9

I'm only just beginning to read about diodes. I'm looking at the 1N400x series, where all specs are the same except the reverse voltage (link: https://www.mouser.com/ProductDetail/Diodes-Incorporated/1N4003-T?qs=rGAXPo9uwV0nfQ36LZW%252BLg%3D%3D for example). The reverse voltage increases as x increases, all else is the same (even price).

From what I understand, the reverse voltage is the maximum voltage drop the diode can have (in reverse) before it fails. My question is, is there a reason I shouldn't just buy the highest reverse voltage I can find? Is the N4001 not the same in every way as the N4003 except that the N4003 can work with higher voltages without failing?

Lespiegle
  • 93
  • 3
  • Are they still the same price when you buy tens of thousands at a time? I've never had any idea why all the specs should be the same though and I only know physical reasons why they wouldn't be. – DKNguyen May 28 '20 at 21:53
  • 1
    Looks like it. Ordering 5000 pieces for x = 1 to 3 is $150 for all 3. The lab I'm doing calls for the N4001 but seeing as these are parts I'm going to be re-using would it make sense to buy the ones with higher reverse voltages? They'll function the same, won't they? – Lespiegle May 28 '20 at 22:00
  • 1
    @DKNguyen Nice to see that "discussion" there. I like Spehro's answer better -- it includes some details I wasn't aware of (PIN as well as RF use, though as soon as he said PIN my mind was cast in that direction and another, which make me want to open one up to expose it to light.) I also took note that Dave either chose not to reply or else deleted his replies. I consider it bad form and perhaps a mark of insecurity to fail to respond to reasoned challenges from informed quarters. – jonk May 28 '20 at 22:44
  • Sometimes components with higher, better ratings are the same price or even cheaper than those with lower ones. Because they are more used, I think. – Fredled May 29 '20 at 00:04

1 Answers1

5

You can see my answer linked by DKNguyen about the real differences based on voltage. Realistically, for mains frequency normal applications you may as well buy 1N4007 or the SMT equivalents as the price difference is pretty small, at least for non-consumer quantities. In a million quantity the difference between a 1N4002 and an 1N4007 might be a total of $1,000, so it's worth spec'ing the cheaper one if it works and nobody wants to throw away a kilobuck.

The situation is rather different for Schottky diodes- higher rated voltage generally leads to a higher Vf at a given current. Compare the differences between 1N5817, 1N5818 and 1N5819 rated at 20, 30 or 40V.

Vf at 1A is max 0.45, 0.55 and 0.6V. So using a 1N5819 where a 1N5817 will do, means about 1/3 more power dissipation.

Spehro Pefhany
  • 376,485
  • 21
  • 320
  • 842
  • Thank you, your link led to exactly what I was looking for. It's too bad I couldn't find that on my own with my initial searches. – Lespiegle May 29 '20 at 10:42
  • For Diodes Inc, the 1N400x series is obsolete and there is no difference in cost at Qty=1M for their replacements 1.27 cents/pc https://www.digikey.ca/en/products/filter/diodes-rectifiers-single/280?s=N4IgjCBcoJxgLFUBjKAzAhgGwM4FMAaEAeygG1wwYB2AVgAYQiwrqA2GEAXSIAcAXKCADK-AE4BLAHYBzEAF8iAWgBMSEKkjiAroRLkQtbn0GQQC5Uegb02fEVKQKEHiAFCFikGqcgcLojZ1CQATISUweggTDyIAR34ATyFI%2BjS0phAk3jwhDBxUeXkgA What does that tell you? The price is artificial based on supply & demand & qty on order !!! – Tony Stewart EE75 May 29 '20 at 23:34
  • @TonyStewartSunnyskyguyEE75 I’m using Shenzhen pricing. I don’t think Digikey.ca gets many orders for 10^6 parts. 1.27 cents is ridiculously expensive. – Spehro Pefhany May 29 '20 at 23:38
  • Of course, but not as much as the cut-tape pricing 50x higher with some differences in pricing. The only difference is the bulk resistance (Rs) vs PIV and it's not that significant. But at 5A flyback pulses at 1% duty cycle, it makes a difference. You have a product using 1m diodes. Congrats. I only had one with that many low profile metric flathead screws – Tony Stewart EE75 May 29 '20 at 23:43
  • @TonyStewartSunnyskyguyEE75 nobody should be using those diodes where speed is a factor, but the fwd and reverse recovery times are about twice as slow for the higher voltage types. I guess they’d create more noise as a result even at mains frequency. – Spehro Pefhany May 29 '20 at 23:48
  • Is that because the lighter doping causes the forward voltage capacitance to be greater while the reverse bias capacitance to be lower? – Tony Stewart EE75 May 29 '20 at 23:50
  • @TonyStewartSunnyskyguyEE75 My understanding is that they’re more heavily doped P&N with an intrinsic layer between but I could be wrong. – Spehro Pefhany May 29 '20 at 23:52
  • It's only a 3dB difference in worse recovery time for 10x higher PIV, and about the same is Rs bulk resistant so it's not significant for a general purpose 1A diode. – Tony Stewart EE75 May 30 '20 at 00:30