Another goof. Again, I left the maximum timestep at 10ps for testing.
I don't think posting the entire file again is needed. Please set the maximum timestep to 1ps to show the window effect on the avgx signal.
The timestep should read .tran 0 200n 0 1ps. Then change it to 10ps and
100ps to see the effects on the avgx signal.
In all these cases, you can get a very different result in the avgx signal when making small changes to the parameters.
Again, the Bessel (or any other linear phase filter of your choice) is independant of the parameter settings. It also gives the fastest risetime with the lowest ripple.
If you happen to need a free filter program that is easy to use, try Elsie:
I guess Larkin believes the simulator can see into the future ;-) ...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| San Tan Valley, AZ 85142 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.
That depends on what you expected. A decent FFT usually includes windowing options, to remove the sort of crud that Jim got from rectangular averaging.
As far as the time shift goes, it would be easy to center the window on the actual data, and eliminate the width/2 time lag.
Pspice is expensive, isn't it? They could provide something better than a dumb laggy sinc averager.
--
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
In spite of Larkin's fudging LIMIT expression, the simulation is reporting greater than 100% ?? ...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| San Tan Valley, AZ 85142 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.
How would that give me better information? I'm NOT doing an FFT, I'm doing a running average.
I think I see a way to display that in PSpice. PSpice has a useful feature called Performance Analysis. ...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| San Tan Valley, AZ 85142 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.
It's the same problem. A rectangular window creates ugly artifacts, averaging or FFT. In your sims, look at how averaged waveform ripple varies with tiny changes in window width. Note how much smoother and repeatable a filter is. A filter is mathematically identical to using a smarter windowing function.
--
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
Yep, shown in my first post examples, where I showed various window widths. Did you not see that? ...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| San Tan Valley, AZ 85142 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 I saw was that you manually forced the test signal and the averaging window to align. Will you do that manually, locally, for every current and power average that you do on a varying-frequency switching regulator?
Filters just always work.
--
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
Nope. Where did you get that idea? I only selected the window width equal to the period. I suspect I can do it for a non-periodic situation as well, just do it in the Probe "Performance Analysis" window where searches can be done.
On a clocked system (as yours supposedly is), don't you know the period?
Sure they do. How come you got greater than 100% efficiency?
And my energy out / energy in approach is absolutely accurate, independent of period.
I think your "spot" efficiency concept is totally flawed... at least as you implemented it... individually "filtering" each signal, THEN dividing. And what absurd results do you get if you don't LIMIT?
Bwahahahaha! Now THIS is funny. Take away that LIMIT function and it's well-behaved (probably because the spike Larkin fretted over is quite narrow)...
Also one should not add the filters inside the actual functional schematic since they influence the result. Instead add the filtering off to the side as a measuring device. You don't need an R to measure current... just use -I(V2)*9
As for all your machinations to avoid a surge current, why not use one of your favorite depletion mode FET's feeding a common flyback switcher and avoid all your excess folderol? ...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| San Tan Valley, AZ 85142 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.
Some switchers are clocked, most aren't. None of the converters we've worked with lately are constant-frequency. Even if they were, independent of Vcc and temperature and load, you'd have to accurately measure the period and set your averaging to some exact multiple.
There are times, during startup and top-off idle, when the numerator/denominator of the divide both decay towards zero. I understand what's happening so it doesn't confuse me.
And doesn't display what I want to see.
OK, invent a better way to do it. Show us.
Occasional teravolt glitches for picoseconds at startup, crazy values after shutdown when Iin and Iout are both zero, stuff like that. LIMIT keeps Spice fron rescaling the graphs to silly values.
Some sims need LIMIT, some don't. So why not leave it in?
It's only a simulation; it's not real life.
Asshole.
--
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
It gave you >100%, leaving it out didn't. Explanation??
Your fall-back retort.
Designed anything recently that didn't require a tweak ?:-)
I knew you'd revert to name-calling, sooner or later :-( And you did copy it, didn't you ?:-)
You're not interested in a true face-to-face engineering discussion. You only want to dance around and self-declare yourself as the prima donna. And you absolutely can't tolerate criticism. ...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| San Tan Valley, AZ 85142 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.
Are you suggesting that the LIMIT function can generate bigger values than if it were not used? Post an example.
Only after you went into insult mode. I was talking on-topic.
No. Unless you think that computing 100 * Pout/Pin is some sort of violation of some trade secret. The first and second order lowpass current shunts are original to me, but hardly Nobel work. It's just a pi filter with the cap low side on Vin instead of ground.
Face to face? No thanks. You're a lot uglier than I am. Besides, you've threatened to shoot me, beat me up, and run over me with your walker.
You don't like being wrong any more than I do, but you do it more often.
--
John Larkin Highland Technology, Inc
jlarkin at highlandtechnology dot com
http://www.highlandtechnology.com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom laser drivers and controllers
Photonics and fiberoptic TTL data links
VME thermocouple, LVDT, synchro acquisition and simulation
Which suggests there's probably an LT part off-the-shelf plus a depletion-mode FET that fits the bill perfectly... no amateurish "Skeptic" configuration required. ...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| San Tan Valley, AZ 85142 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.
If I used a flyback, there would be no big powerup surge. That's unique to a single-inductor buck.
Show us.
--
John Larkin Highland Technology, Inc
jlarkin at highlandtechnology dot com
http://www.highlandtechnology.com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom laser drivers and controllers
Photonics and fiberoptic TTL data links
VME thermocouple, LVDT, synchro acquisition and simulation
--
John Larkin Highland Technology, Inc
jlarkin at highlandtechnology dot com
http://www.highlandtechnology.com
Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom laser drivers and controllers
Photonics and fiberoptic TTL data links
VME thermocouple, LVDT, synchro acquisition and simulation
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.