Matched PNP/NPN dual transistors

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Which are the options available for dual matched transistors?

The only ones I know are MAT02 and MAT03, but they seem to be a vanishing thing. Those are also low voltage.

Toshiba made (still does?) the 2SC3381 and 2SA1349, but they were DUAL transistors, not necessarily matched.

Harman Kardon used some dual pairs named uPA74V and uPA75V on their hk770 amp, but they are not to be found, or even their data.

Any suggestions?



Carlos
 
Re: MAT02/03 substitute

Lenin said:
Try LM194/LM394 (pin for pin MAT02/03 replacement)

if you really must have a higher voltage capable device, better brains than mine will appear shortly.

The LM394 has worked fine for me in the front end of MC circuits.:nod:


Both 194 and 394 are NPN types. So they would replace MAT02 but not MAT03.

If my search was for low voltage types I would stay with the MATs. Unfortunately this application demands higher voltages.


Carlos
 
"Monolithic" dual type.

Hey Carlos,

The two Toshiba parts that you mention are monolithic in construction meaning that they were constructed on the same substrate and should be very, very closely matched. Every pair that I have used has been within 1% or less when comparing electrical characteristics. I have not put them on the curve tracer, but I guess that I could to see just how close they really are.

If you are working at voltages that are higher than 80 V this is not the transistor for you. It is also low power and primarily designed to be used as an audio input stage device. You can find the .pdf here if you do not have it already:

http://www.milton.arachsys.com/nj71/pdf/2sc3381.pdf,
http://www.milton.arachsys.com/nj71/pdf/2sa1349.pdf.

By the way, I believe that the National LM194/394 came before the Analog Devices MAT series. I have been considering using both for a couple of designs that I am working on but can't seem to find the time.

Later,
 
Toshiba duals

stadams said:


The two Toshiba parts that you mention are monolithic in construction meaning that they were constructed on the same substrate and should be very, very closely matched. Every pair that I have used has been within 1% or less when comparing electrical characteristics. I have not put them on the curve tracer, but I guess that I could to see just how close they really are.

If you are working at voltages that are higher than 80 V this is not the transistor for you. It is also low power and primarily designed to be used as an audio input stage device. You can find the .pdf here if you do not have it already:

http://www.milton.arachsys.com/nj71/pdf/2sc3381.pdf,
http://www.milton.arachsys.com/nj71/pdf/2sa1349.pdf.

By the way, I believe that the National LM194/394 came before the Analog Devices MAT series. I have been considering using both for a couple of designs that I am working on but can't seem to find the time.


Thanks very much for the info on how the two internal transistors compare eletrically. That's what I wanted to know.

The next question would be to know how NPN and PNP duals compare between them.

BTW: Reichelt Elektronik in Germany stocks those duals and for less than 1 Euro each.

You're right about the 194/394 age: they came before. But some consider the MAT a bit better. The fact that there's a PNP option is also good.

Now if I could get info on the uPA74/uPA75 that would be great!...



Carlos
 
Yes, but Reichelt won't tell you the hfe group + they will deliver second sources without notice which really don't fulfill the original spec.

These duals that I received half a year ago seemed to be Toshiba, but were from completely different gain groups (something like PNP highest and NPN lowest). Some of the Toshiba single transistors they sell are actually CDIL.
 
capslock said:
Yes, but Reichelt won't tell you the hfe group + they will deliver second sources without notice which really don't fulfill the original spec.

These duals that I received half a year ago seemed to be Toshiba, but were from completely different gain groups (something like PNP highest and NPN lowest). Some of the Toshiba single transistors they sell are actually CDIL.

Any other place to get them? Borbely sells them, but a bit more expensive.

Any USA source for them?

As I already have several MAT02 and MAT03, it would be great if I could use them, but their voltage limitations would demand cascodes to take care of the extra voltage.



Carlos
 
There are transistors like BFX36 and 2N2914, dual NPN/PNP which can handle much higher voltages than the MAT02/03. Elektor used the former in their 100watt Hi-End amplifier. There is the SSM2210 and SSM2220, also from AD. I have seen data relating to something in the BCW series like BCW32 etc., which are also dual matched transistors.

Carl, would you be willing to sell/trade a few of the MAT03s in your possession. I bought 4 pairs of MAT devices a few years ago. It turned out that what was supplied as MAT03 were actually slightly low hfE MAT02s that were reprinted (the print is so fine that it is hard to tell from the original print). I think I have all 8 MAT02 transistors intact and would like to use them in a Class-A preamp and Ribbon Speaker Amplifier, both published by Elektor. Thanks.
 
For matched transistors

Check out

http://www.linearsystems.com/

They are a small manufacturer and are open (at the time I used them)
to small runs of custom components as well as precision matching of transistors and other components. They have some pretty impressive "off the shelf" also;

LS310, 311, 312, 313 Tightly Matched, Monolithic Dual, NPN Transistor
LS318 Log Conformance, Monolithic Dual, NPN Transistor
LS3250 Higher Current, Monolithic Dual and Single, NPN Transistor
LS350, 351, 352 Tightly Matched, Monolithic Dual, PNP Transistor
LS3550 Higher Current, Monolithic Dual and Single, PNP Transistor
LS358 Log Conformance, Monolithic Dual, PNP Transistor


Hope this helps...
 
Leolabs said:
If I was not mistaken,RS Components does have the SSM2210/2220.
Digikey too.

I found that the LM devices and MAT devices are fairly widely available, and the SSM ones are there with one or two well-known retailers. Where do you get the Japanese high-Vceo devices? I too wanted to use them in the input LTP of amps instead of 2N5551/5401, but they're so impossible to find.
 
AndrewT said:
Hi Tcpip,
just cascode them and gain in voltage handling, constant voltage loading and in frequency extension.
Oh, I agree 100%, but I don't know how to design amps. I have to stick to ready-made tested designs. :( So, when the designer tells me that I can replace his input devices with a dual-device chip, I go and make the swap. But I don't know how I'll have to recalculate degeneration resistors or change CCS settings if I introduce a cascode stage. That's the problem which stops me.
 
tcpip said:
......how I'll have to recalculate degeneration resistors or change CCS settings if I introduce a cascode stage...
more research on the topic of amp design!
Adding a cascode is easy. You can leave the degeneration and LTP tail current exactly as is.
But, and a very important but, the extra high frequency gain extracted from the stage will change the requirements of the stability components.
I suspect that the only sure way of getting this right is oscilloscope, debugging and skill.
 
AndrewT said:
Adding a cascode is easy. You can leave the degeneration and LTP tail current exactly as is.
Great. This is good news.
But, and a very important but, the extra high frequency gain extracted from the stage will change the requirements of the stability components.
I suspect that the only sure way of getting this right is oscilloscope, debugging and skill.
I guess this means that the compensation capacitor value may need to be changed or experimented with. Simulation may help to get a first-cut approximation, followed by playing around with a prototype?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.