John Larkin has obviously never been badgered by corporate patent lawyers, nor corrected their "legally acceptable" but sadly meaningless claims.
-- Bill Sloman, Nijmegen
John Larkin has obviously never been badgered by corporate patent lawyers, nor corrected their "legally acceptable" but sadly meaningless claims.
-- Bill Sloman, Nijmegen
On a sunny day (Tue, 29 May 2012 09:43:48 -0700 (PDT)) it happened snipped-for-privacy@gmail.com wrote in :
bipolar transistor will outperform a depletion mode FET.
Indeed, just a few hundred mv in a PNP pass. Use it all the time. And no pre-load needed. Seems like one of those self-torture excercises to use a depletion mode :-) And then fight about it...
I actually prefer NOT to patent... once I use my "art" a few times it's in the public domain, so I can continue to use it forever... I just keep it quiet... copy cats tend to be idiots and don't recognize sleights of hand in electronics. ...Jim Thompson
I have one patent, which I don't recall being hard to do. It was assigned to Imago Corporation, but I insisted that I retain rights to use in non-atom-probe applications. Which I have.
All of Jim's patents seem to be assigned.
plain
loop.
The product at issue will go in a small extruded box, not a rack. It gets local power and Ethernet connections. All our standard products are rated for 0-50 or 0-60 C.
That'll drive up the cost of the opamp. Because a NPN will have to be driven to 10.7V at the base here. Meaning you realistically need an RR opamp. With a depletion mode FET you don't, you can use ye olde jelly-bean amp.
Hey, Fred, "outperform" in what sense?
LM7301 is RRIO. The depletion fet can give free current limiting and lower dropout. The depl fet costs roughly 0.01 per cent of the selling price of the box. I'm in a business where we can have fun with circuits and not worry much about parts cost.
But I think I'll use the servoed LM1117.
All you need is to find an _available_ depletion-mode FET ;-)
I've been doing custom CMOS almost since its inception... ONLY ONE process had a depletion mode FET available. :-( ...Jim Thompson
A trick worth remembering is to simply add a pullup resistor to the output of a jellybean op-amp such as an LM324.
No single-sourced depleted unobtainium required.
You need to get out more.
And as Joerg notes, Infineon has parts. And IXYS.
Those are discrete parts... I'm talking ASIC processes.
BTW, Thanks, again! I can now compensate a resistive pass device AND its moving pole, in an ASIC ;-) ...Jim Thompson
Over 70c for an opamp?
My last two designs where in this catergory as well. I had to pay attention at the Dollar level but not at the cent level.
Ugh. That's like mounting a top on a Morgan sports car. Real men don't do that :-)
That takes me five seconds tops, wanna bet?
Well, us board level designers have to have our field day once in a while, with things you's guys can't have :-)
Yes, but it does increase quiescent power burn-off and it's one more part to stuff. Also, no free current limiting.
Wot's unobatium about a BSP149? Digikey has 38,000, Avnet has 70,000 of them.
I know. In my crowd, everyone cringes at _any_ external part requirement. ...Jim Thompson
The LM7301 is a nice part, RRIO, good specs, 32 volt supply.
That was the only car that you sent to the shop for an oil change and termite treatment.
How many wheels did yours have?
Right. It's a very different thing. And we get to solder, too.
Most had five (four on the ground and one spare). They seem to make electric ones these days:
Yeah, and that's sad because it is exactly what screwed up LDOs.
Yep! ...Jim Thompson
Have something to add? Share your thoughts — no account required.
Ask the community — no account required