That's all you can do is throw derogatory remarks.
But never expose yourself to REAL technical evaluation.
I'm still trying to find someone who knows why, when you try to force a fine time-step on LTspice, you get "fill-ins". I suspect that part of LTspice's algorithm ignores what it "thinks" is numeric noise, to dramatically improve the speed. I further suspect that PSpice caught a problem that LTspice can't. I have had a handful of cases over the years where PSpice caught instabilities that very expensive simulators missed.
In a post in the past few days, while I was traveling, you responded to a comment of mine...
[JT] The Bode plots of your circuit definitely show marginal phase margin.
[JL] It's a little under-damped. I've said that from the beginning. But a Bode plot is a bad way to evaluate a nonlinear system for step response.
Indeed, As I referenced a P-E-T Magazine article "When Bode Plots Fail Us".
You are correct, per the P-E-T article, a GOOD Bode plot does not guarantee stability; but, conversely you are wrong in trying to play your extraordinarily poor Bode plots as not portending ringing. (See Tim Wescott's remarks.)
Your system has three poles and one delay (*)... and a zero only from ESR... not a very good approach to stability...
(*) One pole from Cgs of the FET, one pole and one delay (excess phase (**)) from the OpAmp, and a MOVING pole from the variable R of the FET and the filter caps.
Some time back Jamie suggested adding a zero and I confirmed that approach as heading in the right direction. To make it comfortably stable, but still with a teeny bit of hunting, you do feed-forward...
formatting link
Looks pretty nice, but not what I'd call perfect, you and I have different perceptions of what perfection means... yours is that "it works"... whatever that means :-) The variable pole is the nasty... but, as I was tooting along at 100MPH coming back from Yuma, I think I see a really easy solution.
(**) Properly "pole-split" OpAmps have only one dominant pole, the other pushed out so far it has no effect on amplitude as the GBW crosses the axis, so the best model match is a pole plus an all-pass to set the excess phase, otherwise unaccounted-for at the axis crossing... in my configurable OpAmp model _and_ in Mike Engelhardt's.
So go skid on a condom in front of your luxurious San Fransicko home. Then post a further slur... keep 'em coming, makes you looks so professional ;-) ...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
Oohhh, did I hurt your feelings? Who woulda thunk that you had feelings?
But seriously, I can also correctly simulate and build working LDOs. And a lot of other electronics that you can't do, or even discuss.
That's what customers are for.
formatting link
But the Pspice sim was nonsense, and the (or at least my) LT Spice was right on, ran in under a second, had no warnings, results confirmed by the hardware.
I have had a handful of cases over the
My dual LDO has no instabilities. It works great. The instabilities you found were illusions caused by bad driving.
It's stable, you moron, all the way down to zero ESR. There is no weird noisy-envelope oscillation, no wild shots, none of the things you gloated about in your ludicrous 18-page PDF.
You can't worm out of it: you screwed up big-time.
--
John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators
You posted a circuit which I, initially, gave it no never-mind.
Then Spehro asked about the excessive pre-load.
So I looked at your circuit said it's probably unstable.
Here's where you blew it... if you'd posted the scope photos, however much you'd fudged the circuit, I would have lost interest.
Instead you perpetuated the lie by posting an LTspice circuit DIFFERENT from what you claimed as in production.
That circuit works because of the OpAmp specs... but you don't even know why.
My evaluation of your original post shows gawd-awfully-poor phase margins, irrespectively of load.
Then I "compensated" it in the style of the LTspice schematic, where I found it to "work", but it hunts, because of all those large delays.
You now conveniently snip (and ignore) technical questions, and treat us to "Tourette's by keyboard".
So you are indeed a massive SERIAL LIAR. Such a shame, you should seek treatment.
BTW, Helmut Sennewald (master LTspice guru) responded to my question about "fill-ins"... note the "...I think LTspice HAS TROUBLE WITH CONVERGENCE of this circuit and thus it fill-ins estimated values in the matrix."....
--- >
"Hello Jim, I don't really know what it does, but I think LTspice has trouble with convergence of this circuit and thus it fill-ins estimated values in the matrix.
You could try with the Alternate solver or other settings to help LTspice.
Best regards, Helmut"
So LTspice couldn't zero-in on the hunting, but PSpice could :-)
I know you will now snip all the above truthful statements and throw us some more "Tourette's by keyboard". Will you call Helmut an "idiot", too? You might, your ego claims competency above all others.
What a loser! John Larkin is his name, in case anyone has forgotten :-) ...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
Idiot. Of course it's different. The production schematic doesn't show cap ESR or the additional off-sheet bypass caps. And the posted sim is just one snapshot of the cases and corners that I explored for several minutes. Your compulsion with them matching perfectly may be necessary in your world, but it's not in mine. It works, and you are arguing against physical reality, seldom a smart thing to do.
Of course the opamp matters some. Duh. The circuit works the way I expected it to work. I guessed the comp values and simulated to fine-tune. My first guess was about 4:1 low, and was overdamped. I used 1n+2K because those values were used elsewhere on the board, and we like to keep our BOM line items down.
Have I mentioned lately that you are an idiot?
It's a little underdamped, about the same amount in the sim and on the board. I said that from the first. I could change the comp cap to damp it, but why bother to write an ECO when it doesn't matter? It actually gets more ringy at lighter loads, and if it were a lot worse, I might change the comp cap.
Artifact. Operator error.
Calling you an idiot is simply stating obvious truth.
You are incompetant at simulation, and published an 18 page proof. Neither my sim, nor my working hardware, do anything as stupid as the junk you published. The waveforms on pages 13 and 14 are hilarious.
formatting link
Anybody but an amateur would have looked at those insane waveforms and questioned the simulation. You were so eager to catch me doing something wrong that you quit thinking and made an ass of yourself.
Suits me.
--
John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators
Note the selective snipping without attribution... particularly avoiding Helmut's comments which could get Larkin slammed big time.
But, as you would expect, Tulane graduates don't even know where farts come from ;-)
In my "wasted time" I have found the way to handle a variable RC corner in CMOS, so I thank you John "Tourette's by keyboard" Larkin.
I've certainly enjoyed taunting you... next trip, hopefully a stroke, so you can kiss your own ass good-bye ;-)
I no longer get emotional, I just enjoy the kill :-p ...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
Ooops! I missed complimenting John "Tourette's by keyboard" Larkin's peer group, Jason Betts and Ian Field, for their contributions to John's technology base :-) ...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
You snipped it out, you dummy. You never want to face real technological challenges. So you snip them out so Jason and Ian won't notice :-)
You really ought to learn to read. I guess they didn't teach that at Tulane. I think I'll send your posted circuits to Tulane and ask them if that's the sort of shit they teach there in trailer-trash land ;-)
Gawd! John "Tourette's by keyboard" Larkin must also have Alzheimer's... he trimmed out Helmut's comments (because it specifies the problem... LTspice fudges when it can't find convergence :-), then claims no remembrance of them :-p ...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
Sorry, Daddio, they're not wrong. LTspice can't converge on it (the Helmut explanation that you keep trimming, thinking no one will notice), and you can't see it because of the long time constants... shorten up everything and you'll see the "hunting", which, if you had half-a-brain (but you don't), you'd know come from asymmetric rise and fall at the output.
It's really troubling to me that someone can be so stubbornly ignorant. Do you think anyone is impressed except, perhaps, Jason and Ian, who are both as ignorant as rocks ?:-) ...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
What do you mean "shorten up everything" ? If I set the max time step to 1 ns, it runs much, much slower but the waveforms look the same. Why would I want to do that?
I think your simulation is broken somehow.
Why do you keep insisting that a simple circuit that works fine, doesn't work?
Do you contend that pages 13 and 14 of your PDF are realistic waveforms?
formatting link
My sim and my hardware agree. The sim runs in a fraction of a second with no visible warnings. The LDO works fine in real life. I poked it with load transients and posted the oscilloscope pics. They look nothing like the crazy waveforms you published. I see a little nicely damped ringing but no "hunting", on the sim or on the board.
formatting link
Do you stand by your waveforms, pages 13 and 14, as correct?
--
John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.