6
\$\begingroup\$

I was trying out a hobby project and I was using the IRFD120 and IRFD110 power MOSFETs to control a motor. The transistors worked pretty fine and we simply ran out of them. When I went to buy some more I found out that they are outrageously expensive for a transistor. However, I found the FQP30N06L and FQP27P06 power MOSFETs for normal cheap transistor price. This raised my "sanity" flag. I thought that I knew which parameters matter, yet it seems that there is another parameter so important that it affected the price this much price. I ran a small specs comparison and I got this (comparing N-channel only for simplicity):

  1. FQP30N06L:

    • Current Rating = 60 A
    • Voltage Rating = 30 V
  2. IRFD110:

    • Current Rating = 1 A
    • Voltage Rating = 100 V

This clearly shows that the FQP30N06L can withstand more power, yet it is ~5 times cheaper than the IRFD110. Is it only the 100 V rating that is making the IRFD110 so special?

I am asking this question because I need to buy large quantities and I haven't used the FQP30N06L yet. Would it be a good idea to replace the stock of IRFD110 and IRFD120 with FQP30N06L and FQP27P06 for hobby project related applications similar to this: youtube.com/watch?v=Y-afnY32RrY?

\$\endgroup\$
  • 3
    \$\begingroup\$ Have you tried looking at other parameters, like operating temperature? Also did you recently have a look at the pricetag for a brand new 1969 ford mustang? \$\endgroup\$ – PlasmaHH Apr 22 '16 at 13:47
  • \$\begingroup\$ Without knowing the channel resistance when the device is fully enhanced, your claims about how much power it can "withstand" are meaningless. -1 and vote to close because it's a shopping question and an answer would be opinion-based. \$\endgroup\$ – EM Fields Apr 22 '16 at 14:04
  • 3
    \$\begingroup\$ @EMFields this is not a shopping question as it's not asking where to buy them, it's a question about what the parameters mean. \$\endgroup\$ – pjc50 Apr 22 '16 at 15:40
  • \$\begingroup\$ Sorry guys for the confusion, I will update my question. @pjc50 is right. I basically wanted to know what parameters matter because I thought I knew what matters yet it seems that something else is very important and the price raised my curiosity flag and I had to ask \$\endgroup\$ – himura Apr 23 '16 at 19:24
12
\$\begingroup\$

The 100V could be a reason: high voltage MOSFET tend to be quite expensive. But I think in this case, the reason is the strange package the IRFD110 uses. This is non standard and certainly uses specific production lines.

And usually, the reason is just that: they use older technology that is more expensive to produce than newer technology. The older products then become more expensive, although they are not better speced (usually, they are worse, as you have noticed for the current handling ).

Now, if you look for a replacement MOSFET, look at the datasheet

  • The voltage handling (both in VGS and VDS) should be the same or above (in your case, it's below, so double check that you can do this in your circuit)
  • The current handling should be the same or above.
  • The VGs gate threshold should be similar (in your case, it slightly below, but that is less of a problem than if it's above)
  • The RDSon should be below
  • The gate charge should be in the same range (if it's a lot more, you may have to use bigger drivers).
  • Make sure the dissipation is appropriate.

If you check all this, I think you're clear.

\$\endgroup\$
  • \$\begingroup\$ Also it'd be helpful if the voltage on the mosfet does not exceed 30V.. \$\endgroup\$ – Barleyman Apr 22 '16 at 14:06
  • \$\begingroup\$ After seeing all the answers I think this one has the most information. I would like to also point out that @Barleyman 's answer have the other half of the answer. The IRFD110 and IRFD120 are simply rare to find as they are old. \$\endgroup\$ – himura Apr 23 '16 at 19:36
3
\$\begingroup\$

That package is not widely available from different makers (though I see plenty in stock- eg. 20.5 cents Qty 50 from Future).

If you want the lowest price, now and in the future, newer parts in SMT will likely be the best value, with the bulky TO-220 package being the main through-hole alternative.

The original part has quite a high Rds(on), a not-so-low gate charge and an oddball package. Not great performance.

Generally, all other things being equal, the die size (and thus manufacturing cost) of a MOSFET is related to maximum Rds(on) and Vds rating. But the acreage of silicon does not necessarily track selling price, especially for parts on extremes of the life cycle and in tiny quantities. You may find the price dips as a part is going obsolete, as the stock is dumped onto the market, followed inevitably by unavailability at any kind of reasonable price (brokers and middlemen as well as users doing lifetime buys soak up the stock, with the former looking to make a tidy profit by waiting until the users run out and are forced to choose between huge engineering and testing costs vs. paying a fortune for obsolete parts).

MOSFET technology in particular has been evolving steadily, and newer better and cheaper parts are regularly introduced.

Try doing a parametric search at a distributor- of course you have to understand all about the circuit to know which parameters are important- Vds, Rds(on) at what gate voltage, gate charge, power dissipation capability in a realistic situation (not the stupidly high number on the datasheet), and a number of other factors.

Obviously if you can relax the Vds requirement to 50V or 60V you can get a better performing MOSFET and/or a better price. If the power dissipation turns out to be very low, you might even be able to use an SOT-23 MOSFET which could be very small and cheap.

But it`s always possible, in any given circuit, there is some parameter that does not allow safe substitution and serious problems will arise, as they tend to do if they have not been fully examined quantitatively.

\$\endgroup\$
  • \$\begingroup\$ SOT-23 mosfets tend to die if you look at them sideways. I'd take the stated power/current handling capabilities with artery-busting amounts of salt. Yes, they have their places but only where they're not subject to any electrical abuse. \$\endgroup\$ – Barleyman Apr 22 '16 at 16:17
  • \$\begingroup\$ @Barleyman You should take the datasheet capabilities claimed for TO-220s with a least as healthy dose of skepticism. Hundreds of amperes continuous drain current (with an asterisk) or hundreds of watts continuous dissipation (417W?!) based on you maintaining the case at 25°C... not likely. \$\endgroup\$ – Spehro Pefhany Apr 22 '16 at 18:16
  • 1
    \$\begingroup\$ And let's not forget volume. 30V parts are far more commonly used than 100 volt parts, and the larger volume pulls price down. \$\endgroup\$ – WhatRoughBeast Apr 22 '16 at 18:24
  • \$\begingroup\$ @SpehroPefhany at least you can bolt some fairly hardcore cooling on larger mosfet cases. And even if you don't it's got orders of magnitude more mass so if you made a mistake you have time to switch off the lab supply etc. Sot-23 (and Sot363) parts die out of anything. Stick a probe in with shaky hands? Dead. Put in the plug with power on? Dead. Touch a conductive part with oscilloscope probe? Dead. Plop the PCB on desk too hard? Dead. Pick your nose? Dead. SOT-223 et al are by far more forgiving of everyday mishaps. \$\endgroup\$ – Barleyman Apr 22 '16 at 22:16
  • \$\begingroup\$ I had this dinky little -5V Sot-23-5 charge pump on a 2kV X-ray sensor bias circuit I was replacing constantly during testing. Anything at all went wrong with that thing and that little bug gave up the smoke. Or, usually no smoke at all, just silently died of some transient a larger chip would've shrugged off without breaking a sweat. In this case the tricky part was getting from 2kV to GND and back in under 1ms without over/undershoot and not using exotic parts. Or dissipating a lot of heat. It worked just fine in the field, but testing.. \$\endgroup\$ – Barleyman Apr 22 '16 at 22:22
2
\$\begingroup\$

It sounds like you're being ripped off. In Digi-key the IRFD110 is considerably cheaper than FQP30N06L. 21p versus 35p each for 1000 pieces.

That being said, the different manufacturers do use different pricing strategies. The right cost of a component is as much as you can possibly ask to maximize the profit. c.f. Apple.

\$\endgroup\$
  • \$\begingroup\$ I like your answer. I do. From the bottom of my heart. But I have to accept another answer because it had more information. Here is the digipart link for reference: digipart.com/part/IRFD110. \$\endgroup\$ – himura Apr 23 '16 at 19:18
2
\$\begingroup\$

Would it be a good idea to replace the stock of IRFD110 and IRFD120 with FQP30N06L and FQP27P06 for hobby project related applications?

Nobody but a fool is going to say yes to that question. It's all down to the application, the load, the power rails, the switching frequency and the driving circuit, none of which is specified in your question.

\$\endgroup\$
  • \$\begingroup\$ I tried to specify the applications i was talking about. I agree with you that no one can answer that question with certainty unless full application specs are defined. In fact, a sane person would say just get both, they have different rated voltages and currents. To make it clearer, i meant hobbyist projects similar to this one: youtube.com/watch?v=Y-afnY32RrY \$\endgroup\$ – himura Apr 23 '16 at 19:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.