Member
Joined 2009
Paid Member
Member
Joined 2009
Paid Member
Agree. Using a micro allows you to do everything needed to protect an amplifier. Except of course it cannot protect relays when hey are called upon to switch high current at high voltage. That's a job for a solid state switch.
Hi,
No-body prevents you from implementing an equivalent solution using two relays... You can use a big bruiser automotive relay designed to interrupt routinely massive currents at 24V for the big relay and a second smaller gold on silver signal relay.
You need a wee bit extra delay for the signal relay (well, not really, you should only switch after there is neither DC nor signal) and on release the smaller signal relay usually releases much faster, so no issues there...
Or just use the relay as crowbar... I do it in my commercial Amp's, with minimal failures or course, as soon as DC Protection kicks in I also drop the mains power supply out...
Ciao T
Well, I just went to Amplimo to order my 3 relays only to discover that their shipping costs are higher than the value of the 3 relays and these aren't the cheapest relays. No wonder people in Canada prefer to buy local/US. I'll be looking at those MOSFETs now
No-body prevents you from implementing an equivalent solution using two relays... You can use a big bruiser automotive relay designed to interrupt routinely massive currents at 24V for the big relay and a second smaller gold on silver signal relay.
You need a wee bit extra delay for the signal relay (well, not really, you should only switch after there is neither DC nor signal) and on release the smaller signal relay usually releases much faster, so no issues there...
Or just use the relay as crowbar... I do it in my commercial Amp's, with minimal failures or course, as soon as DC Protection kicks in I also drop the mains power supply out...
Ciao T
Member
Joined 2009
Paid Member
How about this baby, less than $2 and you get a pair of N-FETs in a flat pack,
BSC750N10ND - is it tough enough for the job at hand ?
add one of these... TLP190B and the cost of the 'relay' is $5
Am I looking at the right parts to make this work ?
The diagram is from Geofex, do you think its omitting key components or is it really this simple ?
I'll be using a 'C1237' protection i.c. from fleabay to drive the 'relay's
BSC750N10ND - is it tough enough for the job at hand ?
add one of these... TLP190B and the cost of the 'relay' is $5
Am I looking at the right parts to make this work ?
The diagram is from Geofex, do you think its omitting key components or is it really this simple ?
I'll be using a 'C1237' protection i.c. from fleabay to drive the 'relay's
Attachments
Last edited:
The dual FET wouldn't be my choice tbh even though it is space saving. Are you OK working (soldering) that package (PG-TSDSON).
Compared to an IRF2907 which I have used the dual has a very high on resistance (62mΩ vs 3.5mΩ). For normal operation that has to be one of the most important parameters for anything other than a very low powered amp.
Tha max current ratings, in particular peak currents and "maximum pulsed energy" currents are way way down. These are very important if the FET has to handle a fault condition and break the resulting current flow.
That really is all that's needed as shown on the diagram. The photovoltaic coupler drives the gates directly.
Compared to an IRF2907 which I have used the dual has a very high on resistance (62mΩ vs 3.5mΩ). For normal operation that has to be one of the most important parameters for anything other than a very low powered amp.
Tha max current ratings, in particular peak currents and "maximum pulsed energy" currents are way way down. These are very important if the FET has to handle a fault condition and break the resulting current flow.
That really is all that's needed as shown on the diagram. The photovoltaic coupler drives the gates directly.
In general, you are seeking for a MOSFET with very high Vdsmax, very low Rdson and very high Idmax. Regarding Vdsmax, it must be higher than the sum of positive and negative supply rail voltage, and the output should be provided by clamping diodes connected to supply rails, reverse polarized. My tips are
Type Vdsmax Rdson Idmax
IRFB1227 200V 0.0197ohm 130A
IRFP90N20D 200V 0.023ohm 94A
IRF3710ZG 100V 0.018ohm 59A (too low Vdsmax to me)
IRFS59N10D 100V 0.025ohm 59A (too low Vdsmax to me)
The BSC750N10ND part has Vdsmax 100V (quite low), Rdson 0.075ohm (quite high) and Idmax 13A (very low). This would be very unreliable solution to use this part. I also warn against strange manufacturers that are offering this part. You have to count with much higher price than 2 USD per part, in case you require quality.
Good question.
The turn off time for the TLP190B that Bigun mentions is quoted as a 1 millisecond.
The Avago ASSR-V621 that I used specifically has an "active turn off circuit" and is quoted at 0.03 millisecond off and 0.3 millisecond turn on. Substantially better.
I would say even 1ms is quick enough though, particularly when you factor in a conventional DC offset integrator detect circuit.
The turn off time for the TLP190B that Bigun mentions is quoted as a 1 millisecond.
The Avago ASSR-V621 that I used specifically has an "active turn off circuit" and is quoted at 0.03 millisecond off and 0.3 millisecond turn on. Substantially better.
I would say even 1ms is quick enough though, particularly when you factor in a conventional DC offset integrator detect circuit.
Further to my above post.
I think I see what you are getting at Andrew.
I would refer you back to here,
http://www.diyaudio.com/forums/solid-state/191449-output-relays-14.html#post2661555
As for the TLP190. That could be considerably slower if there were significant G-S capacitance.