.com/cosaquitos/ |
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007 Posts: 1734 Location: NM USA
Audio files: 54
|
Posted: Thu May 24, 2007 8:57 pm Post subject:
error in matching schematic |
|
|
There is a (well-known) mistake in the second schematic. Add a 10k resistor between the transistors as in the first schematic.
One other point -- if you buy a modern batch of devices they will usually all be matched to within ~2 mV, in my experience. If you get a late array such as the CA3083 matching is usually better than .2 mV.
For accurate matching for critical applications such as roll-your-own OTA's, you can use bridge-type circuits. With care, you can measure within .01 mV. |
|
Back to top
|
|
|
Pehr
Joined: Aug 14, 2005 Posts: 1307 Location: Björkvik, Sweden
Audio files: 2
|
|
Back to top
|
|
|
dave_b
Joined: Apr 01, 2007 Posts: 42 Location: Kansas City
|
Posted: Fri May 25, 2007 9:54 am Post subject:
|
|
|
Perhaps a silly question: Regarding the Moog test circuits, is it necessary to remove power when swapping transistors? |
|
Back to top
|
|
|
funkyfarm
Joined: Jan 21, 2007 Posts: 583 Location: France
Audio files: 3
|
Posted: Fri May 25, 2007 1:43 pm Post subject:
Re: error in matching schematic |
|
|
Can you explain ?
next to "100uA" ?
I saw 10k on both...
dave_b wrote: | Perhaps a silly question: Regarding the Moog test circuits, is it necessary to remove power when swapping transistors? |
I guess no !
but do it quickly |
|
Back to top
|
|
|
Mooger5
Joined: May 02, 2007 Posts: 199 Location: Portugal
Audio files: 8
|
Posted: Fri May 25, 2007 5:34 pm Post subject:
Re: error in matching schematic |
|
|
funkyfarm wrote: |
Can you explain ?
next to "100uA" ?
|
I think it´s right there by looking at the symmetry of both circuits. In the NPN circuit the reference transistor is between 51k and 10k resistors. In the PNP the 10k is missing. |
|
Back to top
|
|
|
funkyfarm
Joined: Jan 21, 2007 Posts: 583 Location: France
Audio files: 3
|
Posted: Fri May 25, 2007 10:05 pm Post subject:
|
|
|
Ooops, sorry (refering to another source) |
|
Back to top
|
|
|
Funky40
Joined: Sep 24, 2005 Posts: 875 Location: Swiss
Audio files: 1
G2 patch files: 5
|
Posted: Sat May 26, 2007 12:00 pm Post subject:
|
|
|
frijitz wrote: | There is a (well-known) mistake in the second schematic. Add a 10k resistor between the transistors as in the first schematic.
|
luckily i had to use the upper one.
Thanks for the tip.
dave_b wrote: | Perhaps a silly question: Regarding the Moog test circuits, is it necessary to remove power when swapping transistors? |
No, good question.
I did this things without any knowledge. sometimes thats better, sometimes not .
One should not touch the Transistors by Hand when matching because of heat , so i used a plier.
This tourned out as not so easy and very Timeconsuming. ( really Timeconsuming )
Once i took the Transistor on its legs, crosswise, so i did a short. To be save, i throwed it away.
Next: to allways pull down the Power would be also very Timeconsuming.( The power i use is a long Powercord for Modules, coming from my modularrack )
I did not |
|
Back to top
|
|
|
Fernando
Joined: Dec 30, 2006 Posts: 286 Location: Barcelona & Emporda, Spain
|
Posted: Sat May 26, 2007 12:52 pm Post subject:
|
|
|
I think I'll use the James Patchell circuit when I need to (simpler), unless someone advise me to use the Moog one for any reason...
Am I right in reading that one can use the Moog matcher at +-15V without changing any resistor? |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Sat May 26, 2007 2:25 pm Post subject:
|
|
|
Fernando wrote: | Am I right in reading that one can use the Moog matcher at +-15V without changing any resistor? |
The 100 uA current is determined by the voltage divider at the opamps input and the 51 k resistor going to the neg. supply. The opamps + input is set at -5V for a 10V supply, it will try to regulate it's output in such a way that the - input will also be at -5V. This means there is 5V over the 51 k resistor and that means that about 100 (98) uA will flow.
For a 15 V supply the voltage at the opamp's + input wil be -7.5V resulting in 7.5 V to be present over the 51 k resistor resulting in approx. 150 uA of current to flow. So you'll measure at another set point then.
When you want that 100 uA again the resistor should be around 75 k for a +/- 15 V supply. the current does not really depend on the positive supply voltage, but it's linear with the negative supply.
You could also replace the 10 k resistor going from the opamp + input to the negative supply with a 5V1 zener, that would make the circuit independent from the negative supply voltage as well. A 5v1 zener coincidentally also has a near zero temperature coefficient. In that case keep the 51 k resistor. _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
Fernando
Joined: Dec 30, 2006 Posts: 286 Location: Barcelona & Emporda, Spain
|
Posted: Sat May 26, 2007 5:39 pm Post subject:
|
|
|
Blue Hell wrote: | Fernando wrote: | Am I right in reading that one can use the Moog matcher at +-15V without changing any resistor? |
The 100 uA current is determined by the voltage divider at the opamps input and the 51 k resistor going to the neg. supply. The opamps + input is set at -5V for a 10V supply, it will try to regulate it's output in such a way that the - input will also be at -5V. This means there is 5V over the 51 k resistor and that means that about 100 (98) uA will flow.
For a 15 V supply the voltage at the opamp's + input wil be -7.5V resulting in 7.5 V to be present over the 51 k resistor resulting in approx. 150 uA of current to flow. So you'll measure at another set point then.
When you want that 100 uA again the resistor should be around 75 k for a +/- 15 V supply. the current does not really depend on the positive supply voltage, but it's linear with the negative supply.
You could also replace the 10 k resistor going from the opamp + input to the negative supply with a 5V1 zener, that would make the circuit independent from the negative supply voltage as well. A 5v1 zener coincidentally also has a near zero temperature coefficient. In that case keep the 51 k resistor. |
Thanks a lot. I'll try to digest it and make further questions |
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007 Posts: 1734 Location: NM USA
Audio files: 54
|
Posted: Sun May 27, 2007 12:48 pm Post subject:
Re: error in matching schematic |
|
|
Mooger5 wrote: | I think it´s right there by looking at the symmetry of both circuits. In the NPN circuit the reference transistor is between 51k and 10k resistors. In the PNP the 10k is missing. |
Yes, right. Sorry not to be clearer. If you look (literally) between the two transistors on the schematic you will see that one circuit has a 10 k resistor and the other one has just a wire. Put the additional 10k where the wire is.
Ian |
|
Back to top
|
|
|
LetterBeacon
Joined: Mar 18, 2008 Posts: 454 Location: London, UK
|
Posted: Tue Feb 17, 2009 6:08 am Post subject:
|
|
|
*bump*
My DMM has a DC voltage range of 200mV to 1000V and a resolution of 1mV, that is what I need for this circuit, right?
Does anyone have a stripboard layout for this circuit at all?
Thanks!
EDIT - Oh yes, can this circuit work on a +/- 9 volt supply? |
|
Back to top
|
|
|
kkissinger
Joined: Mar 28, 2006 Posts: 1354 Location: Kansas City, Mo USA
Audio files: 42
|
Posted: Tue Aug 10, 2010 8:31 am Post subject:
Bump |
|
|
If you've done any DIY lately you've likely discovered that it is becoming more difficult to find thru-hole matched transistor pairs. The solution is either to use SMT (and the SMT transistor pairs are almost microscopic!) or to resort to hand-matching discreet transistors.
This thread has some test circuits to allow one to match transistors. It will be useful to me and I decided to bump this thread for others to see. _________________ -- Kevin
http://kevinkissinger.com |
|
Back to top
|
|
|
emdot_ambient
Joined: Nov 22, 2009 Posts: 667 Location: Frederick, MD
|
Posted: Tue Aug 10, 2010 3:48 pm Post subject:
|
|
|
Thanks, it is very useful!
My 52 year old eyes can barely see SMT components, let alone see where I'm supposed to solder them! |
|
Back to top
|
|
|
adambee7
Joined: Apr 04, 2009 Posts: 420 Location: united kingdom
|
Posted: Tue Aug 10, 2010 3:52 pm Post subject:
|
|
|
I built ray wilsons transistor matching circuit. Pretty good. Just pop over to music from outerspace site and the links on the homepage. |
|
Back to top
|
|
|
|