Hello,
I need help in finding a device that can comfortably switch a high voltage (upwards to 30kv) and relatively high current (5 amps) at upwards of 120 Hz, for very short periods of time (enough to make a spark much like a distributor does). I thought about a relay but the contacts would surely wear out quite fast. Solid state seems pretty hopeless as well, I don't think it could handle the high voltage. The only thing I can think of would be to generate it as AC and use a diode to clip it off at the top or something. I really don't know much about what's out there (as far as switching devices go). Could anyone give me some advice?
@Super God,
I think you will need a rotating spark gap.
http://www.richieburnett.co.uk/rotary.html
http://www.ias.ac.in/sadhana/Pdf2001Oct/pe941.pdf
Groundloop.
Quote from: Super God on February 04, 2008, 05:58:14 PM
Hello,
I need help in finding a device that can comfortably switch a high voltage (upwards to 30kv) and relatively high current (5 amps) at upwards of 120 Hz, for very short periods of time (enough to make a spark much like a distributor does). I thought about a relay but the contacts would surely wear out quite fast. Solid state seems pretty hopeless as well, I don't think it could handle the high voltage. The only thing I can think of would be to generate it as AC and use a diode to clip it off at the top or something. I really don't know much about what's out there (as far as switching devices go). Could anyone give me some advice?
I need to know the frequency of interruption and the duty cycle. i.E. how many times per second do you need to switch on and for how long. This is critical. Perhaps mercury can help here.
Hans von Lieven
I would suggest you use an argon discharge tube. The 2 plates are placed a certain distance apart as 2 capacitor plates.
Then you charge up the plates until the argons brakedown voltage is achieved.
You just need to adjust the plate distance to get the frequency you need at the voltage you need.
Do some research on discharge tubes to get an idea of the plate distance and such.
Quote from: Super God on February 04, 2008, 05:58:14 PM
Hello,
I need help in finding a device that can comfortably switch a high voltage (upwards to 30kv) and relatively high current (5 amps) at upwards of 120 Hz, for very short periods of time (enough to make a spark much like a distributor does). I thought about a relay but the contacts would surely wear out quite fast. Solid state seems pretty hopeless as well, I don't think it could handle the high voltage. The only thing I can think of would be to generate it as AC and use a diode to clip it off at the top or something. I really don't know much about what's out there (as far as switching devices go). Could anyone give me some advice?
Hello Super God,
There are two solid state solution possibilities, one that relatively cheap and one that is expensive.
Here is a link to a relatively cheap circuit: http://bromine.cchem.berkeley.edu/grppub/frbm2.pdf this shows 10 power MOSFETs connected in series to increase working voltage up to 8kV with fall time of 230 nanosecond. This circuit also gives a good example of how to drive the MOSFET switches correctly to separate the low voltage side of the switch from the high voltage side. Of course if you need a switch up to 30kV, the total number of components needed to build the series connections may not be the cheapest choice but perhaps still reasonable.
The other solution can use high voltage MOSFETs directly, no need for series connection because they are directly manufactured for really high voltage switching. A German manufacturer site is here: http://www.behlke.de/
For instance here is a data sheet for a 33, 50 and 65 kV solid state switch family with a min 200ns on-time for max 17-30 Amper peak currents** and here is a push-pull device for switching 2x16kV and 2x30kV http://www.behlke.de/pdf/301-03-gsm.pdf . There are many devices with fixed on-time and variable on-time possibilities too. Be prepared for the prices though...
rgds, Gyula
**EDIT: I supplied the missing link http://www.behlke.de/pdf/331-03-lc.pdf
if you will use fets use the type that i give you by mail 1800volts.
the fet-switches that you shown he DONT WORK !
Give attention this are fet swiches for 110 and 230volt (ac) application.
this are "optocopler" that say:
the ISOLATION Votage between input and output can be up to 100kV
YOU CANT SWITCH this Tension (voltage).
Its wvry old "goodknow" technology .
Pese
To switch hi Kvolts .
use rotary contct wit air-gap
Quote from: pese on February 09, 2008, 09:16:19 AM
if you will use fets use the type that i give you by mail 1800volts.
the fet-switches that you shown he DONT WORK !
Give attention this are fet swiches for 110 and 230volt (ac) application.
this are "optocopler" that say:
the ISOLATION Votage between input and output can be up to 100kV
YOU CANT SWITCH this Tension (voltage).
Its wvry old "goodknow" technology .
Pese
To switch hi Kvolts .
use rotary contct wit air-gap
Hi Pese,
I fully understand you offer MOSFETs of 1800V drain source voltage ratings for sale, this is OK.
With due respect indeed, but I do not think it is ok you say the MTP1N100 MOSFET types (that are included in series in the 8kV switch) do not work, when they are actually do. They have 1000V drain source voltage ratings and the parallel MOVs with their 920V ratings should protect them in case the 1000V limit is approached.
Finally I do not understand at all where is the 100kV opto coupler mentioned or applied in this paper???
( http://bromine.cchem.berkeley.edu/grppub/frbm2.pdf ) Please refer to text where you found it? Or did you mistyped it?
Regards
Gyula
If yur looking for a cheap solution, find you one of these bucket-men, that climb the utility poles outside, and get a switch from him,
the ones they put on the transformers before a residential drop should be perfect for what you need.
you can operate it with a 5v IC chip and set whatever frequency you want on the thing, up to some limit which is ungodly higher than 120hz, these switches handle a 3-phase signal at a MUCH faster rate.
Maybe I could use the distributor from a car? The source is gonna be a capacitor bank (big) and when the distributor closes the circuit the capacitor will discharge through a spark gap.
Quote from: gyulasun on February 09, 2008, 10:55:51 AM
Quote from: pese on February 09, 2008, 09:16:19 AM
if you will use fets use the type that i give you by mail 1800volts.
the fet-switches that you shown he DONT WORK !
Give attention this are fet swiches for 110 and 230volt (ac) application.
this are "optocopler" that say:
the ISOLATION Votage between input and output can be up to 100kV
YOU CANT SWITCH this Tension (voltage).
Its wvry old "goodknow" technology .
Pese
To switch hi Kvolts .
use rotary contct wit air-gap
Hi Pese,
I fully understand you offer MOSFETs of 1800V drain source voltage ratings for sale, this is OK.
With due respect indeed, but I do not think it is ok you say the MTP1N100 MOSFET types (that are included in series in the 8kV switch) do not work, when they are actually do. They have 1000V drain source voltage ratings and the parallel MOVs with their 920V ratings should protect them in case the 1000V limit is approached.
Finally I do not understand at all where is the 100kV opto coupler mentioned or applied in this paper???
( http://bromine.cchem.berkeley.edu/grppub/frbm2.pdf ) Please refer to text where you found it? Or did you mistyped it?
Regards
Gyula
Soory for wast time.
this you can make yourself. !!
Constructing an CASCADE with 10 pcs of 1000 volt transistors.
(20 pcs 500volt transistors) .
all collector to the emitter of next transistor.
ech transistor will be driven from an smal coupled windings (secondaries 1 to 10). the connect emitter an base trom each single transitor.
this arrangement is drive from an primary that is connected to this DRIVER Transformer..
each winding must be good isolated from ech others.
THIS sort of CIRCUIT can only PULSING the transistor arrangement for short time ON.
That will be OK for the use that you need.
Gustav Pese