GeorgesWelding
Member level 3
Yeah I guess that the low side of the POT would essentially be connected to the mains rectifier's 0V rail. I can't have that in a case where the internals of the POT would short to the case, w/o a plastic knob on the POT's axle it would then transfer that voltage into the person grabbing the knob which would not be desirable.
Do you know of any type of optocoupler that is fast enough to run at 25KHz that I can place between the TL494 and the IR2110? If I can have isolation between those two devices then I can run them from separate power supplies with the supply for the TL494 grounded to the earth(allowing my POTs to be connected to earth) and the supply for the IR2110 grounded to the 0V rail of the switching bridge/mains like it needs to be.
I found an optocoupler that says it's capable of 2.5KV/uS switching, which would be 2.5V/nS correct? If that is true then that means it can switch a 15V logic signal in 6nS. It also says it can handle 5Mbit/s on a 20MHz clock. It says it has galvanic isolation with a 5300V isolation voltage. The optocoupler is a digital logic level isolator but the datasheet says its capable of running on up to 15V. It says it can supply 25mA of output current while the IR2110's logic input at 15V only draws ~18uA. I have an application paper that shows many ways of driving the LED with speed in mind. I think this optocoupler, if it's fast enough, would be a great solution to the problem of having the POTs connected to the mains. The optocoupler is the SFH6732 Dual Ultra High-Speed unit from Vishay.
Datasheet: https://www.vishay.com/docs/83685/sfh6731.pdf
Application Guide: **broken link removed**
If you have time, could you take a look at those papers and tell me if that optocoupler would be fast enough for interfacing between the TL494 and the IR2110?
- - - Updated - - -
Thanks for all of the other info you've provided me as well. You are an absolute savior in helping me make this welder a possibility.
- - - Updated - - -
About the separate supply for the high side of the IR2110. I was running an isolated supply for the high-side drive but since then I've gotten it to work with the bootstrapping method of supplying it. I didn't have a big enough CAP across VCC/COM pins, or it wasn't close enough to the IC and the voltage to the VCC pin was dropping and screwing up the output signal to the low-side IGBT/FET. Since then I've dropped the size of the boot-strap cap to 1uF and the VCC/COM cap to 4.7uF. I did study the datasheet/application guide more and found that those caps should be placed directly to the IR2110's pins and the VCC/COM cap should be ~10x the size of the bootstrap cap. I'm going to make the BS cap even smaller, I'm going to go with a 470nF cap so that the 4.7uF cap across the VCC/COM pins is 10x the size yet the physical size of these caps will still be small as I'm using tiny yellow ceramic caps.
Do you know of any type of optocoupler that is fast enough to run at 25KHz that I can place between the TL494 and the IR2110? If I can have isolation between those two devices then I can run them from separate power supplies with the supply for the TL494 grounded to the earth(allowing my POTs to be connected to earth) and the supply for the IR2110 grounded to the 0V rail of the switching bridge/mains like it needs to be.
I found an optocoupler that says it's capable of 2.5KV/uS switching, which would be 2.5V/nS correct? If that is true then that means it can switch a 15V logic signal in 6nS. It also says it can handle 5Mbit/s on a 20MHz clock. It says it has galvanic isolation with a 5300V isolation voltage. The optocoupler is a digital logic level isolator but the datasheet says its capable of running on up to 15V. It says it can supply 25mA of output current while the IR2110's logic input at 15V only draws ~18uA. I have an application paper that shows many ways of driving the LED with speed in mind. I think this optocoupler, if it's fast enough, would be a great solution to the problem of having the POTs connected to the mains. The optocoupler is the SFH6732 Dual Ultra High-Speed unit from Vishay.
Datasheet: https://www.vishay.com/docs/83685/sfh6731.pdf
Application Guide: **broken link removed**
If you have time, could you take a look at those papers and tell me if that optocoupler would be fast enough for interfacing between the TL494 and the IR2110?
- - - Updated - - -
Thanks for all of the other info you've provided me as well. You are an absolute savior in helping me make this welder a possibility.
- - - Updated - - -
About the separate supply for the high side of the IR2110. I was running an isolated supply for the high-side drive but since then I've gotten it to work with the bootstrapping method of supplying it. I didn't have a big enough CAP across VCC/COM pins, or it wasn't close enough to the IC and the voltage to the VCC pin was dropping and screwing up the output signal to the low-side IGBT/FET. Since then I've dropped the size of the boot-strap cap to 1uF and the VCC/COM cap to 4.7uF. I did study the datasheet/application guide more and found that those caps should be placed directly to the IR2110's pins and the VCC/COM cap should be ~10x the size of the bootstrap cap. I'm going to make the BS cap even smaller, I'm going to go with a 470nF cap so that the 4.7uF cap across the VCC/COM pins is 10x the size yet the physical size of these caps will still be small as I'm using tiny yellow ceramic caps.