PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
OverUnity Research > Benches > Itsu (Moderator: Itsu) > Itsu's workbench / placeholder.
2024-05-02, 13:07:17
News: A feature is available which provides a place all members can chat, either publicly or privately.
There is also a "Shout" feature on each page. Only available to members.

Pages: 1 2 [3] 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 ... 39
Author Topic: Itsu's workbench / placeholder.  (Read 107679 times)
Topic Options img
img  poynt99
Re: Itsu's workbench / placeholder. « Reply #50 on: 2019-01-20, 14:47:55 »

Group: Administrator
Hero Member
*****

Posts: 3208
It's not as complicated as it may seem...
Quote from: TinMan on 2019-01-20, 12:12:40
Collector voltage is 0v + V/drop across CVR2 when transistor is on.
Base voltage is only 3v,as it is 6VPP,so 3v is a negative value.


Brad

Could you post a scope shot of your collector voltage? One or two pulses, thanks.

On a side note, what is the DC resistance and inductance of your coil?


---------------------------
"Some scientists claim that hydrogen, because it is so plentiful, is the basic building block of the universe. I dispute that. I say there is more stupidity than hydrogen, and that is the basic building block of the universe." Frank Zappa
   
img  TinMan
Re: Itsu's workbench / placeholder. « Reply #51 on: 2019-01-20, 14:58:57 »

Group: Elite Experimentalist
Hero Member
*****

Posts: 4609


Buy me some coffee
 author=poynt99 link=topic=3691.msg71360#msg71360 date=1547995675]

Quote
On a side note, what is the DC resistance and inductance of your coil?

Resistance is 1.9 0hms.
Inductance-no idea,as meter will not give me one.

Quote
Could you post a scope shot of your collector voltage? One or two pulses, thanks.

Yes,see below with attached schematic.


Brad


---------------------------
Never let your schooling get in the way of your education.
   
img  gotoluc
Re: Itsu's workbench / placeholder. « Reply #52 on: 2019-01-20, 15:10:36 »
Group: Elite Experimentalist
Hero Member
*****

Posts: 901
Hi Itsu and guys,

Since all the previous scope data is useless I decided to modify the input to the circuit to get a better scope reading.

All is described in this video: https://youtu.be/lBq0hV3VW-g

Also, all component values are in the schematic.

Regards
Luc
   
img  TinMan
Re: Itsu's workbench / placeholder. « Reply #53 on: 2019-01-20, 15:32:57 »

Group: Elite Experimentalist
Hero Member
*****

Posts: 4609


Buy me some coffee
Quote from: gotoluc on 2019-01-20, 15:10:36
Hi Itsu and guys,

Since all the previous scope data is useless I decided to modify the input to the circuit to get a better scope reading.

All is described in this video: https://youtu.be/lBq0hV3VW-g

Also, all component values are in the schematic.

Regards
Luc

Great video Luc
Opto isolated mosfet --well i guess that takes care of that  O0'



Brad


---------------------------
Never let your schooling get in the way of your education.
   
img  gotoluc
Re: Itsu's workbench / placeholder. « Reply #54 on: 2019-01-20, 15:33:15 »
Group: Elite Experimentalist
Hero Member
*****

Posts: 901
Quote from: Grumage on 2019-01-20, 12:10:38
Hi Guys.

Luc, are you pulsing the coil, Steel core and PM as per Brad's motor drive?

Could the tapering be the PM demagnetising?

Cheers Graham.

Hi Graham,

I am pulsing a coil of an off the shelf steel lamination toroid transformer, no magnets involved.
I use a very fast rising narrow pulse that is very accurately adjusted to bring the steel core to full magnetic saturation and instantly shut off.
This combination replicates what Brad was demonstrating but on a larger scale.
No magnets are required and if Brad removed his magnet I'm quite sure it would preform the same way as it did with the magnet.

Regards
Luc
   
img  gotoluc
Re: Itsu's workbench / placeholder. « Reply #55 on: 2019-01-20, 15:34:26 »
Group: Elite Experimentalist
Hero Member
*****

Posts: 901
Quote from: TinMan on 2019-01-20, 15:32:57
Great video Luc
Opto isolated mosfet --well i guess that takes care of that  O0'

Brad

It sure does mate ;)

Luc

PS. The MOSFET switch was designed by electronic wizar Jason Owen
   
img  poynt99
Re: Itsu's workbench / placeholder. « Reply #56 on: 2019-01-20, 15:36:44 »

Group: Administrator
Hero Member
*****

Posts: 3208
It's not as complicated as it may seem...
Quote from: TinMan on 2019-01-20, 14:58:57
author=poynt99 link=topic=3691.msg71360#msg71360 date=1547995675]

Resistance is 1.9 0hms.
Inductance-no idea,as meter will not give me one.

Yes,see below with attached schematic.


Brad

What was the FG setting (T and PW)? It looks to me like about a 300us PW.


---------------------------
"Some scientists claim that hydrogen, because it is so plentiful, is the basic building block of the universe. I dispute that. I say there is more stupidity than hydrogen, and that is the basic building block of the universe." Frank Zappa
   
img  Itsu
Re: Itsu's workbench / placeholder. « Reply #57 on: 2019-01-20, 15:41:13 »

Group: Moderator
Hero Member
*****

Posts: 4102
Here the LTspice sim of Luc his circuit.

Nice to see the sinewave like signal coming back on both csr signals.
So we have to let things settle down to get a good average reading, so i toke a 2s run.

First screenshot is with 200ms runtime (still settling down, so currents are still different).
Second screenshot is with 2s runtime, now things are stable, and current show the same average.

My floating FG will have the same effect i guess as Lucs Opto isolated MOSFET.
I used the 86Hz as we had it,  so not your ±10Hz

Itsu
   
img  poynt99
Re: Itsu's workbench / placeholder. « Reply #58 on: 2019-01-20, 16:18:10 »

Group: Administrator
Hero Member
*****

Posts: 3208
It's not as complicated as it may seem...
Itsu,

Another better option perhaps is to tell the simulator to ignore the first portion of the run.

It looks like things settle down pretty well after about 400ms, so try changing your .tran statement to the following:

.tran 0 600m 400m 10u

You only need about 200ms worth of data (600m-400m).


---------------------------
"Some scientists claim that hydrogen, because it is so plentiful, is the basic building block of the universe. I dispute that. I say there is more stupidity than hydrogen, and that is the basic building block of the universe." Frank Zappa
   
img  Itsu
Re: Itsu's workbench / placeholder. « Reply #59 on: 2019-01-20, 16:30:44 »

Group: Moderator
Hero Member
*****

Posts: 4102
Quote from: gotoluc on 2019-01-20, 15:10:36
Hi Itsu and guys,

Since all the previous scope data is useless I decided to modify the input to the circuit to get a better scope reading.

All is described in this video: https://youtu.be/lBq0hV3VW-g

Also, all component values are in the schematic.

Regards
Luc


Luc,

still seeing this fluctuating csr2 peaks which seems strange to me, not sure if they are like Poynt says artifacts. 
Not sure they will influence the readings.

Concerning the minimum cycles on screen, perhaps you should go in your trigger menu to NORMAL instead of AUTO mode
My scope does not enter Roll (your scan) mode then.


Itsu
   
img  poynt99
Re: Itsu's workbench / placeholder. « Reply #60 on: 2019-01-20, 16:31:16 »

Group: Administrator
Hero Member
*****

Posts: 3208
It's not as complicated as it may seem...
Quote from: TinMan on 2019-01-20, 14:58:57
author=poynt99 link=topic=3691.msg71360#msg71360 date=1547995675]

Resistance is 1.9 0hms.
Inductance-no idea,as meter will not give me one.

Yes,see below with attached schematic.


Brad
Plugging known data into the sim, and matching wave forms, I estimate an inductance of 6.5mH.

What is the total cycle time in that screen shot, and can you confirm the pulse width setting? (I estimate 320us-330us or so)


---------------------------
"Some scientists claim that hydrogen, because it is so plentiful, is the basic building block of the universe. I dispute that. I say there is more stupidity than hydrogen, and that is the basic building block of the universe." Frank Zappa
   
img  Itsu
Re: Itsu's workbench / placeholder. « Reply #61 on: 2019-01-20, 16:32:18 »

Group: Moderator
Hero Member
*****

Posts: 4102
Quote from: poynt99 on 2019-01-20, 16:18:10
Itsu,

Another better option perhaps is to tell the simulator to ignore the first portion of the run.

It looks like things settle down pretty well after about 400ms, so try changing your .tran statement to the following:

.tran 0 600m 400m 10u

You only need about 200ms worth of data (600m-400m).

Ok,  changed the run time (tran) to   0 600m 400m 10u

Still seeing the csr1 wobble.

Itsu
   
img  poynt99
Re: Itsu's workbench / placeholder. « Reply #62 on: 2019-01-20, 16:35:06 »

Group: Administrator
Hero Member
*****

Posts: 3208
It's not as complicated as it may seem...
That's fine.

The avg's are pretty close still, and you don't need to do such a long run.


---------------------------
"Some scientists claim that hydrogen, because it is so plentiful, is the basic building block of the universe. I dispute that. I say there is more stupidity than hydrogen, and that is the basic building block of the universe." Frank Zappa
   
img  poynt99
Re: Itsu's workbench / placeholder. « Reply #63 on: 2019-01-20, 16:54:02 »

Group: Administrator
Hero Member
*****

Posts: 3208
It's not as complicated as it may seem...
In case some may be wondering, the difference in power being dissipated when comparing csr1 and csr2 is not being questioned. The rms current in csr2 is much greater than in csr1, and therefore it is dissipating more power (and will be higher in temperature). The sims clearly show this.

What the sims don't show is a difference in avg current between the two csr's.


---------------------------
"Some scientists claim that hydrogen, because it is so plentiful, is the basic building block of the universe. I dispute that. I say there is more stupidity than hydrogen, and that is the basic building block of the universe." Frank Zappa
   
img  gotoluc
Re: Itsu's workbench / placeholder. « Reply #64 on: 2019-01-20, 17:14:05 »
Group: Elite Experimentalist
Hero Member
*****

Posts: 901
Quote from: Itsu on 2019-01-20, 16:30:44
still seeing this fluctuating csr2 peaks which seems strange to me, not sure if they are like Poynt says artifacts. 
Not sure they will influence the readings.

What I'll do is lower the input voltage and increase the on time. That should give the scope more time to provide stable data. However, keep in mind we are now changing the dynamics of the inductors effects just to accommodate the scopes inabilities to provide good data. So to me that would not be conclusive as to what is taking place using a narrow fast rising current pulse. A bit like saying the software simulation is real.

Quote from: Itsu on 2019-01-20, 16:30:44
Concerning the minimum cycles on screen, perhaps you should go in your trigger menu to NORMAL instead of AUTO mode
My scope does not enter Roll (your scan) mode then.

Please find the below shots with the trigger set to NORMAL instead of AUTO mode.

Regards
Luc
   
img  gotoluc
Re: Itsu's workbench / placeholder. « Reply #65 on: 2019-01-20, 17:49:27 »
Group: Elite Experimentalist
Hero Member
*****

Posts: 901
Quote from: poynt99 on 2019-01-20, 16:54:02
In case some may be wondering, the difference in power being dissipated when comparing csr1 and csr2 is not being questioned. The rms current in csr2 is much greater than in csr1, and therefore it is dissipating more power (and will be higher in temperature). The sims clearly show this.

What the sims don't show is a difference in avg current between the two csr's.

Thanks poynt for making that clear.
Now that I know what we are focused on I will try my best to provide that.

Here is a live video with more samples using Normal trigger mode: https://youtu.be/0byRTPYJ9f8

Regards
Luc
   
img  Itsu
Re: Itsu's workbench / placeholder. « Reply #66 on: 2019-01-20, 18:04:28 »

Group: Moderator
Hero Member
*****

Posts: 4102

Luc,

No, its not the meaning to change anything for the sake of correct measurements.
We need accurate data on the signals involved.

What are your FG (arduino) timings?   I see they changed from 10Hz to 30Hz now?
What was the on time, are they the same as Brad used 86Hz @ 5% ontime?

looking at your csr1 signal in your DSCO2106.jpg screenshot, its almost non existing compared to csr2.
So i think there is a problem for the scope to handle it.
Could you try like poynt said use 1 Ohm csr's?.

Or crank up the vertical amplitude on csr1 alone and measure that across many cycles.

So the NORMAL mode allowed you to use more cycles, thats good.

I still see the abnormal fluctuation of the csr2 amplitude, this is not right.
You are triggering on the almost non existing csr1 signal, could you change that to the csr2 signal,
perhaps that stabilizes that fluctuation.

Sorry for the questions etc., but we should be able to match the sim signals and data.


Itsu
   
img  Itsu
Re: Itsu's workbench / placeholder. « Reply #67 on: 2019-01-20, 18:15:16 »

Group: Moderator
Hero Member
*****

Posts: 4102

Forget that last question, i see you already did that (trigger on csr2) in screenshot DSC02103 in your post #52

Itsu
   
img  Itsu
Re: Itsu's workbench / placeholder. « Reply #68 on: 2019-01-20, 18:26:02 »

Group: Moderator
Hero Member
*****

Posts: 4102

Luc,

look at the screenshot below where you used many cycles (not samples).
The amplitude of the csr2 signal varies from 200mV till almost 1.5V in 7 cycles.

This cannot be thru, so there must be something wrong, the difference get worse the more cycles you put up,
so i think its something with the scopes ability to handle this.


Itsu
   
img  gotoluc
Re: Itsu's workbench / placeholder. « Reply #69 on: 2019-01-20, 18:31:59 »
Group: Elite Experimentalist
Hero Member
*****

Posts: 901
I changed C1 from 100,000uf to 3,900uf to allow a faster response between the power supply and CVR1 in hopes to get a better scope average reading.
Even though that has made CVR1's wave much more visible on the scope, the averaging is still not even.

Have a look at it live: https://youtu.be/q3Q9aGyYHNQ

Regards
Luc
   
img  gotoluc
Re: Itsu's workbench / placeholder. « Reply #70 on: 2019-01-20, 19:35:26 »
Group: Elite Experimentalist
Hero Member
*****

Posts: 901
Quote from: Itsu on 2019-01-20, 18:04:28
No, its not the meaning to change anything for the sake of correct measurements.
We need accurate data on the signals involved.

Yes I understand, it just feels like we need to change things a little too much but I know it has to be done.

Quote from: Itsu on 2019-01-20, 18:04:28
What are your FG (arduino) timings?   I see they changed from 10Hz to 30Hz now?
What was the on time, are they the same as Brad used 86Hz @ 5% ontime?

I didn't change it, it's just the scope doing that, so I probed the Arduino output and looks to be at 30.45Hz about 7.5ms on time and 25ms off time. See shot below. So it's not the same as Brad. I could bring it up to that frequency but the input and output would be in the high 40 watt range. I don't think that's needed to study this.

Quote from: Itsu on 2019-01-20, 18:04:28
looking at your csr1 signal in your DSCO2106.jpg screenshot, its almost non existing compared to csr2.
So i think there is a problem for the scope to handle it.
Could you try like poynt said use 1 Ohm csr's?.

A 1 Ohm CSR on the pulse side would affect the Inductors (0.25 Ohms) performance which would be doing what you said we won't do. However, my pi filter I use to accurately measure the input has a precision 1 Ohm resistor between the two pi caps. I can place probe 1 there and see what results that gives.

Quote from: Itsu on 2019-01-20, 18:04:28
Or crank up the vertical amplitude on csr1 alone and measure that across many cycles.

That I can try. So basically measure each separately and maximize the wave size of each so the scope has enough surface area of each.

Regards
Luc
   
img  Itsu
Re: Itsu's workbench / placeholder. « Reply #71 on: 2019-01-20, 19:54:01 »

Group: Moderator
Hero Member
*****

Posts: 4102
Quote from: gotoluc on 2019-01-20, 18:31:59
I changed C1 from 100,000uf to 3,900uf to allow a faster response between the power supply and CVR1 in hopes to get a better scope average reading.
Even though that has made CVR1's wave much more visible on the scope, the averaging is still not even.

Have a look at it live: https://youtu.be/q3Q9aGyYHNQ

Regards
Luc


Thanks Luc,

so we have 270 Ohm load resistor, 3900uF for C1 and still 27000uF for C2
Still running at 30Hz with ?? ontime.

The video shows still something is not right.

The input current you measure with your DMM shows 402mA (average).
The CSR2 with many cycles (plus the artifacts) show average (mean) 412mA (41.2/0.1) which is close, so good.

But csr1 (which should also be the same as your DMM (and csr2)) shows 97.9mA   (9.97/0.1).

You have to find out why csr1 is not giving the same average current as the DMM, you are the only one knowing your setup.


picture 1 is the DMM average input current
picture 2 the csr2 average current across a 0.1 Ohm resistor


Itsu
   
img  Itsu
Re: Itsu's workbench / placeholder. « Reply #72 on: 2019-01-20, 20:10:28 »

Group: Moderator
Hero Member
*****

Posts: 4102
Quote from: gotoluc on 2019-01-20, 19:35:26
Yes I understand, it just feels like we need to change things a little too much but I know it has to be done.

I didn't change it, it's just the scope doing that, so I probed the Arduino output and looks to be at 30.45Hz about 7.5ms on time and 25ms off time. See shot below. So it's not the same as Brad. I could bring it up to that frequency but the input and output would be in the high 40 watt range. I don't think that's needed to study this.

A 1 Ohm CSR on the pulse side would affect the Inductors (0.25 Ohms) performance which would be doing what you said we won't do. However, my pi filter I use to accurately measure the input has a precision 1 Ohm resistor between the two pi caps. I can place probe 1 there and see what results that gives.

That I can try. So basically measure each separately and maximize the wave size of each so the scope has enough surface area of each.

Regards
Luc


Luc,

Ok, you are running 30Hz @ 7.5ms on time =  25% on time.


In your post #28, #35 and #41 screenshots, the time between 2 peaks is about 110ms (10Hz) and in the lower right
corner it also shows 10Hz.
 
In your post #51 and further screenshots it shows about 33ms between 2 peaks, so 30Hz, also shown in the lower right corner.

So you are telling me the scope is doing that?

itsu
   
img  Itsu
Re: Itsu's workbench / placeholder. « Reply #73 on: 2019-01-20, 20:41:17 »

Group: Moderator
Hero Member
*****

Posts: 4102
Quote from: poynt99 on 2019-01-20, 14:47:55
Could you post a scope shot of your collector voltage? One or two pulses, thanks.

On a side note, what is the DC resistance and inductance of your coil?

Poynt99,

here some scopeshots of my 2N3055 transistor setup
Coil is 2.7mH @ 2.6 Ohm

screenshots show Collector / emitter signals (single and multiple) and FG input (base / collector) signals (single and multiple)

Basically 86Hz with 5% on time.


Itsu
   
img  Itsu
Re: Itsu's workbench / placeholder. « Reply #74 on: 2019-01-20, 21:16:48 »

Group: Moderator
Hero Member
*****

Posts: 4102

Zooming in on the bottom part of the transistor Q1 on time collector signal.
The between horizontal cursors part is the base voltage (1V) plus 0.6V  (1.6V).

What did F6FLT say:

Quote
What is the collector voltage of Q1?
If it is less than the base voltage+0.6v, then the pulses are rectified by the base-collector junction and the mean DC current goes through R7-L1-R6-R5 and can add up in R2.


That bottom part is less then the "base voltage+0.6v", so "then the pulses are rectified by the base-collector
junction and the mean DC current goes through R7-L1-R6-R5 and can add up in R2".
(R7, R6 and R5 were some dummy 0.1 csr resistors for current measurements in the simulation, so are not really existing).

So how can we translate this bottom 1.6V rectified base-collector part to an amount of current adding up in csr2?  :o


Itsu
   
Pages: 1 2 [3] 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 ... 39
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2024-05-02, 13:07:17