EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: kenw232 on April 04, 2014, 11:29:30 pm
-
Hi, I have an SCR (VS-40TPS12APBF-ND) that will be working with. The Ih value from the datasheet is 150mA. If I wanted the SCR to switch and turn on a current that will only be 75mA max, is that possible or not then? So the Ih value is 150mA to hold it on when I turn it on via the gate, it won't last and turn right off because 75mA is too low from what I would guess. If so can I force the SCR to continue conducting 75mA by keeping the voltage on the gate? So the applied gate voltage will override the failure to reach Ih then correct?
-
Hi, I have an SCR (VS-40TPS12APBF-ND) that will be working with. The Ih value from the datasheet is 150mA. If I wanted the SCR to switch and turn on a current that will only be 75mA max, is that possible or not then? So the Ih value is 150mA to hold it on when I turn it on via the gate, it won't last and turn right off because 75mA is too low from what I would guess. If so can I force the SCR to continue conducting 75mA by keeping the voltage on the gate? So the applied gate voltage will override the failure to reach Ih then correct?
The holding current is often spec'd as a max value to guarantee staying on. So below this value it *might* stay on, or might turn off when the gate drive is removed.
-
Why would you want to use a thyristor as a really really bad BJT?
A MOSFET or IGBT with the same voltage rating for the current you want to switch costs significantly less.
-
Why would you want to use a thyristor as a really really bad BJT?
A MOSFET or IGBT with the same voltage rating for the current you want to switch costs significantly less.
^^ At currents below I_h, an SCR behaves just as a crappy (slow, particularly nonlinear) BJT.
Tim
-
Why would you want to use a thyristor as a really really bad BJT?
A MOSFET or IGBT with the same voltage rating for the current you want to switch costs significantly less.
I actually am using a MOSFET, but my load required voltage isolation when the mosfet is in the off state - it can't have any voltage at all applied to the + side of the load when off. So I want to do this:
(http://sites.extremehosting.ca/trash/eg-highside.png)
But this requires I float the gate up to 1400V which is difficult. They don't make P-Channel mosfets above 600V so I'm out of luck there too. So I thought use an SCR to block or isolate the load and switch the SCR and mosfet on/off at the same time.
(http://sites.extremehosting.ca/trash/eg-highside-scr.png)
Any thoughts?
-
^^ At currents below I_h, an SCR behaves just as a crappy (slow, particularly nonlinear) BJT.
So there no way to force it open and thereby continue to conduct at a maximum by maintaining the voltage against the Gate?
-
Any thoughts?
Put the MOSFET above the load, put a pull up resistor on the gate and pull it down with a second MOSFET? (The switch will turn on if the control circuitry isn't powered but the high voltage supply is.)
-
Yours schematics are almost wrong and will never work.
Use a bipolar high voltage transistor as 2sc4913.
I hope you know that you are risking your life: 1400V 75mA is lethal.
To see your diagram, it seems to me that you do not have a level of knowledge in electronics and electricity to work with such high voltages ...
Be very careful, ask for help from someone qualified
-
I'm just trying to turn 1400V on and off. It could not be more difficult.
-
It could not be more difficult.
That's because you made it so.
It will help you greatly and potentially save you life and those around you if you just slow down a little bit and start learning basics on electronics, etc. before attempting to design something so lethal.
-
As mentioned in the other thread, please search for "high voltage high side switching". Either find a driver which is able to float at 1400V or buy a proven solution from a vendor like http://www.behlke.com (http://www.behlke.com).
-
If switching speed is not an issue, they make optically isolated MOSFET drivers which simply consist of an LED on one side and a photovoltaic pile on the other side which will generate the voltage necessary to fully turn on a MOSFET. The IRF application note includes an example which exactly matches what you are trying to do.
http://www.irf.com/product/_/N~1njcib#tab-tab1 (http://www.irf.com/product/_/N~1njcib#tab-tab1)
http://www.irf.com/product-info/datasheets/data/pvin.pdf (http://www.irf.com/product-info/datasheets/data/pvin.pdf)
http://www.irf.com/technical-info/appnotes/an-1017.pdf (http://www.irf.com/technical-info/appnotes/an-1017.pdf)
http://www.toshiba-components.com/couplers/photocoupler_photovoltaic.html (http://www.toshiba-components.com/couplers/photocoupler_photovoltaic.html)
-
If switching speed is not an issue, they make optically isolated MOSFET drivers which simply consist of an LED on one side and a photovoltaic pile on the other side which will generate the voltage necessary to fully turn on a MOSFET. The IRF application note includes an example which exactly matches what you are trying to do.
http://www.irf.com/product/_/N~1njcib#tab-tab1 (http://www.irf.com/product/_/N~1njcib#tab-tab1)
http://www.irf.com/product-info/datasheets/data/pvin.pdf (http://www.irf.com/product-info/datasheets/data/pvin.pdf)
http://www.irf.com/technical-info/appnotes/an-1017.pdf (http://www.irf.com/technical-info/appnotes/an-1017.pdf)
Thank you for this, it looks interesting. I've read a lot about high side switches, I just find it difficult to manage a second 1400V rail to float the gate.
These look like they would be great too but have been discontinued.
http://datasheet.eeworld.com.cn/pdf/120667_INTERSIL_MCT3D65P100F2.html (http://datasheet.eeworld.com.cn/pdf/120667_INTERSIL_MCT3D65P100F2.html)
-
We don't even know if the 0V of the 1400Vdc power supply is grounded or floating. |O
kenw232 think it's easy to interrupt 1400Vdc, but it's not.
A commun switch is not enough, and may not interrupt such a voltage (arcing between contacts).
Imagine what could happen:
:palm:"The switch is off, that's secure, there is no voltage." :palm:
but it is still with 1400Vdc, he touch something and he is dead
-
If switching speed is not an issue, they make optically isolated MOSFET drivers which simply consist of an LED on one side and a photovoltaic pile on the other side which will generate the voltage necessary to fully turn on a MOSFET. The IRF application note includes an example which exactly matches what you are trying to do.
http://www.irf.com/product/_/N~1njcib#tab-tab1 (http://www.irf.com/product/_/N~1njcib#tab-tab1)
http://www.irf.com/product-info/datasheets/data/pvin.pdf (http://www.irf.com/product-info/datasheets/data/pvin.pdf)
http://www.irf.com/technical-info/appnotes/an-1017.pdf (http://www.irf.com/technical-info/appnotes/an-1017.pdf)
Thank you for this, it looks interesting. I've read a lot about high side switches, I just find it difficult to manage a second 1400V rail to float the gate.
Common alternatives include high voltage direct level shifting, inductive coupling, and capacitive coupling. That last one is probably what I would do since it in inherently failsafe (If the input signal sticks then the MOSFET would turn off.) and does not require a custom transformer. An AC signal is used across a capacitive barrier to charge the MOSFET gate. DC restorers in CRT circuits work this way and handle thousands of volts with no problems.
These look like they would be great too but have been discontinued.
http://datasheet.eeworld.com.cn/pdf/120667_INTERSIL_MCT3D65P100F2.html (http://datasheet.eeworld.com.cn/pdf/120667_INTERSIL_MCT3D65P100F2.html)
I remember them distantly but did not know they had ever made it into production. I suspect insulated gate bipolar transistors took their wind.
-
The IRF application note includes an example which exactly matches what you are trying to do.
http://www.irf.com/product/_/N~1njcib#tab-tab1 (http://www.irf.com/product/_/N~1njcib#tab-tab1)
So these Photovoltaic Isolators seem to be self adjusting. They drive the gate automatically by dynamically floating the gate relative to the drain voltage applied in real time I think. I mean in the quick schematic below how does the PVI know what the + rail voltage is to float the gate up to? Must be smarter then the average IC.
(http://sites.extremehosting.ca/trash/piv.jpg)
-
Voltages in series add.
Cool, an MCT datasheet: I always thought they were unobtainium 4+ inch hockey puck devices used in power distribution only -- the only context I've seen them discussed in. I guess they really don't exist after all?
Tim
-
The secure solution is to feed the load by a separated 1400Vdc power supply (it's only 105W) and to switch the low voltage (230Vac or 110Vac or 24Vdc, ...and so on) feeding this power supply.
-
a photovoltaic pile on the other side which will generate the voltage necessary to fully turn on a MOSFET.
How fast can it charge up / down a hefty power mosfet? and what would be the power dissipation in that process?
-
The IRF application note includes an example which exactly matches what you are trying to do.
http://www.irf.com/product/_/N~1njcib#tab-tab1 (http://www.irf.com/product/_/N~1njcib#tab-tab1)
So these Photovoltaic Isolators seem to be self adjusting. They drive the gate automatically by dynamically floating the gate relative to the drain voltage applied in real time I think. I mean in the quick schematic below how does the PVI know what the + rail voltage is to float the gate up to? Must be smarter then the average IC.
(http://sites.extremehosting.ca/trash/piv.jpg)
It is not that tricky. Only the gate to source voltage of the MOSFET matters for turning it on. When the PVI is off, the gate to source voltage is zero, the MOSFET is off, the source it pulled to ground via the load, and there is 1400 volts from the drain to the source. When the PVI is on, the gate is pulled above the source by 5 to 10 volts. As the MOSFET turns on, the voltage from drain to source drops and the source voltage rises above ground but as it does so because the PVI output is floating, it brings the gate voltage along with it. The same thing would occur with any "suspended" gate drive scheme.
The PVI output however is low current so the MOSFET turn-on and turn-off times will be slow on the order of 10s to 100s of milliseconds which may be a virtue. For fast transition times, some other scheme is needed which can provide higher charge and discharge currents to the gate.
There actually *is* a way to do this with P-channel MOSFETs. A cascode could be used with multiple P-channel MOSFETs in series but it would still require a high voltage pull-down level shifter although *that* part could be a ground referred N-channel MOSFET.
-
The secure solution is to feed the load by a separated 1400Vdc power supply (it's only 105W) and to switch the low voltage (230Vac or 110Vac or 24Vdc, ...and so on) feeding this power supply.
That sounds far too reasonable and safe. Why do it the sensible way when you can live life on the edge instead? ;D
https://www.eevblog.com/forum/projects/mosfet-as-a-high-side-switch/msg417640/#msg417640 (https://www.eevblog.com/forum/projects/mosfet-as-a-high-side-switch/msg417640/#msg417640)