Hi All,
I am crawling the WWW for days now, often being redirected to our Diyaudio site, but whitout finding the answer I am looking for.
I bought 10 each of N and P Sanken Darlington transistors to replace some TIP142/147 in an amplifier output stage.
There's 1 pair of paralleled devices each (N/P) per channel.
Did anyone match similar devices with any of the suggested methods that are floating around ?
Thanks to all!
Max
I am crawling the WWW for days now, often being redirected to our Diyaudio site, but whitout finding the answer I am looking for.
I bought 10 each of N and P Sanken Darlington transistors to replace some TIP142/147 in an amplifier output stage.
There's 1 pair of paralleled devices each (N/P) per channel.
Did anyone match similar devices with any of the suggested methods that are floating around ?
Thanks to all!
Max
I use my own transistor matcher.
It plots on the pc screen up to 5 transistors at the same time for comparison.
It plots on the pc screen up to 5 transistors at the same time for comparison.
Thanks, I was hoping for something "easier" to cobble together, I don't expect to go through this practice often enough to justify an proper boxed up tester.
What I can't get my head around is how to handle the very high gain of the devices (5000 up) with respect to the applied current for testing...
Should I apply current to the collector or emitter with that high gain ?
What I can't get my head around is how to handle the very high gain of the devices (5000 up) with respect to the applied current for testing...
Should I apply current to the collector or emitter with that high gain ?
I applied base current through a resistor to keep the current down.
You can then compare current in base resistor and collector resistor current to find Hfe.
You can then compare current in base resistor and collector resistor current to find Hfe.
Thanks Nigel,
I'll try to get my head around it.
I suppose it's advisable to use the current that will be used in the final circuit ?
Or is that going over the top ?
Max
I'll try to get my head around it.
I suppose it's advisable to use the current that will be used in the final circuit ?
Or is that going over the top ?
Max
I am pretty sure the original mfr of the amp did not match p and n pairs, not sure why you think it is necessary now. Imo just use them and do not concern yourself with that requirement
rsavas,
thanks for that input, but I am not sure if I expressed myself correctly.
I "just" want to match the 2 parallel N's and the 2 parallel P's in each channel, not the P's to the N's.
Actually if I believe the designer of the amp they matched the transistors for HfE using pulse testing.
I'll have a look at the different circuits on the net and will provide a schematic here before I set the shop on fire...
Max
thanks for that input, but I am not sure if I expressed myself correctly.
I "just" want to match the 2 parallel N's and the 2 parallel P's in each channel, not the P's to the N's.
Actually if I believe the designer of the amp they matched the transistors for HfE using pulse testing.
I'll have a look at the different circuits on the net and will provide a schematic here before I set the shop on fire...
Max
I would just compare Vbe using a DMM in diode test mode and match them up that way.
If the design has emitter resistors, measure mVDC across them, as you would to set the OPS bias current, check each emitter R to determine what the differences are for each device. +/- 10% diff is a good target.
Rick
If the design has emitter resistors, measure mVDC across them, as you would to set the OPS bias current, check each emitter R to determine what the differences are for each device. +/- 10% diff is a good target.
Rick
Hi Rick,
thanks for your answer, actually I just ran them through one of those chinese component testers and also through the DMM in Diode mode.
The design has 0.33R emitter resistors, but they are 5% unmatched parts anyway...
I'll will follow your advice and select close readings and confirm later measuring across the emitter resistor once in circuit.
I'll post the measurements later for what it's worth.
Thanks,
Max
thanks for your answer, actually I just ran them through one of those chinese component testers and also through the DMM in Diode mode.
The design has 0.33R emitter resistors, but they are 5% unmatched parts anyway...
I'll will follow your advice and select close readings and confirm later measuring across the emitter resistor once in circuit.
I'll post the measurements later for what it's worth.
Thanks,
Max
Would the internals make any difference if all other parameters are close ?The TIP147 has an extra resistor and diode.The Sanken not.In this case 2SB1559.
Attachments
Last edited:
I am not proficient enough to answer that question I am afraid...
That said the selected Sanken parts have more than once been substituted for the TIP by various builders/service technicians/tweakers, you name it, so I just followed that path (read: "I am a painting by numbers guide").
I figured (worst case) the diode could be added on the PCB if needed, at least that's my understanding when comparing the 2 datasheets, but again, I am NOT the guy to ask...
EDIT: adding the 2nd resistor externaly is not possible, but after reading various threads for nights, I understand it's not suitable for audio anyway..
That said the selected Sanken parts have more than once been substituted for the TIP by various builders/service technicians/tweakers, you name it, so I just followed that path (read: "I am a painting by numbers guide").
I figured (worst case) the diode could be added on the PCB if needed, at least that's my understanding when comparing the 2 datasheets, but again, I am NOT the guy to ask...
EDIT: adding the 2nd resistor externaly is not possible, but after reading various threads for nights, I understand it's not suitable for audio anyway..
Last edited:
Enzo, this is why I'm asking. I had to fix a Harmon Kardon AV amp a week or two ago. I decided to order the original Sankens from Digikey.
"I bought 10 each of N and P Sanken Darlington transistors to replace some TIP142/147 in an amplifier output stage."
Coolnose, the diode is there for protection. There is also a big difference in gain.
Regards
Jan
"I bought 10 each of N and P Sanken Darlington transistors to replace some TIP142/147 in an amplifier output stage."
Coolnose, the diode is there for protection. There is also a big difference in gain.
Regards
Jan
Last edited:
My question would be "did you match them in any way" ?
EDIT: yes, for the diode thanks! Needed ? Maybe...
As for the gain I understand the gain on the Sankens is much higher, but looking at the datasheet it climbs with bias, I will be putting 20mA on each transistor only, but maybe I am not understanding this correctly anyway 😉
EDIT: yes, for the diode thanks! Needed ? Maybe...
As for the gain I understand the gain on the Sankens is much higher, but looking at the datasheet it climbs with bias, I will be putting 20mA on each transistor only, but maybe I am not understanding this correctly anyway 😉
Last edited:
You might actually put a pair in parallel in a similar circuit to the amp- with whatever base and emitter resistors they use, put a current meter in series with each, and drive them with a pulse or continuous (may need sink). A wise person once said, "test it the way you use it."