PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2024-12-03, 21:55:26
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 ... 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 34 35 36 37 38 39
Author Topic: Itsu's workbench / placeholder.  (Read 138173 times)

Group: Moderator
Hero Member
*****

Posts: 4160
Here i measured the input power and ouput power when in shorted situation (output bulb becomes the load).

First screenshot is the input power:
yellow: input voltage (20V rms)
green:  input current (31mA rms)
red:    input power (calculated by scope yellow x green: 506mW)


Second screenshot is the output power:
yellow: output voltage across the output bulb (2.2V rms)
green:  output current (50mA rms)
red:    output power (calculated by scope yellow x green: 111mW)

Itsu
   

Group: Moderator
Hero Member
*****

Posts: 4160
Hi Itsu,

Thanks for referring back to the earlier graph on the bulbs V-I characteristic. 

Regarding efficiency: we know that no claim was made on "OU" so this is a secondary question now,  I believe. 
There are oscillators with measured 89 - 90% efficiencies (in some cases 92-93%) like Class-E oscillators, here is a link to such:
https://www.researchgate.net/publication/326616551_A_Self-Tuned_Class-E_Power_Oscillator   

But let's stay with Nelson's interesting oscillator.   Could you charge a 470 uF capacitor from the output and monitor input current during the charging process like Nelson showed in his video series? 

Gyula




the below screenshot is the input current in green when charging a 470uF cap from 0 to 200V

Time base is 4s/div,   we start the circuit at 4s and it reached 200V about 29s later where i disconnected the cap.

I had to modify the base potmeter as in the earlier setting it would not oscillate with the 470uF cap attached. 

will do some more tests tonight.


Itsu
   
Full Member
***

Posts: 115


Buy me a drink

the below screenshot is the input current in green when charging a 470uF cap from 0 to 200V

Time base is 4s/div,   we start the circuit at 4s and it reached 200V about 29s later where i disconnected the cap.

I had to modify the base potmeter as in the earlier setting it would not oscillate with the 470uF cap attached. 

will do some more tests tonight.


Itsu

Hi Itsu ,

 forgive me for the late reply, but since yesterday I went back to my job after several days of quarantine , and I only get home from 19:00, which makes it harder to follow your work on the circuit like i wish ☺ .

I already had the opportunity to see a summary of your tests, and frankly, the efficiency of the circuit is apparently disappointing, however, I would like to ask your opinion regarding the following:

Since the circuit apparently has a high impedance output, and with a relatively high voltage, when the small lamp is used as a load (shorted output) , given its low resistance, will be the most correct approach to correctly measure at the output?

I say that because when bulb is short is equivalent to have a low impedance load, and the bulb becomes connected in parallel offering a very low resistance in the output  .

From what I could understand the output from the rectifier, it shows a high ripple voltage, and this ripple voltage is certainly non-sinusoidal AC with harmonic voltages (In your last video is visible the frequencies merged in many frequencies values)  , so the measurement  of small values of the ripple will be inaccurate i think.
I don't know if you are following my reasoning, but this observation seems pertinent to me.
I would also like to mention the following test that I had the opportunity to do that can somehow validate my point.

I used an industrial reference relay RM505730 (SCHRACK) ,which has a coil resistance of 7500 Ohm, and needs 2690 mW with minimal 184V to work;
To my amazement, the circuit powered with 21v and 20 milliamps, which is a mere 420mW, manages to operate in the output the relay without any type of limitation in their operation, which makes me think about the points I mentioned earlier regarding the circuit impedance.

I would like to hear your opinion on this topic.
Thanks in advance for your work;)


---------------------------
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: 115


Buy me a drink
I’m so sorry for your loss Nelson. My heart goes out to you and your family.

Many thanks Jim , its life, and we need to learn to deal with the adversities .
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: Moderator
Hero Member
*****

Posts: 4160

Hi Nelson,

good to hear you are resuming your regular job, i guess its the best for everyone to go "back to normal" quickly if possible.


Concerning the first measurement results of your circuit, let me point out that those are some initial measurements.

I think the idea needs to be like you mention yourself that we have to do a series of tests to see if we can exploid this somehow.

Matching the ouput for optimum power transfer is key here.

Hopefully some interest is invoked and we can get more reactions and suggestions on how to maximize the effect seen.

One thing i did not expected to see was that ripply output voltage / current.
Perhaps the used bridge is not able to handle the frequency used.



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

Posts: 1199

the below screenshot is the input current in green when charging a 470uF cap from 0 to 200V

Time base is 4s/div,   we start the circuit at 4s and it reached 200V about 29s later where i disconnected the cap.

I had to modify the base potmeter as in the earlier setting it would not oscillate with the 470uF cap attached. 

will do some more tests tonight.


Itsu

Hi Itsu,

Thanks for doing the capacitor charge-up test.   Just curious again: in the new base potmeter setting good for the cap charge, does it remain good for driving the bulb too?  Or you had to reset the pot again? 

When I earlier asked Nelson to load the output with resistors to halve his unloaded 203 VDC output, it was to get an impression on the approximate output impedance. If I recall, it was in the range of 20-25 kOhm? sorry I cannot recall more precisely. But it showed the output has a relatively high impedance which surely has to be matched.

The ripple voltage across the output may come via capacitor C7 (3.5 nF) which is in parallel with a diode of the diode bridge.  Nelson surely had a reason to use that capacitor.  8)   The ripple voltage can also come about due to the low value puffer capacitor, C6, 97.2 nF.  No need for changing the diode bridge for fast switching diodes yet. 

Nelson, it is very good you can carry on your job.

Greetings
Gyula
   
Full Member
***

Posts: 115


Buy me a drink
Hi Nelson,

good to hear you are resuming your regular job, i guess its the best for everyone to go "back to normal" quickly if possible.


Concerning the first measurement results of your circuit, let me point out that those are some initial measurements.

I think the idea needs to be like you mention yourself that we have to do a series of tests to see if we can exploid this somehow.

Matching the ouput for optimum power transfer is key here.

Hopefully some interest is invoked and we can get more reactions and suggestions on how to maximize the effect seen.

One thing i did not expected to see was that ripply output voltage / current.
Perhaps the used bridge is not able to handle the frequency used.



regards Itsu

Hi Itsu ,
Yes, despite being a person that love being at home, the lack of routines is necessary to maintain mental and social balance,in that way I can say, that I am happy to be back at work.
Concerning the measurements you have right . But I’m curious about that frequencies oscillations  :) that I saw in your shots , is really frustrating to me not able to track those points!:)

Well , seems not many people apparently interested in participate until now with exception to Gyula ; hope is not because my participation; I know I’m not so Popular :)  by the eyes of some people in this Forum , but that should not be a reason to people not get involved is my opinion, I hope I'm wrong …
Hope see more advances in next days Itsu , and I thank you by that .

Have a good night and 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: 4160

Gyula,

the new potmeter setting is OK for driving the bulb, just the input currents and ouput voltage are different, like 54mA unloaded, and 33mA shorted, while the unloaded output went down from 450V to 350V.


Uploading another video right now showing the input current / ouput voltage during a timed charge of the 470uF cap again to 200V.

The output bulb  barely lights up during charge of the cap.

Like to test out something with high impedance, but cannot come up with something i could use right now.

itsu
   

Group: Moderator
Hero Member
*****

Posts: 4160

Here again the timed charging of the 470uF cap to 200V on video:
 
https://www.youtube.com/watch?v=FbGbp9WBPMQ&feature=youtu.be


I zoomed in in the dark on the start moment only here:

https://www.youtube.com/watch?v=gw48IbNSJQQ&feature=youtu.be

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

Posts: 1199
....
I would also like to mention the following test that I had the opportunity to do that can somehow validate my point.

I used an industrial reference relay RM505730 (SCHRACK) ,which has a coil resistance of 7500 Ohm, and needs 2690 mW with minimal 184V to work;
To my amazement, the circuit powered with 21v and 20 milliamps, which is a mere 420mW, manages to operate in the output the relay without any type of limitation in their operation, which makes me think about the points I mentioned earlier regarding the circuit impedance.
....


Hi Nelson,

Regarding your test with that relay type, we need to focus on some details.  Here is the data sheet https://media.digikey.com/pdf/Data%20Sheets/Tyco%20Electonics%20AMP%20PDFs/RM5,6.pdf  for that relay. 

  Rated coil  power at 50/60Hz   for coil type 730 is  2.69 VA at 50 Hz and at 230 VAC rated voltage.  The 184 V is the minimal operational AC voltage at 50 Hz.  So the 2.69 VA power is valid for the coil DC resistance of 7500 Ohm in series with the coil L inductance which has an inductive reactance at 50 Hz and we do not know the phase angle between the current and the voltage. 

All I mean is that more rigorous output power measurement is needed,  you need to check the DC output voltage across the relay when you hook it up and also the DC current the relay coil draws from the output of the oscillator.  Ripple voltage may also be present.

But you may as well abandon the relay test and use a normal known value resistor as the load and check the voltage across the resistor, then calculate the power. (provided the ripple voltage is negligible small, this would need a scope I know).

Greetings, and Good night,
Gyula

   
Full Member
***

Posts: 115


Buy me a drink
Here again the timed charging of the 470uF cap to 200V on video:
 
https://www.youtube.com/watch?v=FbGbp9WBPMQ&feature=youtu.be


I zoomed in in the dark on the start moment only here:

https://www.youtube.com/watch?v=gw48IbNSJQQ&feature=youtu.be

Itsu


Hi Itsu ,
Just by curiosity :
Why your multimeter in first video at minute 1:45 show a voltage of 349V ? Seem happens when you stop the oscillator .. Really strange ... Did you know what happen ?
https://youtu.be/FbGbp9WBPMQ?t=105

Many thanks by the videos :)

Have a good night   


---------------------------
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: 1199
Gyula,

the new potmeter setting is OK for driving the bulb, just the input currents and ouput voltage are different, like 54mA unloaded, and 33mA shorted, while the unloaded output went down from 450V to 350V.


Uploading another video right now showing the input current / ouput voltage during a timed charge of the 470uF cap again to 200V.

The output bulb  barely lights up during charge of the cap.

Like to test out something with high impedance, but cannot come up with something i could use right now.

itsu

Okay Itsu, thank you for the tests.   

The output bulb has the chance to come up with a faint brightness when the empty 470 uF cap starts charging, this is ok. As it charges, the current exponentially reduces via the bulb, giving less and less chance to give light.

The charging time and the 470 uF capacitor are in relation with the output resistance of the oscillator, RC time constant calculation helps to approach the resistance.

Must finish now, thanks for your time.

Gyula
   
Full Member
***

Posts: 115


Buy me a drink
Hi Nelson,

Regarding your test with that relay type, we need to focus on some details.  Here is the data sheet https://media.digikey.com/pdf/Data%20Sheets/Tyco%20Electonics%20AMP%20PDFs/RM5,6.pdf  for that relay. 

  Rated coil  power at 50/60Hz   for coil type 730 is  2.69 VA at 50 Hz and at 230 VAC rated voltage.  The 184 V is the minimal operational AC voltage at 50 Hz.  So the 2.69 VA power is valid for the coil DC resistance of 7500 Ohm in series with the coil L inductance which has an inductive reactance at 50 Hz and we do not know the phase angle between the current and the voltage. 

All I mean is that more rigorous output power measurement is needed,  you need to check the DC output voltage across the relay when you hook it up and also the DC current the relay coil draws from the output of the oscillator.  Ripple voltage may also be present.

But you may as well abandon the relay test and use a normal known value resistor as the load and check the voltage across the resistor, then calculate the power. (provided the ripple voltage is negligible small, this would need a scope I know).

Greetings, and Good night,
Gyula

Hi Gyula ,
Yes you have right , I just use the example of relay because coil impedance theme and by their coil.
Maybe I will do that tests tomorrow like you suggest just by curiosity but only the voltage and current consumed by relay with my multimeter .
Many thanks Gyula  Have a good night  .


---------------------------
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: 4160

Hi Itsu ,
Just by curiosity :
Why your multimeter in first video at minute 1:45 show a voltage of 349V ? Seem happens when you stop the oscillator .. Really strange ... Did you know what happen ?
https://youtu.be/FbGbp9WBPMQ?t=105

Many thanks by the videos :)

Have a good night


Nelson,    well spotted,  i have the DMM across the ouput leads, with the 470uF parallel to that.

At 200V i disconnect the 470uF / 200V cap, so then the DMM only measures the unloaded output leads voltage which directly jumps to its max. 350V with this base pot setting.

itsu
« Last Edit: 2020-05-13, 09:59:27 by Itsu »
   

Group: Moderator
Hero Member
*****

Posts: 4160
I used a range of resistors on the output as a load and noted down the input and output voltages, currents and power levels., see graph below.


Resistors used are 0, 1, 10, 100, 1K, 10K, 100K, 1M and 10M ohm.
For calculating the input power i used P=U*I using my current probe to measure the rms current and de Fluke 179 DMM for the voltage.
For calculating the ouput power i used P=U²/R using a voltage probe /scope to measure the rms voltage.


EDIT: i added an efficiency colum and graph and changed the load range into a logarithmic scale.
         Also i measured the 470uF cap to be 431.8uF

Regards Itsu   
« Last Edit: 2020-05-13, 20:01:06 by Itsu »
   
Group: Tech Wizard
Hero Member
*****

Posts: 1199
Hi Itsu,

By using this calculator here:  https://www.digikey.com/en/resources/conversion-calculators/conversion-calculator-time-constant

the 29-31 second charging time you measured for the 470 uF capacitor at 200 V  gives a resistance range between 62 to 65 kOhm for the oscillator output impedance.
So the load resistance in that range looks like as optimum for the highest power output as per the measured charging time but is a little off for the latest in / out power results, but the difference is not that high at all and the output resistance can be nonlinear too. 
This means the efficiency of your oscillator can be in the range between 50-60 %.

Thanks again for your efforts.

EDIT: the tolerance of the 470 uF electrolytic capacitor normally has a wide range so it may also cause or contribute to the difference.

Gyula
« Last Edit: 2020-05-13, 19:16:27 by gyula »
   

Group: Moderator
Hero Member
*****

Posts: 4160

Gyula,

see some EDITs above, like the measured cap value (should make it worse).




I toke a look at the output voltage signal when loaded with a 1K load with and without C7 (4nF).

White is with C7,
yellow is without C7.

Shape and amplitude of the signals are almost identical, only the frequency differs (goes up without).

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

Posts: 1199
Hi Itsu,

Okay, thanks.  It is unusual that C7 pulls the oscillator frequency down but with its capacitive reactance it seems to connect the 1 kOhm load 'heavier' into the oscillator by shunting the diode in the diode bridge. 

This means now that the roughly 8 Vpp  ripple voltage is present because the puffer capacitor 97 nF is too small for the 1 kOhm load but if you increase it, then the loading and pulling effect may also increase what both the 1 k load and the presence of C7 causes. 

It is interesting that the highest efficiency comes at the 1 kOhm load this indicates a nonlinear, load dependent output resistance but this is eventually not surprising for an oscillator. 

Greetings
Gyula
   

Group: Moderator
Hero Member
*****

Posts: 4160

Thanks Gyula,

yes, it seems there is a nonlinear, load dependent output resistance which also causes the oscillator frequency to change.

Perhaps i can chart that frequency change and see if it might change any reactance which influences the load.

Itsu
   

Group: Moderator
Hero Member
*****

Posts: 4160
I measured the oscillating frequency under severall load conditions and added that data and graph to the below picture.
C7 was included again.

Its not that of a dramatic change that could account for a heavy reactance influence imo.

 
itsu
   
Group: Tech Wizard
Hero Member
*****

Posts: 1199
Hi Itsu,

You measured a higher than 2 times change in frequency and IMHO this is quite a change.   C.C   ;)
Coils change their inductive reactances more than 2 times while capacitors reduce their capacitive reactances to less than half for such > 2 times frequency increase.
Anyway,  many thanks for all your kind efforts!

Gyula
   

Group: Moderator
Hero Member
*****

Posts: 4160

Gyula,

you are quite right when looking at it that way (2 times change).

But the difference from 12Khz to 17Khz where the major efficiency change occures is not that big.

Itsu   
   
Full Member
***

Posts: 115


Buy me a drink
I measured the oscillating frequency under severall load conditions and added that data and graph to the below picture.
C7 was included again.

Its not that of a dramatic change that could account for a heavy reactance influence imo.

 
itsu

Hi Itsu,
Hope you're well,
My apologies for not being so active these last few days, but I feel exhausted after the end of the day since i return to work and with 10 hours of work .... uffff ....
But What a beautiful job you did :) It is very interesting data that you share in relation to the circuit. I think, I have entertainment for the weekend to analyze this data :) .

In my perspective after a superficial analysis of the data you shared,  the coil L2 / L3 is equivalent to 1:1 winding, and seems the inductance value of coil L2 change with relation to the load resistance value connected in the output of the L3 coil, causing the tank voltage in C4 increase or decrease depending on the resistive value of the load making the bias in base of Transistor being regulated by this action, make frequencies change . Just a superficial opinion :) .
Did you verify if the base of transistor have present negative values when the circuit has a load connected ? It will interest know that .

I will try in weekend make some experiments with my circuit; since you have now 1 operational circuit I think I could test more freely in mine .
Many thanks , by all your efforts , I really appreciate yours and Gyula, collaboration.
Have a good night Itsu and Gyula







---------------------------
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: 1199
Hi Nelson,

Yes, L2 / L3 is a 1:1 turns ratio transformer so the inductive reactance of the 13 mH coils reduces below 1 kOhm from the 2.2 kOhm (that manifests at 27 kHz) when the 1 kOhm load resistance is connected across the so far unloaded output.  This change then certainly affects the whole circuit operation. 

I wish you and also Itsu good night too.   O0

Gyula
   

Group: Moderator
Hero Member
*****

Posts: 4160
Hi Itsu,
Hope you're well,
My apologies for not being so active these last few days, but I feel exhausted after the end of the day since i return to work and with 10 hours of work .... uffff ....
But What a beautiful job you did :) It is very interesting data that you share in relation to the circuit. I think, I have entertainment for the weekend to analyze this data :) .

In my perspective after a superficial analysis of the data you shared,  the coil L2 / L3 is equivalent to 1:1 winding, and seems the inductance value of coil L2 change with relation to the load resistance value connected in the output of the L3 coil, causing the tank voltage in C4 increase or decrease depending on the resistive value of the load making the bias in base of Transistor being regulated by this action, make frequencies change . Just a superficial opinion :) .
Did you verify if the base of transistor have present negative values when the circuit has a load connected ? It will interest know that .

I will try in weekend make some experiments with my circuit; since you have now 1 operational circuit I think I could test more freely in mine .
Many thanks , by all your efforts , I really appreciate yours and Gyula, collaboration.
Have a good night Itsu and Gyula

Hi Nelson,

I am fine thanks.   No problem with your activity, i still remember how it was when worked 10 hours  ;)

I can make some measurements on the base during loading off course, will that be compared to ground or to emitter or?

 
The base signal compared to ground looks very much like the blue base signal shown in the bottom screenshot here:
https://www.overunityresearch.com/index.php?topic=3691.msg80986#msg80986

have a good night both.

Regards Itsu
   
Pages: 1 ... 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 34 35 36 37 38 39
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2024-12-03, 21:55:26