Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
IssacD
Level 1
Level 1
5 replies posted 10 sign-ins First reply posted

Hello Community

If the source of a N-channel mosfet is connected to a negative high voltage, and the drain is connected to ground, is there any method to drive the mosfet, without using a isolation transformer ?

this is the problem I encountered  when I try to design a fast pulse generator to produce an edge rising from -4.5kV to 0.

please share your solution.

thanks 

0 Likes
1 Solution

Great!

If set current clamp loop for MOSFET, the idea will very will.

Recommend a simpler idea:use 6500V IGBT. This is IGBT like with reference:

https://www.infineon.com/dgdl/Infineon-FF225R65T3E3-DataSheet-v01_11-CN.pdf?fileId=5546d4627a0b0c7b0...

View solution in original post

6 Replies
喜马拉雅之雪
Employee

For high voltage switching output, we have five solution for different. Would you like share your application?

For one of the drive and only focus on testing, they maybe can try:

_0-1677465835113.png

Link:https://www.infineon.com/cms/cn/product/power/gate-driver-ics/eicedriver-compact-300-mil/?filterValu...

0 Likes

Thanks for your reply.

The purpose is to generate a pulse with around -4KV amplitude, less than 50ns rising and falling time, flat top voltage, and variable width from 300ns to 450ns. 

my idea is below.

gate_pulse.PNG

step1 :when  Q1 is off and  Q2 is on, the capacitor is charged.  At the output, a fast falling edge from about 0 to  -4KV is generated. 

step 2 :hold the state for 300~450ns, the -4KV flat voltage is generated.

step3: when Q1 is on and Q2 is off  ,the rising edge from -4KV to 0 is generated.

Of coures, there are 2 or 3 mosfets in serial in place of the Q1 and Q2 to hold the -4KV.

Here is the problem when considering  the driver .

If we don't use isolation transfomers,

The reference ground of driver 2 connected to Q2  is  -4KV. Then the  power supply should begin at -4KV+18V ,or the driver 2 is hold a high voltage.

But whe Q2 is switched on ,the reference ground  is  changed suddenly from -4KV to 0,and then fall to -4KV. the power supply should change simultaneously.

The  similar problem happen at the  driver2.

 So when the source of  N-mosfet is connected to a negative voltage and the drain is connected to 0, the  level of the source is jumping.

I want to find a solution of the driver without using isolation transformers.

 

0 Likes

Great!

If set current clamp loop for MOSFET, the idea will very will.

Recommend a simpler idea:use 6500V IGBT. This is IGBT like with reference:

https://www.infineon.com/dgdl/Infineon-FF225R65T3E3-DataSheet-v01_11-CN.pdf?fileId=5546d4627a0b0c7b0...

Thanks.

In fact, there is a restriction I didn't mention that the delay should be less than 150 ns.

So the IGBT may be not  suitable in this application.

 

0 Likes

 

This is 225A switching time. Your switching current is about 1A. It will go short.

_2-1677568721992.png

 

 

 

0 Likes

thanks for your suggestion.

No matter use a mosfet stack or an IGBT,I still have to keep the power supply of the output of the driver around 20V respecting to the reference all the time  ,as the reference changes from -4kv to 0 ,  without using an isolation transformer to connect the driver and the switch.

Maybe a bootstrap circuit works.

Anyway, using IGBT simplify the circuit design. Thanks.

0 Likes