Silicon Carbide Transistors and Diodes... Worth the extra $$? / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Silicon Carbide Transistors and Diodes... Worth the extra $$?

Move Thread

LAN_403

jdub1581

Sat Dec 19 2015, 01:57PM

Registered Member #55205 Joined: Mon Jun 08 2015, 06:06AM
Location:
Posts: 15

I was curious to see if anyone had a review on any these devices.

I am tired of losing Transistors and after looking at many different products, the SIC seems that it would be the most user friendly for newbies at added cost, regarding switching losses, drive, and frequency

Cree has a couple $10 range MFets I am considering, but would like another opinion(s)

hen918

Sat Dec 19 2015, 10:39PM

Registered Member #11591 Joined: Wed Mar 20 2013, 08:20PM
Location: UK
Posts: 556

The value is entirely dependant on the application, but generally, yes they are nice! I recommend C3M0065090D : 900V MOSFETs with better specs than most 200V MOSFETs, they work well in my inverter (very efficient due to low switching losses)

jdub1581

Sun Dec 20 2015, 12:27AM

Registered Member #55205 Joined: Mon Jun 08 2015, 06:06AM
Location:
Posts: 15

Sweet, I was looking at the C3M0065090D as well, also the 'matching' Schottky's and UFast's.

I've been using salvaged parts mostly, as they're free, bought some HF IGBT's and only have 2 of 8 left.

Still a newbie, been trying to become a coiler over the past year.. I have no scope to view waveforms, so I want to remove as much of the common trouble area's as possible.
And don't mind paying a little extra if the end product is more robust and efficient.

With that said I'd like to have a modular half bridge to start.
2 of the transistors, $10ea
2 Schottky's CVFD20065A $10ea
2 UF's :: DSEI120-12A (1200v 120a 40ns) $7.50 ea (probably overkill but price and ratings are nice)

So for $60 seems like it should be fairly bullet proof and extensible.

Justin

Sun Dec 20 2015, 04:29AM

Registered Member #46164 Joined: Wed May 07 2014, 08:16AM
Location: California, USA
Posts: 89

Honestly, until you get a scope I wouldn't bother with expensive silicon.

jdub1581

Sun Dec 20 2015, 06:02AM

Registered Member #55205 Joined: Mon Jun 08 2015, 06:06AM
Location:
Posts: 15

Well, $60 for the main parts is not so bad... Most of my hobbies are not cheap.
So I did make a small scope with my arduino, just to see the gdt signals. Not extremely accurate, and single channel only so I don't really rely on it.
But it was accurate enough for me to tune my gdt.. Although I was thinking about using an ADUM4223 4amp Isolated Half-bridge driver rather than a gdt.

The ADuM3223/ADuM42231 are 4 A isolated, half-bridge gate
drivers that employ the Analog Devices, Inc., iCouplerÂ® technology to provide independent and isolated high-side and low-side
outputs. The ADuM3223 provides 3000 V rms isolation in the
narrow body, 16-lead SOIC package, and the ADuM4223 provides 5000 V rms isolation in the wide body, 16-lead SOIC package.
Combining high speed CMOS and monolithic transformer technology, these isolation components provide outstanding performance characteristics superior to the alternatives, such as the combination of pulse transformers and gate drivers. The ADuM3223/ADuM4223 isolators each provide two independent isolated channels. They operate with an input
supply ranging from 3.0 V to 5.5 V, providing compatibility with lower voltage systems. In comparison to gate drivers employing
high voltage level translation methodologies, the ADuM3223/
ADuM4223 offer the benefit of true, galvanic isolation between the input and each output. Each output may be continuously operated up to 537 V peak relative to the input, thereby supporting low-side switching to negative voltages. The differential voltage
between the high-side and low-side may be as high as 800 V peak.
As a result, the ADuM3223/ADuM4223 provide reliable control
over the switching characteristics of IGBT/MOSFET configurations over a wide range of positive or negative switching voltages.

Sounds like they should work well and are only $5. having logic level input and I assume clean wave forms. Thus reducing part count and chance for drive errors.
Anyone have exp. with those drivers?

Wolfram

Sun Dec 20 2015, 12:59PM

Registered Member #33 Joined: Sat Feb 04 2006, 01:31PM
Location: Norway
Posts: 971

SiC devices have amazing performance advantages over regular silicon, but their speed also makes it harder to not kill them with voltage spikes caused by bad layourt, and they're also easier to kill with bad gate drive. If you are currently killing MOSFETs, just replacing them with SiC devices is likely to result in more broken MOSFETs. When parts break, the reason is often in the design and implementation of the circuit itself, and the real solution is to find the reason your MOSFETs are blowing, and fixing that problem. Saving the fifty bucks towards buying an oscilloscope is much more likely to result in a working design.

jdub1581

Sun Dec 20 2015, 09:18PM

Registered Member #55205 Joined: Mon Jun 08 2015, 06:06AM
Location:
Posts: 15

Wolfram wrote ...

SiC devices have amazing performance advantages over regular silicon, but their speed also makes it harder to not kill them with voltage spikes caused by bad layourt, and they're also easier to kill with bad gate drive. If you are currently killing MOSFETs, just replacing them with SiC devices is likely to result in more broken MOSFETs. When parts break, the reason is often in the design and implementation of the circuit itself, and the real solution is to find the reason your MOSFETs are blowing, and fixing that problem. Saving the fifty bucks towards buying an oscilloscope is much more likely to result in a working design.

Thanks for that, You answered what I was assuming... regarding ease of use..

As far as Killing the Fets I have, Literally all but the 8 cheap IGBT's I payed for, ALL silicon (diodes too) were all taken from free Tv's on CraigsList.. Aside from Zeners ..
So that I could learn on the cheap. Their performance reliability had more than likely been compromised from the start...
Also they all had rather low ratings.. 6amp max, 1ohm rds etc... I am about 90% sure that most of the fast diodes were failing first.

I do absolutely agree I need a real scope.. I look daily..lol... New is not in my budget though..

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob