Dear fellows,
Lately protection solutions fill my mind and it is my impression that it is excessive current that kills our beloved trannies. Since is current faster than heat, it should be the agressor to be contained.
After a number of different methods I came up with this one:
The goal is to be able to cut off the current to the power amp in a matter of microseconds, which I'm probably not quite achieving yet
.
The LTC6101HV outputs a small voltage which will be compared to a reference of about 140mV (14A). When Vy exceeds Vx the LM393's output goes low, shutting down the current path to the amplifier. With the output going low, Q3 pulls up Vy to 17V, latching the output low and keeping the current path shut. Q4 disables the latching during the first millisecond to avoid triggering during startup. This latching delay is probably the main catch since in 1ms alot of damage can be done.
Please note that the LTC6101HV and VOM1271 will be used inreallife but are not included in simulation as they are not in the libraries.
Also the VOM1271 is not the fastest but friend of mine got a pair. What are your favorite opto-isolated mosfet relay drivers? I might accept using a battery for more Vgs. Any suggestions on the big relay mosfets are really welcome as I'm not that familiar with them yet(learning student here). And of course any other thoughts and opinions for that matter. After all, I'm posting this here since I'm interested in your opinions.
Much thanks and much cheers,
Ruben
Lately protection solutions fill my mind and it is my impression that it is excessive current that kills our beloved trannies. Since is current faster than heat, it should be the agressor to be contained.
After a number of different methods I came up with this one:
The goal is to be able to cut off the current to the power amp in a matter of microseconds, which I'm probably not quite achieving yet
.The LTC6101HV outputs a small voltage which will be compared to a reference of about 140mV (14A). When Vy exceeds Vx the LM393's output goes low, shutting down the current path to the amplifier. With the output going low, Q3 pulls up Vy to 17V, latching the output low and keeping the current path shut. Q4 disables the latching during the first millisecond to avoid triggering during startup. This latching delay is probably the main catch since in 1ms alot of damage can be done.
Please note that the LTC6101HV and VOM1271 will be used inreallife but are not included in simulation as they are not in the libraries.
Also the VOM1271 is not the fastest but friend of mine got a pair. What are your favorite opto-isolated mosfet relay drivers? I might accept using a battery for more Vgs. Any suggestions on the big relay mosfets are really welcome as I'm not that familiar with them yet(learning student here). And of course any other thoughts and opinions for that matter. After all, I'm posting this here since I'm interested in your opinions.
Much thanks and much cheers,
Ruben
Last edited:
vom1271
Ruben,
I have used lots of vom1271s, generally with the PSMN5r6 To220 mosfet as a `speaker relay'.
I've attached a model and symbol for the vom1271, and also a datasheet and model for the PSMN5r6 mosfets.
The LTC6101 has a model, in the `opamps' section of LTspice.
I normally run vom1271's at 20mA of current, or sometimes initially at 25mA and then let it drop to 15mA. The current doesn't alter the output voltage much (it will be about 8 to 9 volts which is plenty), but the higher currents reduce the source resistance considerably. If you have a dual vom1271, then put the LEDs in series and the outputs in parallel. The vom1271 is a bit slow to turn mosfets on, but very fast to turn them off (which is what you want).
Regards,
Paul Bysouth
Ruben,
I have used lots of vom1271s, generally with the PSMN5r6 To220 mosfet as a `speaker relay'.
I've attached a model and symbol for the vom1271, and also a datasheet and model for the PSMN5r6 mosfets.
The LTC6101 has a model, in the `opamps' section of LTspice.
I normally run vom1271's at 20mA of current, or sometimes initially at 25mA and then let it drop to 15mA. The current doesn't alter the output voltage much (it will be about 8 to 9 volts which is plenty), but the higher currents reduce the source resistance considerably. If you have a dual vom1271, then put the LEDs in series and the outputs in parallel. The vom1271 is a bit slow to turn mosfets on, but very fast to turn them off (which is what you want).
Regards,
Paul Bysouth
Thank you for the input.
I don't see how cutting of the current might cause a raise of voltage. Do you mean this circuit or the amplifier? However, in simulation I did notice that when loadcurrent is shorted completely, I_load= 2kA, the voltage over vom1271's diodes will be >700V. I assumed this won't happen working with not ideal voltage sources.
Could you perhaps elaborate more?
cheers!
Si8752 very much noted. Thanks! IV circuits nicely apply to single output pairs, though the amp in question has 4 pairs, complicating this matter.
Youre right! didnt expect to find it there. Thanks for adding the models, great help.
Much cheers to you mate!
The standard method of protecting the output transistors against inductive kickback is to put free wheeling diodes in parallel with the output transistors. This clamps it to the supply voltage plus Vf of the diode. The diodes don’t need to be anything special. The kickback is of limited energy (whatever is stored in the speaker VC), and once activated, you’re already going to hear distortion. Even the old bog standard VI limiters need these diodes, for the same reason.