Author Topic: Comparison of power switches for least on voltage drop  (Read 1599 times)

0 Members and 1 Guest are viewing this topic.

Offline ZeroResistanceTopic starter

  • Frequent Contributor
  • **
  • Posts: 585
  • Country: gb
Comparison of power switches for least on voltage drop
« on: October 06, 2018, 06:48:18 am »
I'd like to know which power switch among
1. Transistor
2. IGBT
3. Mosfet
4. Triac
would have least forward voltage drop when used as a power switch in power applications at voltages in the range of 300 to 500VDC.
From my understanding a Mosfet should have been capable of this, but from what I've seen Power mosfets generally have a higher Rds on.

Thanks in Advance.
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15149
  • Country: de
Re: Comparison of power switches for least on voltage drop
« Reply #1 on: October 06, 2018, 08:08:01 am »
A triac is not really working with DC - it needs some extra tricks to turn them off. It is not that good with voltage drop either.

IGBTs have a significant minimum drop ( > 1 V), similar to the triac, even at low currents. It might still be acceptable at a relatively high voltage (e.g. > 600 V).

For MOSFETs the voltage drop is just from R_on. So it depends on the current and size of the MOSFET. It can be very low drop with a large one, but also higher than that of a IGBT with a rather small one optimized for high voltage switching. R_On goes up with higher voltages, so that very low R_on may not be practical.

BJT Transistors have a small fixed minium drop of around 50-100 mV combined with on Resistance. Depending on the size it can be reasonably low, even at high voltage.
 
The following users thanked this post: ZeroResistance

Offline 6PTsocket

  • Regular Contributor
  • *
  • Posts: 212
Re: Comparison of power switches for least on voltage drop
« Reply #2 on: October 06, 2018, 09:23:36 am »
IGBT's are used in high power situations. Where I worked, a bank if them was used to switch a long string of big motors. The hybrid design allows fast switching so the switch does not dissipate a lot of heat during the the transition.

Sent from my SM-G900V using Tapatalk

 
The following users thanked this post: ZeroResistance

Offline ZeroResistanceTopic starter

  • Frequent Contributor
  • **
  • Posts: 585
  • Country: gb
Re: Comparison of power switches for least on voltage drop
« Reply #3 on: October 06, 2018, 09:27:50 am »
A triac is not really working with DC - it needs some extra tricks to turn them off. It is not that good with voltage drop either.

IGBTs have a significant minimum drop ( > 1 V), similar to the triac, even at low currents. It might still be acceptable at a relatively high voltage (e.g. > 600 V).

For MOSFETs the voltage drop is just from R_on. So it depends on the current and size of the MOSFET. It can be very low drop with a large one, but also higher than that of a IGBT with a rather small one optimized for high voltage switching. R_On goes up with higher voltages, so that very low R_on may not be practical.

BJT Transistors have a small fixed minium drop of around 50-100 mV combined with on Resistance. Depending on the size it can be reasonably low, even at high voltage.

It would be interesting to know the voltage drop of the Triac for AC systems, from hat I've read they have round 1 to 2V drop.
Also interesting that BJT's can acheive low drops, are there any good devices at > 400V that have a low saturation volateg. in the range you mentioned (50 to 100mV) and reasonably priced.
High voltage mosfets seem to have larger Rds on and the ones that have low Rds on for higher voltages need you to spend more money.
 

Offline ZeroResistanceTopic starter

  • Frequent Contributor
  • **
  • Posts: 585
  • Country: gb
Re: Comparison of power switches for least on voltage drop
« Reply #4 on: October 06, 2018, 09:33:08 am »
IGBT's are used in high power situations. Where I worked, a bank if them was used to switch a long string of big motors. The hybrid design allows fast switching so the switch does not dissipate a lot of heat during the the transition.

Sent from my SM-G900V using Tapatalk
I guess they don't dissipate heat during the transition. But when used as a static switch, it might dissipate considerable power, looking at the high voltage drop that they have,based on what Kleinstein pointed out.
 

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15149
  • Country: de
Re: Comparison of power switches for least on voltage drop
« Reply #5 on: October 06, 2018, 10:57:46 am »
IGBTs and triacs (not DC suited) the drop out is some 1.2-1.8 V typical. So something like 1.2 V + current dependent part.
For BJTs it is some 50-100 mV + current dependent part, something like 0.2-0.5 is practical, could be 1-1.5 V at maximum current.
For MOSFETs it is only R_on. So only current dependent, but for a given chip size more than with BJTs/IGBTs. With higher voltage ones this could be up to about 1-2 V for a practical current. However for infrequent switching relatively large FETs can be feasible and thus a low drop.
Lowest drop is usually a mechanical relay  :-DD.

For IGBTs and BJTs there is also quite some switching loss especially during turn off, as they are not that fast. BJTs also may need quite some base current, as the gain is low (e.g. around 5-10 for hard switching).

There are still quite some high voltage switching BJTs available and sometime for cheap from old TV designs, like BU508 and similar.
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 20357
  • Country: gb
  • 0999
Re: Comparison of power switches for least on voltage drop
« Reply #6 on: October 06, 2018, 12:42:45 pm »
An SCR is more suited to DC, than a TRIAC. What''s the application? It's true that SCRs latch on, until the current is interrupted, but that's not a problem if it's in a circuit which naturally interrupts the current such as charging a capacitor, via an inductor.
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Comparison of power switches for least on voltage drop
« Reply #7 on: October 06, 2018, 11:05:39 pm »
MOSFET of course.  But you already knew this.  What you are missing is, what other constraints do you have?

600V MOSFETs that are suitable for synchronous rectification, say (i.e., Von < 1V so it's worthwhile over using a dumb diode), are very large indeed (large die, big Qg and Coss, big $).  Consider SiC and GaN, if size is an issue (and cost is not, though some SiC is now undercutting Si on resistance per buck).

If cost is the issue, then you must give up something else.  Thyristors are harder to use except in special cases (mains AC usually), so we can eliminate those options.  That leaves IGBTs, which are indeed quite good on power per buck at modest switching frequencies.  The losses are higher so you will pay some for heatsinking and hardware, and maybe other ventilation (holes, fan..), but that is still a better outcome than the huge cost and some bother of a big transistor.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf