Hi Jonathan,
Yes, that will work. I've seen it done with a triac. Basically, anything that removes the energy from the load (speaker) is what is needed. A shunt to ground works for me.
Some amps will go into oscillation with the feedback removed and so this ensures the destruction of that type of amp. Still cheaper than good speakers.
-Chris
Yes, that will work. I've seen it done with a triac. Basically, anything that removes the energy from the load (speaker) is what is needed. A shunt to ground works for me.
Some amps will go into oscillation with the feedback removed and so this ensures the destruction of that type of amp. Still cheaper than good speakers.
-Chris
If that is sufficient to protect the speakers, it would be a much better option if the relay that is triggered by a dc sensing circuit connects the output of the amp to ground in case of fault. That way you have two advantages: no relay in the signal path and if the relay were to be welded by high currents, it would still be protecting the the speaker. Why haven't I seen circuits like that among the available schematics?anatech said:
I'm still stuck on relays becuase I don't have the faintest idea how to make a protection circuit with a mosfet. If any of the knowledgeanle guys could share a circuit with some point and hints how to adapt it for another situation, that would be purrfect.
Hans,
Use a triac to ground from the speaker output. Fire the triac in a fault condition. You need to watch for dv/dt tripping the triac. I have seen this in some really big pro amps and it works. Can't remember the name but I recall the destruction on the output PCB.
-Chris
Can we get another post for echowars please? He wants 50 posts.
Use a triac to ground from the speaker output. Fire the triac in a fault condition. You need to watch for dv/dt tripping the triac. I have seen this in some really big pro amps and it works. Can't remember the name but I recall the destruction on the output PCB.
-Chris
Can we get another post for echowars please? He wants 50 posts.
Hi EchoWars,
I take my feedback from the output terminal, and boldly pull and replace the fuse in the loop. Of course, the feedback is only around 7dB and has nothing to do with DC stability.
With higher feedback and the DC servo, I'm sure the spkrs would go bang and the contacts would pit.
Can we watch ?
-Chris
I take my feedback from the output terminal, and boldly pull and replace the fuse in the loop. Of course, the feedback is only around 7dB and has nothing to do with DC stability.
With higher feedback and the DC servo, I'm sure the spkrs would go bang and the contacts would pit.
Can we watch ?
-Chris
R.G. said:
On investigation, this scheme really is not very practical if your MOSFETs have 1 ohm Rds(on)....(most do at operating temps) as this means they will dissipate 400W(!!!) for 20A from the supply...This sort of current level is perfectly normal with some loads....
Serious heatsinking is therefore required for your MOSFETs....Not good.
Thanks for the update Mike. I did a little investigation and found that by the time you sorted the gate drive circuitry and sensing logic it added complexity, board estate and cost far beyond that offered by relays, so I binned the idea.
At the end of the day it still leaves you with the onboard bypass caps to discharge into the load which I am also not happy with.
At the end of the day it still leaves you with the onboard bypass caps to discharge into the load which I am also not happy with.
Hi pinkmouse,
The relay used depends on the output power. Suitable relays would be in the 5 to 10 or 20A range. They normally do not switch hot current if the volume is down when you power up or down. We were taught that in the early seventies as stereo salesmen. Those were the days when salesmen actually knew the product.
When the power gets up there you may be further ahead with a triac to ground across the output. This eliminates the contact for the purists and can be used at lower power too. BTW, a relay has one contact, a fuse has two plus a non-linear resistive element.
The Carver PM 2.0 uses a high frequency switching supply and electronic shut down of same. This works well also as the current in the AC leg is lower than the rectified DC path.
Adcom amplifiers killed the current source for the diff pair and the vas. That works unless an output turns into a piece of wire. They had a backup fuse in the output.
It doesn't matter what method is used as long as there is some way to disconnect the load from a fault condition. Coupling the output disconnect with an input mute is a good idea. Small relay to ground is best I feel here.
What do you think?
-Chris
The relay used depends on the output power. Suitable relays would be in the 5 to 10 or 20A range. They normally do not switch hot current if the volume is down when you power up or down. We were taught that in the early seventies as stereo salesmen. Those were the days when salesmen actually knew the product.
When the power gets up there you may be further ahead with a triac to ground across the output. This eliminates the contact for the purists and can be used at lower power too. BTW, a relay has one contact, a fuse has two plus a non-linear resistive element.
The Carver PM 2.0 uses a high frequency switching supply and electronic shut down of same. This works well also as the current in the AC leg is lower than the rectified DC path.
Adcom amplifiers killed the current source for the diff pair and the vas. That works unless an output turns into a piece of wire. They had a backup fuse in the output.
It doesn't matter what method is used as long as there is some way to disconnect the load from a fault condition. Coupling the output disconnect with an input mute is a good idea. Small relay to ground is best I feel here.
What do you think?
-Chris
" Contact purists "
Correct relay for this application have nominal current cca 20 - 30 A of AC or DC current and contact resistance after 10 000 cycles of switch on - off maximal 100 mOhm ( Finder 65 ..., Matsushita / Nais TV 15 ), but normal value will be probably much lower. Is funny, if " contact purists " argue with this resistance, while their boxes are connected by cabels with higher resistance than resistance of contacts.
Correct relay for this application have nominal current cca 20 - 30 A of AC or DC current and contact resistance after 10 000 cycles of switch on - off maximal 100 mOhm ( Finder 65 ..., Matsushita / Nais TV 15 ), but normal value will be probably much lower. Is funny, if " contact purists " argue with this resistance, while their boxes are connected by cabels with higher resistance than resistance of contacts.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.