I am using this PS regulator daily in my TT amp, no plops on switch on or off.
The speaker hot line does not switch, just the power lines simultaneously switch off in case of overload or DC on the power amp output.
P.S. One more feature, if one power line polarity sag under load, other line polarity sag simultaneously.
Last edited:
I think it's a good idea to clarify something here, lest we risk confusing future readers. If your amp plops on turn-on before you add any type of relay in the rails, it's going to plop afterwards. If it doesn't plop before, it probably won't plop just because you switch the rails hard on with a SSR.
In my case, my amp plops because of design considerations, it's not something easily "fixed", it was a tradeoff at design-time, and i'm inclined to accept that. Switching the speaker lines has the potential to "fix" the plop in that it isn't audible any more, something a relay in the rails couldn't.
In my case, my amp plops because of design considerations, it's not something easily "fixed", it was a tradeoff at design-time, and i'm inclined to accept that. Switching the speaker lines has the potential to "fix" the plop in that it isn't audible any more, something a relay in the rails couldn't.
Why does your amp do this, what was your trade-off?
Not an output capacitor surely?
Would it help to power up the front end first and then switch the power to the output power transistors?
The front end could tap power before the power rail MOSFETs and be easily protected some other way. Switch the MOSFETs on after the front end stabilizes. Should be simple to do.
Best wishes
David
has anyone found a cheap way to trigger the mosFETs off?
All the photovoltaic driver chips seem to be too expensive.
Only Mouser can search and find that term.
The isolated transformer driver by Janneman (is it Jan.didden) is not cheap either, particularly PCB estate.
BTW,
I want to try one in each supply rail and one in the speaker return line.
All the photovoltaic driver chips seem to be too expensive.
Only Mouser can search and find that term.
The isolated transformer driver by Janneman (is it Jan.didden) is not cheap either, particularly PCB estate.
BTW,
I want to try one in each supply rail and one in the speaker return line.
Last edited:
If you want to switch off the supply rails you don't need to switch off the speaker line, just don't put to big caps after the mosfets. To switch off the supply rails you don't need a photovoltaic chip or an isolating transformer.
Can you show me how?
BTW, I want to use N channel mosFET on the +ve Rail, as well as the -ve rail.
Maybe this could help you. If you want the same polarity mosfet in both rail you have to change it yourself, as I am going on ten days vacation and no time to do that.
It's simple using N-channel MOSFET to switch the -Ve, not so to switch +Ve as you will need a boosted supply that rides about 10V above +Ve to provide the needed Vgs as the source of the MOSFET is at +Ve potential.
Just look at the 02 headphone amp Andrew. That uses switched rails driven via a comparator, so both rails switch together and cleanly. I'm sure you could just use resistive/zener level changing for the gate drive to allow control of higher voltage rails while retaining the control circuitry running on low voltage. It does need complementary FET though.
The issue is not about clean switching, but topology and how it behaves during power up and down. Balanced amps are better in this regard compared to blameless types.
If you have power rail decouplers after the switches, and the front end does not have full DC control during power up and power down, you are likely to get large offsets as the amp powers up or down, leading to plops.
One way to mitigate this is to use a DC offset servo.
If you have power rail decouplers after the switches, and the front end does not have full DC control during power up and power down, you are likely to get large offsets as the amp powers up or down, leading to plops.
One way to mitigate this is to use a DC offset servo.
Last edited:
That does not help.
It uses Pchannel on the +ve rail.
It uses a transistor off the input to trigger.
I want to use heatsink over temperature and/or output excess current and/or output excess IV and/or output excess DC to trigger and maybe a time delay.
Maybe after you come back.
Last edited:
That's why the photovoltaic driver is so good.
It is isolated and it provides it's own drive voltage.
Thanks for your time.
Last edited:
Andrew,
I suggest you use the following concept. All the fancy stuff goes in the boxes labelled X3 and X4. The vom1271 will be a bit slow to turn the mosfets on (10's of mSec), but will turn off very fast (a few uSec). Run them at 20 to 25mA LED current.
Note! If you use a speaker offset detector, then it must latch on! If it doesn't latch on, then the offset turns the rail switches off, the offset disappears, then the rail switches turn on and dump the full rail voltage onto the (faulty) amplifier. This will continue till the amp and/or speakers fail. An opto triac is useful here.
Paul Bysouth
I suggest you use the following concept. All the fancy stuff goes in the boxes labelled X3 and X4. The vom1271 will be a bit slow to turn the mosfets on (10's of mSec), but will turn off very fast (a few uSec). Run them at 20 to 25mA LED current.
Note! If you use a speaker offset detector, then it must latch on! If it doesn't latch on, then the offset turns the rail switches off, the offset disappears, then the rail switches turn on and dump the full rail voltage onto the (faulty) amplifier. This will continue till the amp and/or speakers fail. An opto triac is useful here.
Paul Bysouth
Attachments
Thanks again Paul.
The wiring/interconnection of the modules looks similar to what I imagine I want to do. Except I plan to put the speaker relay in the Return route and avoid on photovoltaic.
And yes I agree on latching. two transistors is all it takes.
Fitted by Crimson in my amp modules from way back and in a more recent Cyrus output detector.
The Cyrus version switches off at crossover.
But you are using Photovoltaic drivers.
Why the transorb across the speaker relay?
The wiring/interconnection of the modules looks similar to what I imagine I want to do. Except I plan to put the speaker relay in the Return route and avoid on photovoltaic.
And yes I agree on latching. two transistors is all it takes.
Fitted by Crimson in my amp modules from way back and in a more recent Cyrus output detector.
The Cyrus version switches off at crossover.
But you are using Photovoltaic drivers.
Why the transorb across the speaker relay?
Last edited:
Andrew,
Why not use the photo-voltaic drivers? The vom1271's are cheap ($2.50 aprox), available (Digikey have 22,678 at the moment, Mouser has 5,152), and save a large amount of PCB space. Also they are not likely to become obsolete in the near future.
The tranzsorb is to safeguard against exceeding the breakdown voltage of the mosfets - probably not really required unless using mosfets that have a breakdown voltage close to the rail to rail voltage.
Paul Bysouth
Why not use the photo-voltaic drivers? The vom1271's are cheap ($2.50 aprox), available (Digikey have 22,678 at the moment, Mouser has 5,152), and save a large amount of PCB space. Also they are not likely to become obsolete in the near future.
The tranzsorb is to safeguard against exceeding the breakdown voltage of the mosfets - probably not really required unless using mosfets that have a breakdown voltage close to the rail to rail voltage.
Paul Bysouth
I have never used photo voltaic couplers.
Are these suitable, I can source them easily. Moose and Digikey is sadly not an option for me.
I'm sure about turn off of this one
http://docs-europe.electrocomponents.com/webdocs/0af4/0900766b80af4d0a.pdf
Regards
Are these suitable, I can source them easily. Moose and Digikey is sadly not an option for me.
I'm sure about turn off of this one
http://docs-europe.electrocomponents.com/webdocs/0af4/0900766b80af4d0a.pdf
Regards
Last edited:
Vostro,
The Toshiba TLP3914 opto appears to be reasonable, but would require an external turn off circuit. See diagram for how to do this. I did a simple model to test the circuit with PSMN3r8-100bs mosfets (which have very low Rds-on, but very high input capacitance), and it appears to work as it should.
Paul Bysouth
The Toshiba TLP3914 opto appears to be reasonable, but would require an external turn off circuit. See diagram for how to do this. I did a simple model to test the circuit with PSMN3r8-100bs mosfets (which have very low Rds-on, but very high input capacitance), and it appears to work as it should.
Paul Bysouth
