Re: Any Good Reason to Plumb Ground Out of Hier. Schem.
I'm surprised no one has worked with the origins of plumb. Plumbum
is the latin name for lead (hence the chemical symbol Pb). The
English word "Lead" is, of course,? a heteronym - a multi-functional
word with different pronunciations.
--
Regards,
Tony
On 04/02/2025 22:46, John Woodgate
wrote:
toggle quoted message
Show quoted text
The
only things that you can plumb in English are 'the depths'.
|
No, the way to deal with "errors" at the start of a simulation is to
just ignore them. Simply use the "Time to start saving data" option
in the .TRAN directive. We're talking about specifying the number of
cycles that .FOUR uses when it processes the waveform for harmonic
analysis.
Even when using the "Time to start saving data" technique, you
should not use (at least) the first cycle in a harmonic analysis
because there are always non-uniform samples. I'm not sure why, but
it is easy to see by switching on the "Mark Data Points" option in
the Waveform Viewer.
The solution to the "non-problem" of not simulating long enough is
to always use a calculated analysis time rather than some arbitrary
time. Most .TRAN analyses use a periodic source. Just make the
analysis time an integer multiple of this source's period. For the
life of me, I can't understand why everyone doesn't do this. At
least make the collected data time, i.e. ("Stop time" - "Time to
start saving data") an integer multiple of the period.
As for the .FOUR directive, I've never seen the "number of cycles"
option make any difference at all to the reported THD level unless
there is non-periodic noise in the waveform.
I never really use the .FOUR option for THD measurements, except for
occasional verification, because it can't be used for plotting. I
have made a this a feature request, but so far it hasn't happened.
There a number of other techniques that can be used for these types
of analysis, some of which I have already uploaded. Using a
fundamental notch filter is a pretty simple solution that also
allows for the simple plotting of the distortion residual, which is
visually informative. (This is also how traditional THD meters
worked.) It is also fairly simple to exactly replicate the .FOUR
process by harmonic decomposition using .MEAS:
.meas tran V1re avg -(V(out)-Vavg)*sin(360*time*Freq)
.meas tran V1im avg? (V(out)-Vavg)*cos(360*time*Freq)
.meas tran V1m param hypot(V1re,V1im)
[phase omitted]
.etc...
..to calculate the magnitude of each harmonic. The advantage of this
is that it can then be plotted, using the Right-click > Plot
.step'ed .MEAS Data process once the log file is viewed with Ctrl-L.
.FOUR is probably quicker because it uses compiled code, but it
isn't transparent.
--
Regards,
Tony
On 04/02/2025 18:18, John Woodgate
wrote:
toggle quoted message
Show quoted text
Using
more cycles gives an average level of each harmonic, which may
compensate for errors due to the 'start' epoch of the simulation.
I find it useful to set the.TRAN simulation to ignore the early
cycles that look distorted, but 'look' is sometimes not good
enough; some transient harmonic levels that are significant are
still being included. If I'm not confident about the reported
result, I change the .TRAN to save results later and/or simulate
for longer. The other aspect of .TRAN simulation time is that if
you set it too short it will not resolve very low frequency
components. For example, a 50ms simulation will not show a 10Hz
component.
|
Re: Any Good Reason to Plumb Ground Out of Hier. Schem.
That's just plumb silly.?
|
Re: Any Good Reason to Plumb Ground Out of Hier. Schem.
The only things that you can plumb in English
are 'the depths'.
On 2025-02-04 21:43, mstokowski via
groups.io wrote:
Oof, forgot the tongue in cheek. Mistakes were made.
Apologies.
?
As noted by others, plumbing is used sometimes to explain
basic DC electronics, not much beyond that.
?
That said, I liked the poetic (incorrect) use of plumb as a
verb in this case, but agree it does not easily translate.
--
Michael Stokowski
LTspice Team
Analog Devices Inc.
--
OOO - Own Opinions Only
Best Wishes
John Woodgate
Keep trying
|
Re: Any Good Reason to Plumb Ground Out of Hier. Schem.
Oof, forgot the tongue in cheek. Mistakes were made. Apologies.
?
As noted by others, plumbing is used sometimes to explain basic DC electronics, not much beyond that.
?
That said, I liked the poetic (incorrect) use of plumb as a verb in this case, but agree it does not easily translate.
--
Michael Stokowski
LTspice Team
Analog Devices Inc.
|
Re: In need of some zeners or maybe model them myself
On Tue, Feb 4, 2025 at 04:34 AM, John Woodgate wrote:
?
Yeah. I can't get anywhere with that site. They have their own simulators now but, nothing remotely hinting towards helping me find the type of old analog component that I need. I even searched through their community forum and some other related link but, nothing. Thanks anyway.
|
I just uploaded a new version of the LM317/LM337 power supply.
?
Fully functional Dual Power Supply using LM3x7 regulators.
Reference diodes replaced by single silicon diodes,
This works but may not provide the full adjustment range intended,
Values for Rpot chosen by successive approximation.
?
Dave
|
Re: Any Good Reason to Plumb Ground Out of Hier. Schem.
What Tony is saying is that you (eewiz) started this topic by replying to a message in another topic and changing the Subject line.? Please don't do that.
?
To those who do not use email programs like Thunderbird that properly thread replies, you probably can't tell the difference.? But to others, your message displays in the wrong place, as a reply to someone else's topic.? You hijacked their topic.
?
New topics must start by clicking "New Topic" (on the web) or "Start new message" or "Compose" or whatever in your email program.
?
This group's email address is simple enough to remember: LTspice (at) groups.io .? Add it to your Address Book.
?
Andy
?
|
Tony asked:
Why do you think more cycles would give you a better result?
That's an interesting question.
?
Theoretically, one ought to see identical results from either 1 cycle or 1000 cycles.? But it is sampled data and it's imperfect.? There might be some logic to using many cycles for the? .FOUR calculation, since it should randomize the samples over a cycle better.? Looking at it another way, it averages the calculated amplitudes over many cycles if (a big IF) the calculation over one cycle could be slightly off.
?
I'm hesitant to mention the following because I don't think it has much impact on this, but maybe it does when you are trying to see components that are way down in the mud, so to speak.? The timesteps are not uniform over a simulation.? LTspice sets them quite small at the start, and lets them expand until they reach Maximum Timestep or another factor puts a cap on them.? It briefly throttles the timestep again at the "Time to start saving data", and again at the end of the simulation.? It needs to hit those points exactly, without having an unintentional "timestep too small" error, so it throttles the timestep briefly as it's approaching either point.
?
Now if you are concerned that unequal timesteps might impair your Fourier analysis - which is perhaps not unreasonable when talking about parts-per-billion or so - then you might try to avoid applying the .FOUR both right after "Time to start saving data", and right before the end of the simulation.? The former, you can control.? The latter, you can't.? .FOUR always takes the last samples in a simulation, which theoretically might be ever-so-slightly altered (I hesitate to say "corrupted") because of the throttled timestep.
?
Averaging Fourier over several cycles might mitigate this effect because most of those cycles have uniform timesteps.? At least on paper it would.
?
Again, I emphasize that we're talking about down-in-the-mud concerns here, which might or might not change the results.
?
I haven't seen your full schematic, but did you remember to use .options plotwinsize=0 and set the maximum timestep to a good fraction of a cycle? I normally set Tmax to 1/period/2^n, with n normally between 8 and 16, depending on the expected level of THD with a suitable trade-off in analysis time. Normally Fourier stuff like things in binary multiples, but I understand LTspice's algorithms are much less fussy.
He did all that.? I think he used 2**16 for period-to-maximum-timestep ratio.? He also used the higher NUMDGT.
?
Andy
?
|
Using more cycles gives an average level of
each harmonic, which may compensate for errors due to the
'start' epoch of the simulation. I find it useful to set
the.TRAN simulation to ignore the early cycles that look
distorted, but 'look' is sometimes not good enough; some
transient harmonic levels that are significant are still being
included. If I'm not confident about the reported result, I
change the .TRAN to save results later and/or simulate for
longer. The other aspect of .TRAN simulation time is that if you
set it too short it will not resolve very low frequency
components. For example, a 50ms simulation will not show a 10Hz
component.
On 2025-02-04 16:57, Tony Casey wrote:
Did you try just letting .FOUR use its defaults?
My experience that it is pretty consistent. Why do you think more
cycles would give you a better result? Just make sure the waveform
display is comfortably more than the number of cycles the .FOUR
uses. I normally use the default of 1 cycle, but display between 3
and 6 cycles for good good visual. .FOUR will use the last cycle
of the waveform.
I haven't seen your full schematic, but did you remember to use
.options plotwinsize=0 and set the maximum timestep to a good
fraction of a cycle? I normally set Tmax to 1/period/2^n, with n
normally between 8 and 16, depending on the expected level of THD
with a suitable trade-off in analysis time. Normally Fourier stuff
like things in binary multiples, but I understand LTspice's
algorithms are much less fussy.
--
Regards,
Tony
On 04/02/2025 16:32, rlim701 via
groups.io wrote:
I spoke too
soon.? Still not getting consistent THD results.? The latest
zipped files (thd_debug.zip) show that changing .param
fft_cycles=67 brought back the 'thd=0.000000% line again even
though I added extra end time and also subtracted 1 cycle from
.FOUR.? Would appreciate other thoughts to at least get some
consistent results.? What I am essentially doing is tuning the
simulator via fft_cycles and fft_throw_cycles to get a
trustworthy thd number, but whenever I try that by increasing
either parameters, I seem to eventually end up back to all
zeros.
--
OOO - Own Opinions Only
Best Wishes
John Woodgate
Keep trying
|
Four hours ago, group member "EE20BT043.alum24" uploaded a schematic and two photos, about a circuit with parallel-connected MOSFETs..? Or maybe it's about series and parallel MOSFETs.
?
But she/he has forgotten to send any message here about it.
?
EE20BT043.alum24, it looks like all your parallel-connected N-channel MOSFETs are biased hard OFF so they never turn ON.? Is that what you wanted?? The gate voltage is much too low to turn them ON.? Their current is essentially zero.
?
EE20BT043.alum24, do you have a statement to make?? Are you asking any questions about your circuit?
?
Without a question or comments here in the messages, nobody can answer your unspoken question or concern.
?
What is going on?
?
?
Andy
?
|
Re: Any Good Reason to Plumb Ground Out of Hier. Schem.
There's always a good reason to start a new topic on a different
subject, rather than hijack existing one that's completely
unrelated. It may not matter to you, but it messes things up for
people that read messages in threads.
Please don't do this!
--
Regards,
Tony
On 04/02/2025 02:26, eewiz via
groups.io wrote:
toggle quoted message
Show quoted text
?
Is there any good reason to plumb the ground net (node 0) out
of a hierarchical schematic block?
Inside the block, the hierarchical label could be called RTN
and outside the block, the hierarchical label could be connected
to ground (node 0).
Assuming it is always connected to node 0 outside the block,
it appears to me that there is no good reason to do the
plumbing.
|
Did you try just letting .FOUR use its defaults?
My experience that it is pretty consistent. Why do you think more
cycles would give you a better result? Just make sure the waveform
display is comfortably more than the number of cycles the .FOUR
uses. I normally use the default of 1 cycle, but display between 3
and 6 cycles for good good visual. .FOUR will use the last cycle of
the waveform.
I haven't seen your full schematic, but did you remember to use
.options plotwinsize=0 and set the maximum timestep to a good
fraction of a cycle? I normally set Tmax to 1/period/2^n, with n
normally between 8 and 16, depending on the expected level of THD
with a suitable trade-off in analysis time. Normally Fourier stuff
like things in binary multiples, but I understand LTspice's
algorithms are much less fussy.
--
Regards,
Tony
On 04/02/2025 16:32, rlim701 via
groups.io wrote:
toggle quoted message
Show quoted text
I spoke too soon.?
Still not getting consistent THD results.? The latest zipped files
(thd_debug.zip) show that changing .param fft_cycles=67 brought
back the 'thd=0.000000% line again even though I added extra end
time and also subtracted 1 cycle from .FOUR.? Would appreciate
other thoughts to at least get some consistent results.? What I am
essentially doing is tuning the simulator via fft_cycles and
fft_throw_cycles to get a trustworthy thd number, but whenever I
try that by increasing either parameters, I seem to eventually end
up back to all zeros.
|
Re: Any Good Reason to Plumb Ground Out of Hier. Schem.
On Tue, Feb 4, 2025 at 11:30 AM, Jerry Lee Marcel wrote:
I must say I'd like a translation? pof plumbing in the realm of electronics.
Maybe I'm dumb, but I don't know what eewiz meant in this context.? Some things that came to my mind are these:
?
To "plumb" might mean to bring something straight to ground.? I'm thinking about a "plumb line", which is a carpenter's tool, not related to plumbing or what a plumber does.
?
To "plumb" might mean to direct or re-route something to somewhere else, through some other path.? This is what a plumber does with pipes.
?
Electronics is somewhat similar to the second version.? The wires are like pipes that move the electrons around.? That is the kind of thing you might hear once, and then never again, so it is no surprise if you never heard that.
?
Andy
?
|
Re: Any Good Reason to Plumb Ground Out of Hier. Schem.
DC circuits can be modelled as the flow of
water along pipes of different diameters and lengths, but
modelling AC circuits becomes very complicated and usually not
useful.
On 2025-02-04 16:30, Jerry Lee Marcel
via groups.io wrote:
I must say
I'd like a translation? pof plumbing in the realm of electronics.
--
OOO - Own Opinions Only
Best Wishes
John Woodgate
Keep trying
|
Re: Any Good Reason to Plumb Ground Out of Hier. Schem.
Not really. Simple DC circuits can use
plumbing analogies, as can EMI shielding, but AC circuits
require little-known fluid dynamics, so that plumbing analogies
are not helpful.
On 2025-02-04 16:20, mstokowski via
groups.io wrote:
“Electronics is
just plumbing,” is universally idiomatic, no?
--
OOO - Own Opinions Only
Best Wishes
John Woodgate
Keep trying
|
Re: Any Good Reason to Plumb Ground Out of Hier. Schem.
Le 04/02/2025 à 17:20, mstokowski via groups.io a écrit?: “Electronics is just plumbing,” is universally idiomatic, no? Certainly not. Never heard of it in 50 years of contact with english and american electronic designers. Honest, I fail to understand what it means.
|
Re: Any Good Reason to Plumb Ground Out of Hier. Schem.
I must say I'd like a translation? pof plumbing in the realm of
electronics.
Le 04/02/2025 à 17:20, mstokowski via
groups.io a écrit?:
toggle quoted message
Show quoted text
On Mon, Feb 3, 2025 at 07:45 PM, Andy I wrote:
Assuming it is always connected to node 0 outside
the block, it appears to me that there is no good
reason to do the plumbing.
There are actually some "tricks" going on, when a subcircuit
(or lower-level schematic) has a "ground" net that comes out
on a .subckt pin.? Some years ago I spent some time
documenting this, but did not finish.
Piqued my curiosity. I'd like love to see this.?
By the way, I am not sure if everyone here understands
your use of the words "plumb" or "plumb out".? Honestly, I
am not sure that I know exactly what you mean either.?
Remember that a sizable percentage of?the members of this
group are not in North America and do not know our slang.
“Electronics is just plumbing,” is universally idiomatic, no?
--
Michael Stokowski
LTspice Team
Analog Devices Inc.
|
Re: Any Good Reason to Plumb Ground Out of Hier. Schem.
On Mon, Feb 3, 2025 at 07:45 PM, Andy I wrote:
Assuming it is always connected to node 0 outside the block, it appears to me that there is no good reason to do the plumbing.
There are actually some "tricks" going on, when a subcircuit (or lower-level schematic) has a "ground" net that comes out on a .subckt pin.? Some years ago I spent some time documenting this, but did not finish.
Piqued my curiosity. I'd like love to see this.?
By the way, I am not sure if everyone here understands your use of the words "plumb" or "plumb out".? Honestly, I am not sure that I know exactly what you mean either.? Remember that a sizable percentage of?the members of this group are not in North America and do not know our slang.
“Electronics is just plumbing,” is universally idiomatic, no?
--
Michael Stokowski
LTspice Team
Analog Devices Inc.
|
I spoke too soon.? Still not getting consistent THD results.? The latest zipped files (thd_debug.zip) show that changing .param fft_cycles=67 brought back the 'thd=0.000000% line again even though I added extra end time and also subtracted 1 cycle from .FOUR.? Would appreciate other thoughts to at least get some consistent results.? What I am essentially doing is tuning the simulator via fft_cycles and fft_throw_cycles to get a trustworthy thd number, but whenever I try that by increasing either parameters, I seem to eventually end up back to all zeros.
Thanks,
Richard
|
Tony Casey
thutches
John Woodgate
Bell, Dave