Unfortunately I only have 8 pin optocouplers. I can include them in my next order, but this will take some time.
With a 4 pin optocoupler there is also a vcc needed in the optocouplers output right? Since it needs a power source to power the optocouplers isolated output. If I'm not mistaken, this is only possible with a power source referenced to secondary GND, which is generated after the transformers/D22, which probably is delayed compared to the remote cable.
After inspecting the schematics, assuming that Q52 is driven by Q51 via the remote, maybe this could be an idea.
Isn't it possible to remove C82 (maybe D51 and R56 as well)? And delay the gate from Q51 during startup (with a capacitor via a transistor), and disable the transistor after remote is removed.
This will delay the startup, and switch off as good as immediately when removing remote.
Since Q51 could be switched off quicker than the Jfet on Q52.
Or is Q52 driven by the SD/RAIL voltage difference and am I totally misunderstanding the Q52 circuit.
With a 4 pin optocoupler there is also a vcc needed in the optocouplers output right? Since it needs a power source to power the optocouplers isolated output. If I'm not mistaken, this is only possible with a power source referenced to secondary GND, which is generated after the transformers/D22, which probably is delayed compared to the remote cable.
After inspecting the schematics, assuming that Q52 is driven by Q51 via the remote, maybe this could be an idea.
Isn't it possible to remove C82 (maybe D51 and R56 as well)? And delay the gate from Q51 during startup (with a capacitor via a transistor), and disable the transistor after remote is removed.
This will delay the startup, and switch off as good as immediately when removing remote.
Since Q51 could be switched off quicker than the Jfet on Q52.
Or is Q52 driven by the SD/RAIL voltage difference and am I totally misunderstanding the Q52 circuit.
Attachments
Last edited:
For a 4-pin optocoupler:
Try inserting the transistor side of the optocoupler between the 4148 and the resistor driving the gate of the jfet.
Power the diode side of the optocoupler with the remote but you'll need a series resistor to prevent damaging the internal LED.
Do you have any junk amps that you could pull the optocoupler out of? The amps like the one below have 8-pin optocouplers that are actually two 4 pin optocouplers in one 8-pin package.
Try inserting the transistor side of the optocoupler between the 4148 and the resistor driving the gate of the jfet.
Power the diode side of the optocoupler with the remote but you'll need a series resistor to prevent damaging the internal LED.
Do you have any junk amps that you could pull the optocoupler out of? The amps like the one below have 8-pin optocouplers that are actually two 4 pin optocouplers in one 8-pin package.
Connected the PC817.
And it works absolutely fantastic!
Not a single drop of rail voltage during power off. It also waits with oscillating until max rail voltage is reached.
Very strange, since I would expect the optocoupler creates an even bigger delay and would not be faster since the optocouplers transistor VCC comes from the same source as before (D22).
Anyway, would it be safe to remove the 15v rails and solder the rectifiers back in place?
Amplifier seems to behave correct currently with both powering on and off. It also produces clean audio.
And it works absolutely fantastic!
Not a single drop of rail voltage during power off. It also waits with oscillating until max rail voltage is reached.
Very strange, since I would expect the optocoupler creates an even bigger delay and would not be faster since the optocouplers transistor VCC comes from the same source as before (D22).
Anyway, would it be safe to remove the 15v rails and solder the rectifiers back in place?
Amplifier seems to behave correct currently with both powering on and off. It also produces clean audio.
I don't know if it's safe to go to full rail voltage but you'll have to try it at some point.
The remote is switching off before the rest of the amp can respond so, even if the optocoupler is adding some slight delay, it's a better point to switch from.
There have been other 21844 substitutes that have been mentioned in other threads. Have you tried any others?
Do you still have the increased turn-on delay that you got by replacing the resistor/Zener/capacitor in the delay circuit?
The remote is switching off before the rest of the amp can respond so, even if the optocoupler is adding some slight delay, it's a better point to switch from.
There have been other 21844 substitutes that have been mentioned in other threads. Have you tried any others?
Do you still have the increased turn-on delay that you got by replacing the resistor/Zener/capacitor in the delay circuit?
I have seen substitutes in other threads like the FAN7393AMX, but those were also not always working properly with the same reason.
In soms amps they work, in some amps they don't work and blow out fets as well.
An other repair tech order the same IX21844N at the same time. He had no problems with the IX21844N and worked properly directly.
So I suppose the IX21844N and FAN7393 works properly in some amps, but need an adjustment in some other amps.
Yes, the delay is still present. The delay works nice with powering ON.
In soms amps they work, in some amps they don't work and blow out fets as well.
An other repair tech order the same IX21844N at the same time. He had no problems with the IX21844N and worked properly directly.
So I suppose the IX21844N and FAN7393 works properly in some amps, but need an adjustment in some other amps.
Yes, the delay is still present. The delay works nice with powering ON.
Removed the 15v rails.
Installed new output fets in the whole output.
Works absolutely perfect, very clean output fet square waves as well.
Waits until 120v+ rail is reached when rising slowly with only 8amps of input and dumps no voltage when powering off.
Thank you very much for the help Perry
Installed new output fets in the whole output.
Works absolutely perfect, very clean output fet square waves as well.
Waits until 120v+ rail is reached when rising slowly with only 8amps of input and dumps no voltage when powering off.
Thank you very much for the help Perry
If this problem applies to other amps and the 21844s (or other workable ICs) aren't available someone could easily produce a small board for this if a neat/easy way of wiring these components in isn't feasible.
Using the optocoupler likely allows you to be able to source the drive for the jfet from the input of the 12v regulator (cathode of D22), eliminating the 1n4148.
The added delay may also be helpful for other amps.
Using the optocoupler likely allows you to be able to source the drive for the jfet from the input of the 12v regulator (cathode of D22), eliminating the 1n4148.
The added delay may also be helpful for other amps.