rajeev luthra,
It makes sense to me. My experience confirms what jneutron said about the unknown element involved with buying second rate parts. I know you have no choice over there, but I purchased some surplus 1N4002 diodes and tested them for Vbr, and stopped at 600v. They may actually be more like 1N4007.
It makes sense to me. My experience confirms what jneutron said about the unknown element involved with buying second rate parts. I know you have no choice over there, but I purchased some surplus 1N4002 diodes and tested them for Vbr, and stopped at 600v. They may actually be more like 1N4007.
The devices have been processed with 90 volts in mind..so, the manufacturer datasheets will have an SOA curve that is bounded by the bonding wire limitation to a maximum current, a theta jc curve maximum, and at the higher voltage, by a secondary breakdown area, then finally, by the device maximum rating in voltage.
By using it over 90 volts, you are heavily into the secondary breakdown area..I would not recommend that on general principles. And it is difficult to say if they will survive, how reliable they will be, and how they will react to the reactance of the speaker load...if you look at the SOA area, extending the vertical voltage line doesn't really buy you too much additional capacity device wise..
I run mobile DJ apps, and prefer reliability over cost..so I would not be comfortable using the devices above ratings. Your's may survive in a well designed circuit, but I'd be wary.
Cheers, John
Hello,
Thanks subwo1 , joan2 , and junetron ,
actually how I came up to this topic of breakdown testing is because ,in another
thread I had mentioned that I was using 90v transistors in an amp with + / - 60v
rails , here it was pointed out by djk that I was lucky that my transistors were
survivng at these voltages , later Ashok also said the same thing , hence I got
worried , but my amp is working fine since rearly last two years , this is because
the working voltages of devices is normally higher than rated and it is even higher
if the emitter base resistor in the circuit is 100 ohms or lower.
junetron
I do not want to take the risk but I have no choice , as here pnp devices are not
avilible easily , however I am looking if I can find a source from where I can get
better devices , and I hope to do so by the time the higher wattage version of my
amp is complete. However this test will always be very valuable to me.
I got a very encouraging e-mail from
Stephen Mantz he gave his breakdown testing method and the follows ;-.
( ( I used to do this 30 years ago with 2N3055's to use them in high powered
amps as 2N3773 and 2N6259s were so expensive in those days.
Typically all good quality trannies will exceed their Vceo ratings by a good margin
- why because the manufacturer must guarantee the spec.
Now remember something. All trannies are operated with some sort of Coll-Base
resistor (especially output types) and the Vcb rating is much higher than the Vceo.
So you can do the same above scope test by simply connecting the alligator leds
to Coll- Base instead.
Do not forget to check SOA if doing this with outputs. I have routinely used
TIP35C/36C on +/- 70 v rails with NO pre-grading done. Why because with
experience the Vceo 100v rating is always higher and the Vcb rating is even
higher again and I have 22 ohm base-emitter resistors in the circuit.
Of course with +/- 70v rails and depending on the lowest impedance being driven,
enough device must be used to ensure safe operation ))
With Good Wishes
Rajeev Luthra
Thanks subwo1 , joan2 , and junetron ,
actually how I came up to this topic of breakdown testing is because ,in another
thread I had mentioned that I was using 90v transistors in an amp with + / - 60v
rails , here it was pointed out by djk that I was lucky that my transistors were
survivng at these voltages , later Ashok also said the same thing , hence I got
worried , but my amp is working fine since rearly last two years , this is because
the working voltages of devices is normally higher than rated and it is even higher
if the emitter base resistor in the circuit is 100 ohms or lower.
junetron
I do not want to take the risk but I have no choice , as here pnp devices are not
avilible easily , however I am looking if I can find a source from where I can get
better devices , and I hope to do so by the time the higher wattage version of my
amp is complete. However this test will always be very valuable to me.
I got a very encouraging e-mail from
Stephen Mantz he gave his breakdown testing method and the follows ;-.
( ( I used to do this 30 years ago with 2N3055's to use them in high powered
amps as 2N3773 and 2N6259s were so expensive in those days.
Typically all good quality trannies will exceed their Vceo ratings by a good margin
- why because the manufacturer must guarantee the spec.
Now remember something. All trannies are operated with some sort of Coll-Base
resistor (especially output types) and the Vcb rating is much higher than the Vceo.
So you can do the same above scope test by simply connecting the alligator leds
to Coll- Base instead.
Do not forget to check SOA if doing this with outputs. I have routinely used
TIP35C/36C on +/- 70 v rails with NO pre-grading done. Why because with
experience the Vceo 100v rating is always higher and the Vcb rating is even
higher again and I have 22 ohm base-emitter resistors in the circuit.
Of course with +/- 70v rails and depending on the lowest impedance being driven,
enough device must be used to ensure safe operation ))
With Good Wishes
Rajeev Luthra
How about stacking one tr on the other?
Hi Rajeev,
I understand your position perfectly.
You could however try using the power transistors stacked like (I think ) in the Ampzilla power amp. There each device sees much less than the supply voltage. I might be wrong but check it out. I saw it a long time ago. I must have the circuit somewhere.
I'll look for it.
But I will repeat. Try to make the amp as bullet proof as possible. It's a waste of time and money to fix broken amps - if it fails due to a 'rating' problem. Not to mention the users reaction.
You can spend that time building more amps or doing constructive development.
Cheers,
Ashok.
Hi Rajeev,
I understand your position perfectly.
You could however try using the power transistors stacked like (I think ) in the Ampzilla power amp. There each device sees much less than the supply voltage. I might be wrong but check it out. I saw it a long time ago. I must have the circuit somewhere.
I'll look for it.
But I will repeat. Try to make the amp as bullet proof as possible. It's a waste of time and money to fix broken amps - if it fails due to a 'rating' problem. Not to mention the users reaction.
You can spend that time building more amps or doing constructive development.
Cheers,
Ashok.
That was a mistake.
I had a look at Ampzilla III just now and it doesn't use two transistors to beat the Vceo limit. Look at the Ampzilla
and Ampzilla II circuits
at
http://www.audio-circuit.dk/Schematics.html
Cheers,
Ashok.
I had a look at Ampzilla III just now and it doesn't use two transistors to beat the Vceo limit. Look at the Ampzilla
and Ampzilla II circuits
at
http://www.audio-circuit.dk/Schematics.html
Cheers,
Ashok.
Digital amps.
Hi Rajeev,
I think you should seriously look at the new generation of Digital amplifiers. There are quite a few chipsets around and plenty of information.
It will probably solve all your problems and provide more power than you want.
You must have seen some of the threads on that subject on this forum. Maybe you should work on that in parallel .
Are the 2SA/2SC devices genuine ? I plan to be in Mumbai sometime in September and would like to get a few pairs.
Cheers,
Ashok.
Hi Rajeev,
I think you should seriously look at the new generation of Digital amplifiers. There are quite a few chipsets around and plenty of information.
It will probably solve all your problems and provide more power than you want.
You must have seen some of the threads on that subject on this forum. Maybe you should work on that in parallel .
Are the 2SA/2SC devices genuine ? I plan to be in Mumbai sometime in September and would like to get a few pairs.
Cheers,
Ashok.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.