Set the var under the vertical amplifier to the straight-up position (to cal). Also set the var for the timebase to cal. Leave those set that way unless you have a very good reason to change them.
Set the timebase to 10us and the vertical amplifier to 5v/div. Set the trace to align to the reference line with the coupling set to gnd. Go back to DC coupling and repost the waveform.
Is the B+ voltage constant when the waveform becomes unstable?
Set the timebase to 10us and the vertical amplifier to 5v/div. Set the trace to align to the reference line with the coupling set to gnd. Go back to DC coupling and repost the waveform.
Is the B+ voltage constant when the waveform becomes unstable?
Check the voltage at pin 4 of the TL494 IC in the PSU. These Chinese know zero about how to control dead time. Even though the TL494 has a built in dead time it is NOT sufficient to control the switching of the MOSFETs in the power supply due to them knowing nothing about grounding and Kelvin connections.
If the voltage at pin 4 of the TL494 is less than 200mV I suggest to remove the resistor to ground from pin 4 and insert a 22K in it's place. Then hard wire a 470K from pin 4 to pins 12/13 of the TL494 and parallel it with a 100mfd 10v for a slow start up.
This 220mV dead time voltage helps to mitigate the bad PCB layout.
A similar issue occurs with their crude class D design as they typically do not have any dead time control and that is why your idle current is so high.
Stephen Mantz
Zed Audio
If the voltage at pin 4 of the TL494 is less than 200mV I suggest to remove the resistor to ground from pin 4 and insert a 22K in it's place. Then hard wire a 470K from pin 4 to pins 12/13 of the TL494 and parallel it with a 100mfd 10v for a slow start up.
This 220mV dead time voltage helps to mitigate the bad PCB layout.
A similar issue occurs with their crude class D design as they typically do not have any dead time control and that is why your idle current is so high.
Stephen Mantz
Zed Audio
Made those adjustments and also set the probe to x1. Thank you for sorting that out.
No, the B+ voltage isn't rock steady once we get over 11.5V on B+ it starts fluctuating ~ .25V as its current draw starts bouncing. This ~.25V swing is consistent all the up to 14.5V on B+.
No, the B+ voltage isn't rock steady once we get over 11.5V on B+ it starts fluctuating ~ .25V as its current draw starts bouncing. This ~.25V swing is consistent all the up to 14.5V on B+.
Attachments
If there is no pulse-width control, the problem may be that the driver IC is slightly less efficient than the original. I've seen the UCC27524 and the MC33152 used. You may want to try those. Maybe someone here knows, definitively, the original part number of the defaced IC.
If you post a photo of the input of the 4427 and the output of the MCU, that may be able to tell you something about the problem with the deadtime.
If you post a photo of the input of the 4427 and the output of the MCU, that may be able to tell you something about the problem with the deadtime.
OK, I will check those two points tomorrow.
I do not have the other two driver IC's, but I need to place an order for other items and will add both to the order.
I do not have the other two driver IC's, but I need to place an order for other items and will add both to the order.
There's no guarantee that there will be a difference but I'd give those a try if no one knows the original driver.
It may also be possible to make enough of a difference in the drive by parallelling a 1N4148 across the series resistors between the MCU and the driver IC inputs (striped end of diode to the MCU). This would be temporary.
It may also be possible to make enough of a difference in the drive by parallelling a 1N4148 across the series resistors between the MCU and the driver IC inputs (striped end of diode to the MCU). This would be temporary.
Following up on this thread, I received additional information via PM on these amplifiers, thank you for that advice.
I placed an order for different parts and have installed those now.
On amplifier A, I returned the gate resistors to 4.7 ohms, replaced the MIC4427 with UCC27524 and also replaced the eight IRF1404 with CSD18502KCS.
The amplifier now has a consistent idle current draw of ~1.7A from 10.5V up to 14.5V
Moving to amplifier B, it already has 4.7 ohm gate resistors and IRF1404 MOSFETS. I swapped out the MIC4427 for the UCC27524 and tested that configuration. It also has consistent idle current of ~1.8A from 10.5V to 14.5V. So, I think that proves the MIC4427's I have, are having trouble driving the eight MOSFETS.
I placed an order for different parts and have installed those now.
On amplifier A, I returned the gate resistors to 4.7 ohms, replaced the MIC4427 with UCC27524 and also replaced the eight IRF1404 with CSD18502KCS.
The amplifier now has a consistent idle current draw of ~1.7A from 10.5V up to 14.5V
Moving to amplifier B, it already has 4.7 ohm gate resistors and IRF1404 MOSFETS. I swapped out the MIC4427 for the UCC27524 and tested that configuration. It also has consistent idle current of ~1.8A from 10.5V to 14.5V. So, I think that proves the MIC4427's I have, are having trouble driving the eight MOSFETS.