janneman said:
I think you'll find that the low Vce part of the SOA curve is not an issue with reactive load-drive....
The key is using as much of the SOA at Vce>=Vcc as is safe...
You may of course wish to utilize more of the SOA in the Vce<<Vcc region in the interest of improving the amp's ability to drive purely resistive loads....
Whether this is worthwhile or even realistic is moot...
Sam,
Good point. But an amp with a well-designed V/I limiter should be able to survive an event as you describe. In fact, it should be able to survive a continuous short, except for the thermal factor. Although the short-circuit dissipation should be within the device capacity, it does raise the heatsink temp until the point where the device still fails. So, for longer-time shorts etc maybe we should include a thermal breaker.
Jan Didden
Good point. But an amp with a well-designed V/I limiter should be able to survive an event as you describe. In fact, it should be able to survive a continuous short, except for the thermal factor. Although the short-circuit dissipation should be within the device capacity, it does raise the heatsink temp until the point where the device still fails. So, for longer-time shorts etc maybe we should include a thermal breaker.
Jan Didden
A well designed VI limiter should be able to do all that AND not effect sound quality. I believe there are some that do all this. I also have experience with some that don't. And as you say some that are OK for a while until thermal failure enters.
BTW, I also note that the voltage and current present at some amplifier outputs, at least the more powerful ones, are potentially hazardous. VI-limiters provide protection from this as well. This is another reason why I'm not fond of bare binding posts. Of course it is good to know that there is protection should I become the "shortest path to ground", I really would prefer not be in that situation. Same for guests, pets, and small children.
I'm all for VI limiters - always use them -- but it should never be forgotten that anytime a VI limiter becomes active it means some other system has failed.
Add the general category of "design fault". Not necessarily of the amplifier. For instance speakers having drastic impedance dips. Very analogous to software that is incompatible with other software or hardware that any reasonable person would expect it to encounter routinely. The reverse is also valid.
This discussion points out one difference in DIY vs. commercial considerations. If you are designing something for commercial use it is wise (if only to limit warranty returns) to build in as many protection from operator error and "other Product" compatability as you can think of. For DIY you can restrict the design to actual requirements.
Hi Mike,
Sorry, but yes. If you are sure of your load and operation you don't need SOA protection. As a DIYer, it's your own call (pro choice
) For a commercial (sold to others) product I fully support your opinion.
Having said that, there are many products without much protection at all. Many of them British over the years, so ....
-Chris (flame suit on)
Sorry, but yes. If you are sure of your load and operation you don't need SOA protection. As a DIYer, it's your own call (pro choice
) For a commercial (sold to others) product I fully support your opinion.Having said that, there are many products without much protection at all. Many of them British over the years, so ....
-Chris (flame suit on)
Hi Chris,
I don't like your answer but your absolutely right, protection is optional. Maybe that's why so many British brands no longer exist!
The only reason any designer might leave all protection out is to preserve the sound quality.
Good design will ensure that sound quality is not compromised.
Oh, there is another reason to omit - cheapskate!!
I don't like your answer but your absolutely right, protection is optional. Maybe that's why so many British brands no longer exist!
The only reason any designer might leave all protection out is to preserve the sound quality.
Good design will ensure that sound quality is not compromised.
Oh, there is another reason to omit - cheapskate!!
There are many simple protections available, that are cheap including fusible resistors which can replace individual Re on the outputs and 'PTC fuses or polyswitches' that can be put inside the feedback loop.
But not many amplifiers even have off cycle tailored SOA limiters, relying on pulse region. Why? Reactive loads are not suddenly pulsed as soon as Vout goes to the off half voltage.
My PA amplifiers had 2 slope SOAR limiters (slowed) in 1974 with MJ15003/4's and supply fuses. We had few problems in practice.
But not many amplifiers even have off cycle tailored SOA limiters, relying on pulse region. Why? Reactive loads are not suddenly pulsed as soon as Vout goes to the off half voltage.
My PA amplifiers had 2 slope SOAR limiters (slowed) in 1974 with MJ15003/4's and supply fuses. We had few problems in practice.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.