PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2022-01-18, 03:54:35
News: Check out the Benches; a place for people to moderate their own thread and document their builds and data.
If you would like your own Bench, please PM an Admin.
Most Benches are visible only to members.

Pages: 1 ... 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 [18] 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
Author Topic: Itsu's workbench / placeholder.  (Read 47757 times)
Group: Tech Wizard
Hero Member
*****

Posts: 948
Hi Itsu,

I am not an expert either in using LTspice, I use it only occasionally.  There was an active forum on it in a separate  and open yahoo group and when Yahoo deleted all technical groups, they moved to here: https://groups.io/g/LTspice     
Sometimes I scroll the messages if I need answers to some problems, lots of topic have been accumulated during the years, lots of simulation problems have been discussed in connection with specific circuits. And there are experts on using the software and give advice. 
Membership is open for anyone.

I have spent some time on this oscillator but could not improve it yet.  Your latest asc file works for me and gives the same (small) output you get.

Gyula
   

Group: Moderator
Hero Member
*****

Posts: 3332

Hi Gyula,

ok,  thanks,  i will have a look overthere.

Concerning my latest .asc file, it still uses some guessed value's for the inductors, so it could be there is something to win.

Especially the T1 (L2/L3) value's could be way off, they even could have some transformer relation as i did find some CMC like components with a 1:4 or 1:10 relation and low ohmage.


Itsu 
   

Sr. Member
****

Posts: 273
Enjoy your trek through life but leave no tracks
Did you watch and pay atention to my short video ?
   

Group: Moderator
Hero Member
*****

Posts: 3332

Hi AG,   i did both.


Was there something special i needed to look at?


Itsu
   
Full Member
***

Posts: 114


Buy me a drink
Hi Itsu,

I am not an expert either in using LTspice, I use it only occasionally.  There was an active forum on it in a separate  and open yahoo group and when Yahoo deleted all technical groups, they moved to here: https://groups.io/g/LTspice     
Sometimes I scroll the messages if I need answers to some problems, lots of topic have been accumulated during the years, lots of simulation problems have been discussed in connection with specific circuits. And there are experts on using the software and give advice. 
Membership is open for anyone.

I have spent some time on this oscillator but could not improve it yet.  Your latest asc file works for me and gives the same (small) output you get.

Gyula


Hi Gyula and Itsu,
I am not a regular user of Ltspice, I usually do not do simulations, but I was curious about the "reliability" of simulation in cases of circuits, said to be less conventional.
Unfortunately I had an episode connected with health issues that made me still not test Ltspice more closely. Maybe Tomorrow i could return again and give a try with the simulation of the circuit .

Many thanks by the links , i already place all the libraries and now only the UF4007 ,  are fault .



---------------------------
Best Rewards
Nelson Rocha

" The goal is not to be successful, the goal is to be valuable.
Once you’re valuable, instead of chasing success,
it will attract itself to you. "
   
Group: Tech Wizard
Hero Member
*****

Posts: 948
Hi Nelson,

I hope you recover soon.
From this link http://ltwiki.org/?title=Components_Library_and_Circuits  you can download this zipped file
http://ltwiki.org/files/LargeCollection.zip  and there is the standard.dio file in it what you can copy into the cmp subdirectory of the LTSpice file in your   /user/Documents directory.   No need to edit the parameters in it, just copy and replace the old standard.dio with this new one, it includes UF4007 too.  I attach that particular file for you in a zipped format, includes only the standard.dio file.

Gyula
   
Full Member
***

Posts: 114


Buy me a drink
Hi Nelson,

I hope you recover soon.
From this link http://ltwiki.org/?title=Components_Library_and_Circuits  you can download this zipped file
http://ltwiki.org/files/LargeCollection.zip  and there is the standard.dio file in it what you can copy into the cmp subdirectory of the LTSpice file in your   /user/Documents directory.   No need to edit the parameters in it, just copy and replace the old standard.dio with this new one, it includes UF4007 too.  I attach that particular file for you in a zipped format, includes only the standard.dio file.

Gyula

Hi Gyula ,
I already did that when you publish your first post about this subject , but seems even  with that zip that include the majority of data the UF4007 give error.
I already delete the lib and put the lib from the zip but the same . I'm working in Mac and could be some problem in permissions of files .  I will need to search about this issue but many thanks ;) .

NEW DATA
I already confirm and the UF4007 is present in the lib but even that ltspice say that it could find it , lol .  i replace the diodes by a standard diodes , by suggestion of LTspice.
So it work now . Lets start test it !

   


---------------------------
Best Rewards
Nelson Rocha

" The goal is not to be successful, the goal is to be valuable.
Once you’re valuable, instead of chasing success,
it will attract itself to you. "
   
Group: Tech Wizard
Hero Member
*****

Posts: 948
Okay Nelson it is good that it works now.  Just an idea: if you right-click any one of those diodes then you can choose "Pick a new diode" icon in the appearing window and you may then choose another diode type from the list you know as a fast switching diode and replace the standard diodes you had to choose.

Here I attach the file I use from Itsu (he uploaded it above) and I noticed now that the 470 uF capacitor after the diode bridge across the DC output "suffocates" the circuit in the simulator and no DC output is received.  I mean if I use 470 pF instead of 470 uF temporarily, then a DC 12 V or so appears immediately across the output.  And we can continue from this to tweak the circuit and evaluate.  So far the 470 uF bogged the circuit down...  :D

In the screenshot below the green waveform is the AC voltage across L1, the blue trace is the DC output voltage and the red waveform is the collector current. 

Regards
Gyula
   
Full Member
***

Posts: 114


Buy me a drink
Okay Nelson it is good that it works now.  Just an idea: if you right-click any one of those diodes then you can choose "Pick a new diode" icon in the appearing window and you may then choose another diode type from the list you know as a fast switching diode and replace the standard diodes you had to choose.

Here I attach the file I use from Itsu (he uploaded it above) and I noticed now that the 470 uF capacitor after the diode bridge across the DC output "suffocates" the circuit in the simulator and no DC output is received.  I mean if I use 470 pF instead of 470 uF temporarily, then a DC 12 V or so appears immediately across the output.  And we can continue from this to tweak the circuit and evaluate.  So far the 470 uF bogged the circuit down...  :D

In the screenshot below the green waveform is the AC voltage across L1, the blue trace is the DC output voltage and the red waveform is the collector current. 

Regards
Gyula


Let's Test it ! thanks


---------------------------
Best Rewards
Nelson Rocha

" The goal is not to be successful, the goal is to be valuable.
Once you’re valuable, instead of chasing success,
it will attract itself to you. "
   

Group: Moderator
Hero Member
*****

Posts: 3332

Way to go guys.

I had the same problem using the big 470uF cap, so i used smaller ones before (6uF).
I think its caused by the slow charging rate (low output current).
If you wait enough (60 second or so) it will slowly have build up.

Not sure you guys already know, but:

doing a "cntr / LEFT mouse click" on the f.i. (red) Ic(Q1) readout in the trace screen, it will display the average and rms value of that trace.

doing a RIGHT mouse click only shows a screen where you can select Attached cursor (none, 1 or 2) to set in the trace screen to measure like frequency between 2 peaks etc.

pushing the left alt key when showing the current symbol over a component (like E1) shows a power meter and creates power trace.

Nice tool.


Itsu
   

Sr. Member
****

Posts: 273
Enjoy your trek through life but leave no tracks
The coil L1 (chokes and coils) some just dont hack it, Nelson used one of the larger cores around 4mh
RS do a 4.7mh part no 675 5324 @ 3.9R or a 3.3mh 675 5311 @ 2.37R nothing in between or wind it your self
but skrimping here will down grade the R BEMF.

The rest of the circuit distorts the wave form and ads another cycle in my device so one in two out
thats why i said look at the output wave.

So the device is nothing special as it stands but it's truly is a traing device, from an educational poit of view!

   
Full Member
***

Posts: 114


Buy me a drink
Way to go guys.

I had the same problem using the big 470uF cap, so i used smaller ones before (6uF).
I think its caused by the slow charging rate (low output current).
If you wait enough (60 second or so) it will slowly have build up.

Not sure you guys already know, but:

doing a "cntr / LEFT mouse click" on the f.i. (red) Ic(Q1) readout in the trace screen, it will display the average and rms value of that trace.

doing a RIGHT mouse click only shows a screen where you can select Attached cursor (none, 1 or 2) to set in the trace screen to measure like frequency between 2 peaks etc.

pushing the left alt key when showing the current symbol over a component (like E1) shows a power meter and creates power trace.

Nice tool.


Itsu

 Itsu , many thanks by the tips about hotkeys ;) really easier that way!  I start think the problem it's related to the fact that I'm running LTspice in Mac Os, because i not able to overcome some small issues .  Now i able to run but without same result you and Gyula have , i will post later some of my shots , but maybe i will try to virtualize a machine with windows and try in windows plataform .
Another question , is how is it possible to simulate the push bottom function, like in the real circuit  to start the oscillation? Is that possible ? Like i told yesterday , I don't usually use Ltspice to simulate circuits, and all free tips are welcome.

PS -Im waiting to receive a call from a friend of mine , to try borrow a LCR Meter . Maybe i could have today the right values  O0

« Last Edit: 2020-04-27, 14:47:34 by nelsonrochaa »


---------------------------
Best Rewards
Nelson Rocha

" The goal is not to be successful, the goal is to be valuable.
Once you’re valuable, instead of chasing success,
it will attract itself to you. "
   
Full Member
***

Posts: 114


Buy me a drink
The coil L1 (chokes and coils) some just dont hack it, Nelson used one of the larger cores around 4mh
RS do a 4.7mh part no 675 5324 @ 3.9R or a 3.3mh 675 5311 @ 2.37R nothing in between or wind it your self
but skrimping here will down grade the R BEMF.

The rest of the circuit distorts the wave form and ads another cycle in my device so one in two out
thats why i said look at the output wave.

So the device is nothing special as it stands but it's truly is a traing device, from an educational poit of view!

Hello AlienGrey, happy to see you here, and to know that you are trying to duplicate the circuit.  I'm trying to obtain a lcr meter to test the inductance from the coils , maybe today i will be able to measure the coils .   It seems to me that your guess about the coils is not very assertive, so I advise you to wait for official values.
If you  consider nothing special the circuit , and you feel is easy explain some characteristics shown, me and probably whoever follows the process of "dissecting" the circuit, would appreciate your opinion, being no less valid than what has already been mentioned.
We must not forget that you were one of those responsible, because I shared this same circuit at your request, right?   In this way, the best contribution you could give me was to participate too.  Many thanks




---------------------------
Best Rewards
Nelson Rocha

" The goal is not to be successful, the goal is to be valuable.
Once you’re valuable, instead of chasing success,
it will attract itself to you. "
   
Group: Tech Wizard
Hero Member
*****

Posts: 948
Hi Nelson,

I removed C9 which I put earlier in parallel with L2 as shown above in my previous post. And with these other capacitor values:
C3=100n
C4=10n ic=8 (initial condition 8V)
C5=10n
C6=220n
C8=470p

and with 1000 kOhm (i.e. no load) across the output, the DC voltage is 28V.

If I load the output with 10k resistor, the DC output voltage reduces to 14.8V. If I reduce the 10k to 9k, then the oscillator bogs down.
Note that the puffer capacitor is C5 only, (220 nF) and if I apply say 1 uF for C8, when the 10 k load is also there, the oscillator bogs down also.

This is a certain progress but a small one.  Perhaps your inductance measurements will bring simulation results much closer to your working circuit.    I attached the circuit file and 2 screenshots, the first is with 1000 k load, the second is with the 10 k load. The latter is intended to see the waveforms with the 10 k load versus that of the first without the load. There seems to be only amplitude differencies and a small frequency difference from no load 9.05 kHz to the loaded 9.175 kHz. 

Green is the voltage across L1, blue is output DC voltage, red is the collector current and yellow is the base current. 

It is interesting, that current flows in diodes D1 and D2 only during the initial condition in C4 takes place which lasts for a few us, then never again during the full 10 ms transient analysis time. 

Gyula
   
Group: Tech Wizard
Hero Member
*****

Posts: 948
Way to go guys.

I had the same problem using the big 470uF cap, so i used smaller ones before (6uF).
I think its caused by the slow charging rate (low output current).
If you wait enough (60 second or so) it will slowly have build up.

Not sure you guys already know, but:

doing a "cntr / LEFT mouse click" on the f.i. (red) Ic(Q1) readout in the trace screen, it will display the average and rms value of that trace.

doing a RIGHT mouse click only shows a screen where you can select Attached cursor (none, 1 or 2) to set in the trace screen to measure like frequency between 2 peaks etc.

pushing the left alt key when showing the current symbol over a component (like E1) shows a power meter and creates power trace.

Nice tool.

Itsu

Hi Itsu,

Thanks for the tips, the power meter and the power trace created is what I did not know about, a useful tool indeed.   O0

One notice to the average and rms value displays:  it always calculates them according to the displayed number of cycles, i.e. if you say expand a waveform to see it closer its shape in detail so that there are say 2 or 3 cycles remaining instead of say the total 15, then the values may differ from the ones calculated for the total simulation time for the 15 cycles. 

Gyula   
   

Group: Moderator
Hero Member
*****

Posts: 3332
Itsu , many thanks by the tips about hotkeys ;) really easier that way!  I start think the problem it's related to the fact that I'm running LTspice in Mac Os, because i not able to overcome some small issues .  Now i able to run but without same result you and Gyula have , i will post later some of my shots , but maybe i will try to virtualize a machine with windows and try in windows plataform .
Another question , is how is it possible to simulate the push bottom function, like in the real circuit  to start the oscillation? Is that possible ? Like i told yesterday , I don't usually use Ltspice to simulate circuits, and all free tips are welcome.

PS -Im waiting to receive a call from a friend of mine , to try borrow a LCR Meter . Maybe i could have today the right values  O0


Nelson,

it could be there are some differences in using LTspice on a MAC, but basically it should be the same i think.

Concerning the switch (push button), there is a switch available in LTspice, but its operated by timing, not by hand.

I tried to incorperate that switch, but it did not work properly, so member Frederik added a working switch and also included some LTSpice statements for using the MJE18008 and UF4007, see his .asc file here:
 
https://www.overunityresearch.com/index.php?topic=3691.msg80989#msg80989


Lateron i received a tip from member poynt99 that you can also use the "initial condition" attribute for a cap or coil (see partzman post also there).
So now i use this IC attribute for C4 (IC=1 or 8 like Gyula did) to set an initial voltage (1V or 8V) on C4 so to simulate the switch putting a voltage on C4.


Great for the LCR meter, i wonder what those value's are.


Itsu
   
Full Member
***

Posts: 114


Buy me a drink

Nelson,

it could be there are some differences in using LTspice on a MAC, but basically it should be the same i think.

Concerning the switch (push button), there is a switch available in LTspice, but its operated by timing, not by hand.

I tried to incorperate that switch, but it did not work properly, so member Frederik added a working switch and also included some LTSpice statements for using the MJE18008 and UF4007, see his .asc file here:
 
https://www.overunityresearch.com/index.php?topic=3691.msg80989#msg80989


Lateron i received a tip from member poynt99 that you can also use the "initial condition" attribute for a cap or coil (see partzman post also there).
So now i use this IC attribute for C4 (IC=1 or 8 like Gyula did) to set an initial voltage (1V or 8V) on C4 so to simulate the switch putting a voltage on C4.


Great for the LCR meter, i wonder what those value's are.


Itsu

Itsu and Gyula ,
Thank you very much for your patience, and yes is a bit different from Windows version ,I can assure you of that.
On Mac, it seems that there are 3 locations for the libraries, and I finally managed to get it working without component limitations as before.
I had to use the netlist command and read the code, to understand where LTspice was going to look for libraries. After that everything became easier, having just copied the libraries that previously included MJ18008 and UF4007. I also added a large collection of information and examples from the link that Gyula provided.
Subject closed on this topic :) from Ltspice  :P
Regarding the measurement of coils inductance, I hope to have some updates today.  O0



---------------------------
Best Rewards
Nelson Rocha

" The goal is not to be successful, the goal is to be valuable.
Once you’re valuable, instead of chasing success,
it will attract itself to you. "
   
Full Member
***

Posts: 114


Buy me a drink
Hi Itsu and Gyula ,
I can now measure the coil inductance values with a LCR meter borrowed from my friend Luis ;) nice Guy the equipment is  VICI DM4070.
The L1 have 3,07mH  and L2/L3 have each one 13mH.  I tried to do some simulations with LTspice, but so far I have not managed to keep the oscillation active.
However, when doing Transient Analysis simulation,
with a lower time value, you can see something strange or promising.
Given my lack of experience in Ltspice, I would like to have your opinion, Gyula and Itsu.
I have attached some shots, which I would like you to comment on.
I am going to dinner now, I am waiting for your comments. Thank you


---------------------------
Best Rewards
Nelson Rocha

" The goal is not to be successful, the goal is to be valuable.
Once you’re valuable, instead of chasing success,
it will attract itself to you. "
   

Group: Moderator
Hero Member
*****

Posts: 3332
Hi Nelson,

thanks for the value's, now we can close in on a real replication, searching my junkbox......


Concerning your sim,  i was able to get it oscillating with your same value's (L's and C's).
Green is the emitter  to ground trace.


But i don't understand your .tran statement.   
Mine is at least 3 attributes (stop time,  time to start saving data and max steptime) while i only see 2 in yours and the 500u seems the be the stop time.

Could you specify which are what?

Also not known in the trace screenshots is what is what.

Anyway, i think you point to the negative value's in your traces, which could be normal, looking to the current symbol through the bulb (E1).


Indeed much different layout on the MAC as in Windows.


Itsu
   

Group: Moderator
Hero Member
*****

Posts: 3332

By the way,  when placing a voltage probe (red probe), it is always compared to ground level.
If you DON'T want it to be compared to ground level, you can HOLD the left mouse button and drag the mouse pointer to the other point you want as reference (voltage probe turns black) and release the button.

It will now measure between those set points.


Itsu
   

Group: Moderator
Hero Member
*****

Posts: 3332
Could it be that your sim is not oscillating because your selected components (L1, L2/L3, C4 etc.) lack some real life properties like series resistance, parallel resistance, parallel capacitance)?

I normally use only components from the sim database which have real life properties specified, like this L1, see screenshot below.

I only editted the inductance value and series resistance to match.


Itsu
   
Group: Tech Wizard
Hero Member
*****

Posts: 948
Hi Folks,

I run the sim with the measured L values and the capacitor values shown in the MAC schematic. (By the way, Nelson, with the meter you may wish to check the nF capacitor values  too.)

I think in the .tran statement in the MAC schema the  50ms means the Stop time and the 50 us means the Maximum Time Step within the 50 ms.

I found the oscillator stops at 18.7 ms as you can see below. When I change the 400 uF puffer cap to 200 uF, then oscillator stops at around 10 ms, strangely enough.

Nelson, I will try to analyze your sim displays later on, please try to make some more comments too.

Gyula
   
Full Member
***

Posts: 114


Buy me a drink
Hi Nelson,

thanks for the value's, now we can close in on a real replication, searching my junkbox......


Concerning your sim,  i was able to get it oscillating with your same value's (L's and C's).
Green is the emitter  to ground trace.


But i don't understand your .tran statement.   
Mine is at least 3 attributes (stop time,  time to start saving data and max steptime) while i only see 2 in yours and the 500u seems the be the stop time.

Could you specify which are what?

Also not known in the trace screenshots is what is what.

Anyway, i think you point to the negative value's in your traces, which could be normal, looking to the current symbol through the bulb (E1).


Indeed much different layout on the MAC as in Windows.


Itsu


Itsu ,
About  tran statement :
Step 50 ms ending at 50us - I will add the same tran statement like yours .

About the trace screenshots:

The are labeled on the top V1 to the input  and E2 is the output bulb .  The other formula you see is the power  calculation that tell you the Wattage output  .
About the properties of components i did not change anything in the values  .  Only by example i left shots of  l1  values .
I go correct the tran statement and i already post .



---------------------------
Best Rewards
Nelson Rocha

" The goal is not to be successful, the goal is to be valuable.
Once you’re valuable, instead of chasing success,
it will attract itself to you. "
   

Group: Moderator
Hero Member
*****

Posts: 3332
Hi Folks,

I run the sim with the measured L values and the capacitor values shown in the MAC schematic. (By the way, Nelson, with the meter you may wish to check the nF capacitor values  too.)

I think in the .tran statement in the MAC schema the  50ms means the Stop time and the 50 us means the Maximum Time Step within the 50 ms.

I found the oscillator stops at 18.7 ms as you can see below. When I change the 400 uF puffer cap to 200 uF, then oscillator stops at around 10 ms, strangely enough.

Nelson, I will try to analyze your sim displays later on, please try to make some more comments too.

Gyula


Gyula,

mine stops also around 20ms when having the same value's and having series resistance of 1.6 Ohm for L1.

But L1 series resistance should be 2.6 Ohm as measured by Nelson.
It then stops even earlier.

But lowering the coupling factor to 0.7 of L2/L3 makes it run again.

Itsu
   
Group: Tech Wizard
Hero Member
*****

Posts: 948
This is the input power during the 10 ms oscillation time (with 200 uF puffer cap across the output).

The negative input power is because the simulator consider power taken out from voltage or current source as consumed hence gives it a negative sign.  (not to be mixed up with the oscillator producing power backwards).
   
Pages: 1 ... 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 [18] 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2022-01-18, 03:54:35