Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Register Log in
  • Congratulations IDS on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Fairchild IGBTs OK? 1

Status
Not open for further replies.
Renovator1

Renovator1

Industrial
Mar 14, 2003
72
Basically just looking for feedback here on the quality level and how far said IGBTs deviate from the datasheet specs (ie - says it will handle 60A at Tc = 100C but lets out the magic smoke at 40A, etc...). I'm mainly interested in non-punch-through (NPT) or "field stop" IGBTs for easy paralleling and several Fairchild parts keep coming up in my searches at, eg, half the cost of IXYS/Infineon/IRF/etc and since I totally believe TANSTAAFL I'm wondering what gives. So, any feedback would be most welcome. (this will be my first use of IGBTs for something besides tinkering, btw).

-Jeff
 
Replies continue below

Recommended for you

Sort by date Sort by votes
Check that your design is correct, that you consider all losses, that you measure case temperature at the proper spot before blaming a manufacturer.
 
Upvote 0 Downvote
Um, I'm not blaming anyone for anything... yet. All I'm looking for at this point is some feedback from people that have used Fairchild IGBTs. This is mainly because I'm having a hard time believing they aren't "too good to be true" because they are so much cheaper than their competitors. E.g., Fairchild part number FGH40N60UFD is a "Field Stop" IGBT that costs around $2US in 100's and has the following specs:

Ic = 40A @ 100C (I never bother with the 25C rating)
Vce[sat] = 1.8V @ 40A
Qg = 120nC @ 400V
*positive TC of Vce[sat] at all current levels shown*

Nobody else has a device with similar specs for less than twice this price; actually the specs are less compelling and the price is usually 3x to 4x higher. I have had good results with Fairchild MOSFETs in the past and have chosen them for the same reasons - they had the best price/performance ratio at the time - but the differences in price were not so stark between their MOSFETs and their competitors' as they are with their IGBTs.
 
Upvote 0 Downvote
The price you have to pay for such a device in small quantities (100s is a small quantity) is more governed by sales channel than by manufactureing and technology.

Most semiconductor manufactueres are just not intrested in selling a few hundered TO-220 parts.
 
Upvote 0 Downvote
Do you have any observations or experience with Fairchild IGBTs, electricuwe? That is what I was asking about, after all...

That said, my experience with many semi manufacturers flatly contradicts what you just stated: they are more than happy to provide technical support and design assistance whether I'm buying a 100 parts a year through a distributor, or 10,000 directly from them.

 
Upvote 0 Downvote
The FGH40N60UFD looks like a really good part. I'm comparing it to the IRGP50B60D which I'm currently using at the 25A average operating point that I have.
$4.52 each vs. $1.79 for the FGH40N60.
The FGH40N60UFD has slightly better gate charge, better Qrr of the diode, and turn-off loss (in mJ/A), and the total switching loss of the IRGP' is 68% of the FGH'.

MMMM, I haven't answered your question about what can be wrong with this part. Looks like a really good price!
 
Upvote 0 Downvote
Exactly, hgldr - it has great specs, the graphs seem to back them up and if you use the usual 40-60% derating of the usually overly-ambitious current capability you still end up with a part that delivers excellent value. I guess I'm just a little bit skeptical *because* the price/performance ratio is so good... I'm guessing/hoping it's because Fairchild is trying to grab market share and not because the specs are pure fiction. :)

 
Upvote 0 Downvote
The price/performance ratio mainly is the result of applying the Fieldstop approach. Compared to NPT losses go down significantly at the price of more difficult control and higher requirements on keeping stray inductance low.

If you can deal with this, even when paralleling, it's fine.

Check Infineon for parts with even lower losses:

Don't know about cost.

Check for conference papers from Infineon on IGBT3 for details of the technology.

Regarding your experience with support: where are you located ?
 
Upvote 0 Downvote
I know folks that are using the Fairchild IGBTs with excellent results. I believe that Fairchild bought the Harris IGBT products and I used those; good parts. Had some analysed once. I recall they had some secret Gold doping somewhere that surprised the analysis folks (International Rectifier I think).
 
Upvote 0 Downvote
electricuwe - the Infineon part you referenced has a slightly lower conduction loss but twice the gate charge. Not a good trade-off for my application. Matching between the built-in anti-parallel diode and the IGBT is exceptionally good, though, and that is a benefit. However, the Infineon part is not available from *any* of the major US distributors and that pretty much eliminates it from the running: I hate being held hostage by a single supplier.

I'm in the US - Florida. Until very recently I have never had a poor technical support experience with any US manufacturer; the recent offender was International Rectifier (and I suspect their being taken over by Vishay has a lot to do with that).

sreid - Thanks much for your observations - that's just the sort of second-hand, totally unsubstantiated data I needed to convince me to buy a hundred of those babies ;)

 
Upvote 0 Downvote
sreid is correct regarding Harris IGBT products beeing bought by Fairchild. I also used Harris parts in the past with good experience. But Fairchild transfered engineering to Korea and manufacturing to China and in the mean time several new technologies have been introduced. So I do not know how much of the Harris genes is still within the Fairchild products.
 
Upvote 0 Downvote
I still like IXYS personally. Good products, good support but they do charge a premium. Sometimes worth it, better sleep.
 
Upvote 0 Downvote
I grew up one block from International Rectifier. As a kid, I used to go and collect selenium rectifier stacks from their trash. Anyone remember those?
 
Upvote 0 Downvote
Here's the update: I went ahead and bought a hundred of the Fairchild FGH40N60UFD but after talking with some of the other folks involved in this venture we thought that there might be an overall economic advantage in going with a single IGBT module (cost of assembly, time to design, etc.). So, despite what I said above about not wanting to be held hostage to a single supplier, I called up Semikron to ask, somewhat hopelessly I'll admit, about the price/availability of their new SKM600GB066D Trench/NPT module - in quantities of 24 (standard box size) they want $100 ea. Sold!

Oh, and electricuwe - Semikron uses Infineon dice to make the modules, so I went ahead and made the trade-off of higher Qg for lower conduction loss (Qg = 4400nC; gonna take 15A current pulses to transition the gate in 200nS...)
 
Upvote 0 Downvote
$100 for a 600A/600V half-bridge module that can switch as fast as 77nS seems pretty good to me. It's not like there was an overwhelming number of choices here, but I gather from the ?? you think I could have done better? If so, do tell. I'm stuck with these hundred but nothing says I have to keep buying them in the future.

 
Upvote 0 Downvote
If you want to use Infineon dies in modules, you can also choose from the Infineon (former eupec) product range:


Regarding current rating:

The manufacturer need a number to print on the package, but for a design you need to do a loss calculation according the specific operating conditions. When comparing just by the nominal rating be aware that some rate at 25°C and others at 80°C..100°C case temperature.

Please consider that with IGBTs you do not have the same loss vs. die size trade off as with Mosfets (due to threshold in forward voltage like a diode). So using 600 A rating instead of 40 A will have only a small benefit in losses.

Make a sound choice technically with two or three alternatives and then care about availability and price.
 
Upvote 0 Downvote
Status
Not open for further replies.

Similar threads

jrick
  • Locked
  • Question
230 single phase A\C running on 460 3 phase circuit- OK? 6
2
Replies
23
Views
348
waross
waross
VEBill
  • Locked
  • Question
Simple series Zener diode to drop voltage - okay? 5
Replies
14
Views
112
danw2
danw2

Part and Inventory Search

Sponsor

Back
Top Bottom