Author Topic: Proper phase lead adjustment at the input  (Read 559 times)

Offline Chrisader

  • High Voltage Enthusiast
  • *
  • Posts: 11
  • Karma: +0/-0
    • View Profile
Proper phase lead adjustment at the input
« on: July 20, 2021, 10:06:14 PM »
Hi!
I decided to use AD790 comparator in my first project of DRSSTC.
Below is circuit of the input stage of the AD790:



I have some strange behavior with my input and problem to get enough phase lead to achieve zero current switching. I need to put 150 ohms in place R24 (burden resistor), because if I put much lower value like 1 ohm or few ohms the all stuff doesn’t run/excite. With 100 ohms of R24 layout will run but slightly with worse performance.  I need to put also R5 potentiometer, primarily to set proper adjustment/delay but if in these place is no any resistance also layout doesn’t want to excite. There need to be between few hundred ohms to 1K or little more.
 Zener diodes cuts excessive voltage to prevent input of AD790 from damage. Maybe they do any shortcut of signal but I’m afraid to remove them and then destroy AD790.
It is conditions before attempts tuning phase lead...
I tried several combinations of burden L and R, but mentioned 150 ohms of R24 and inductor like axial choke which has 1.2mH inductance and 16.9 ohms of resistance. It gives me best results at this time but it is not sufficient. Before tuning I had more than 500nS delay, after add 1.2mH 16.9R inductor I have about 200nS delay. I want achieve perfect zero switching to avoid destroy my power transistors at full power. I run everything on 12V to prevent from any damage.

Below screens of scope before and after my attempts to compensate delay in 1.2mH 16.9R inductor and 150R parallely...
There still 200nS delay but spikes are much smaller...



Anybody had similar problem or know how to solve it?
It is possible to make computer simulation to find best burden L and R or L/R?
Maybe exist any calculator to find best values?
Any advice will be very helpfull...

« Last Edit: July 20, 2021, 10:09:29 PM by Chrisader »

Offline davekni

  • High Voltage Expert
  • ******
  • Posts: 1296
  • Karma: +62/-0
  • Physicist, engineer (electronic), and hobbiest
    • View Profile
Re: Proper phase lead adjustment at the input
« Reply #1 on: July 21, 2021, 04:57:04 AM »
I think your circuit as shown in the right side scope plots is actually working successfully.  It appears that you are scoping current with the same CT output used in the circuit shown (scope across R24).  That CT output will lead the phase of actual current due to the 1.2mH burden inductance, which makes it appear as if you don't have sufficient phase lead.  H-Bridge output scope plot indicates roughly correct phase lead (bridge output voltage transition due to remaining primary current, then current reversing just as the opposite IGBT starts conducting).

BTW, for future posts, it helps to state explicitly which signals are being scoped.  I'm guessing in my above response based on studying plot details.

In spite of correct operation, I'd recommend a few changes.  Most important, the 5V zener diodes D19 and D20 will be marginal protection for comparitor input.  Forward drop of one diode plus zener voltage of the other diode will allow voltage above supply rails.  I suggest instead a set of parallel signal diodes, one each direction, from comparitor input to ground.  That will clamp to much lower voltage, but still plenty sufficient for switching the comparitor.

Look at UD2.7 schematics for ideas, but don't copy everything.  You have an advantage with a comparitor running around ground.  One UD2.7 part I would copy is the added AC-coupled hysteresis, series capacitor and resistor from comparitor output to comparitor + input.  (The above parallel diode clamping is similar to UD2.7, but easier with your ground-referenced input.)

Another UD2.7 piece to copy is removing R24 (or make it a much higher value).  R24 is reducing phase lead.

Concerning failure to start, that issue may be a missing H-Bridge output load resistor.  If there is insufficient DC load on the H-Bridge output, charge on the MMC may leave the H-Bridge voltage initial state matched to the first drive state, generating insufficient feedback signal to start oscillation.  Testing at somewhat higher bridge voltage will help too.

There should be no need to make R5 adjustable once other issues are resolved.  1k is fine.  That presumes updated comparitor input protection diodes are reasonably-low capacitance (ie. signal diodes, 1N4148 or 1N914 or ...).

To scope current properly, either use a second CT, or scope across R1 with R24 removed.  (That is another reason to remove R24.)  UD2.7 uses 51 ohms for R1 burden resistor.  10 ohms may be fine depending on your CT ratio and expected primary current.  Don't know what value you have now.
« Last Edit: July 21, 2021, 05:04:13 AM by davekni »
David Knierim

Offline Chrisader

  • High Voltage Enthusiast
  • *
  • Posts: 11
  • Karma: +0/-0
    • View Profile
Re: Proper phase lead adjustment at the input
« Reply #2 on: July 21, 2021, 08:15:00 AM »
Hi!
Yellow trace is current measured from same CT used for feedback.
Blue trace is voltage measured on fullbridge inverter.
Current transformer has 60 windings on TX32/19-3E25 Ferroxcube core (white painted with orange one side).
Summarizing:
1. Zener diodes 0.5W gives poor protection and are harmful in any way for (distort/delay) signal?
2. Current transformer may not give to much current and voltage, because power stage is feeded from 12V supply until I make proper phase adjustment.
3. AD790 configuration seems to be fine and I based on predikter driver. Should I something to change?
4. In R24 case higher resistance gives more delay of phase lead?
5. It is good to use instead R24, R1 and Burden Coil use RF choke with high L and R just like this kind of chokes:
https://www.bourns.com/data/global/pdfs/70f_series.pdf
6.  I'm just only struggling in innsufficient phase lead. Some people in Russia founded for it special calculator, but I don't know russian and cannot understand how to use it:
https://tqfp.org/calculatoria/sd_predictor/?f=80&t_predict=300&r_predict=50&r_delay=1k
 

Offline davekni

  • High Voltage Expert
  • ******
  • Posts: 1296
  • Karma: +62/-0
  • Physicist, engineer (electronic), and hobbiest
    • View Profile
Re: Proper phase lead adjustment at the input
« Reply #3 on: July 22, 2021, 06:39:37 AM »
Quote
1. Zener diodes 0.5W gives poor protection and are harmful in any way for (distort/delay) signal?
Not likely harmful, although some zener diodes have significant capacitance that would add extra phase lag.

Quote
2. Current transformer may not give to much current and voltage, because power stage is feeded from 12V supply until I make proper phase adjustment.
Yes.

Quote
3. AD790 configuration seems to be fine and I based on predikter driver. Should I something to change?
Mostly fine.  Some AC hysteresis (series R + C from output to positive input would help - prevent possible issues with high-frequency oscillation before feedback starts that could fry the gate driver chip(s).  Look up UD2.7 schematic for reference.

Quote
4. In R24 case higher resistance gives more delay of phase lead?
Higher resistance will increase phase lead (more negative delay).  UD2.7 has no such resistor (infinite R24 value).

Quote
5. It is good to use instead R24, R1 and Burden Coil use RF choke with high L and R just like this kind of chokes:
https://www.bourns.com/data/global/pdfs/70f_series.pdf
No.  You are after about 600ns of phase lead to compensate for 500ns driver delay and end with slight phase lead.  If you remove R24 and use 50 ohms for R1, then you need about 30 ohms of L1 reactance, or slightly over 30uH inductance at your ~155kHz frequency.  Perhaps a bit higher if the inductor adds significant DC resistance in series with R1.

Quote
6.  I'm just only struggling in innsufficient phase lead. Some people in Russia founded for it special calculator, but I don't know russian and cannot understand how to use it:
Your problem is not directly with phase lead.  Your problem is incorrect measurement of phase lead.  You need to scope current as I described in my last post.  One more current scoping option came to mind.  You could add another small (say 10 ohm) resistor in series with the ground lead of the CT secondary, then scope across that resistor.  As long as your CT inductance is high enough (likely fine), the added 10 ohms won't affect feedback behavior.

Quote
Current transformer has 60 windings on TX32/19-3E25 Ferroxcube core (white painted with orange one side).
Is this a single-stage current transformer?  If so, you will have trouble once current gets high.  What primary coil current are you planning?  For example, 300A would be 5A in CT secondary.  Could be workable, but requires much lower R1 value to handle 5A, and correspondingly lower L1 value.  BTW, what is the value of R1 now?
David Knierim

Offline Chrisader

  • High Voltage Enthusiast
  • *
  • Posts: 11
  • Karma: +0/-0
    • View Profile
Re: Proper phase lead adjustment at the input
« Reply #4 on: July 25, 2021, 04:16:11 PM »
Thanks davekni!
Your posts is helpfull:-)
My CT is single stage and at this moment I don’t want excess 300A absolutely. I suppose that is impossible in 12V power supply. It just first step for me to make fully proper work of logic.

I do some changes in my AD790 input and I hope correct scope measurements:






 
I put two resistors 180 ohms to check what happen. One man said me that to make possible phase lead the waveform has to be fully sine. Distorted sine wave in any way cannot be leaded.
I did measurments of bridge voltage (1 blue color on diagram) and current yellow directly on terminals of CT connected to AD790. On terminals in front of R2 180 ohms I have fully sine wave, but through I have distorted by 15V zener diodes what is correct. Some strange thing is with negative half wave, because it has about 8-12 volt. I can remove R2 and feedback works/excites but then I have distorted wave in all circuit.

Questions:
1.   Now measurements of scope are correct?
2.   If low value of R24 delays phase lead then is necessary to put some resistance in series like R2 to prevent wave from distortions and limits input current for AD790 but not to large what I suppose delays the signal?
3.    I have not galvanic isolation between logic state and power state. Everything is on the one board. Can it make any unwanted coupling or influence?
4.   You mentioned about 50 ohms resistor and about 30uH coil with 30 ohms reactance. In my LC circuit frequency is about 70-75 kHz so parameters of the burden coil are depend from frequency. There are inductance, resistance, capacity inductance/resistance, own resonant frequency... I thing any simulation would be helpfull to find or design best coil for it?

Offline davekni

  • High Voltage Expert
  • ******
  • Posts: 1296
  • Karma: +62/-0
  • Physicist, engineer (electronic), and hobbiest
    • View Profile
Re: Proper phase lead adjustment at the input
« Reply #5 on: July 25, 2021, 10:15:30 PM »
Quote
1.   Now measurements of scope are correct?
You have explained well exactly what signals you are measuring.  Thank you.  However, those signals are not useful for measuring phase lead.  The signal in front of R2 leads in phase the actual DRSSTC primary current due to L1.  To measure phase lead, one scope channel needs to measure actual coil current, not a signal after phase lead has been applied.  Use either another CT or a resistor in series with the ground lead of your existing CT secondary.

Quote
2.   If low value of R24 delays phase lead then is necessary to put some resistance in series like R2 to prevent wave from distortions and limits input current for AD790 but not to large what I suppose delays the signal?
Low R24 reduces phase lead.  Most drivers do not have any resistor where you have R24.
R2 along with clamp diodes after it are necessary to protect AD790 input from excess voltage.  Distortion after R2 is normal, not a problem.  Phase lead is done before R2.  To avoid distortion on the phase lead circuit, it is normal for R2 to have a much higher value than R1.  For example, UD2.7 uses 51 ohms for R1 and 1k for R2.

Quote
3.    I have not galvanic isolation between logic state and power state. Everything is on the one board. Can it make any unwanted coupling or influence?
Depends on your layout and where power connections are made.  May be fine or may be a problem.  So far I haven't seen anything in your scope traces to indicate a problem.

Quote
4.   You mentioned about 50 ohms resistor and about 30uH coil with 30 ohms reactance. In my LC circuit frequency is about 70-75 kHz so parameters of the burden coil are depend from frequency. There are inductance, resistance, capacity inductance/resistance, own resonant frequency... I thing any simulation would be helpfull to find or design best coil for it?
Oops, my mistake on frequency.  I see your 70kHz frequency in scope traces.  It doesn't change inductance/resistance calculation very much.  You still want L/R to be about 600ns, a bit more at higher frequency.
Yes, I highly recommend simulation!  Simulation allows understanding the circuit better too.  For example, in simulation, you can plot actual primary current and the CT output voltage to compare phase.  My favorite free simulator is LTSpice, but there are many others.

Are you actually using FETs for a DRSSTC?  Or does your schematic show FETs because simulation is easier with FETs.  For the same package size and price, IGBTs handle much higher peak current.  However, IGBTs need 15Vge or higher, so require increasing your gate drive voltage.

Quote
Some strange thing is with negative half wave, because it has about 8-12 volt. I can remove R2 and feedback works/excites but then I have distorted wave in all circuit.
You need better AD790 input protection.  I'll again suggest antiparallel signal diodes instead of zener diodes.  AD790 is specified for input voltage from -Vs to +Vs-2V, so -5V to +3V in your case.  You are lucky to have not burnt your AD790 already!  Diode structures within the AD790 are clamping negative voltage to -5V.
David Knierim

Offline Chrisader

  • High Voltage Enthusiast
  • *
  • Posts: 11
  • Karma: +0/-0
    • View Profile
Re: Proper phase lead adjustment at the input
« Reply #6 on: September 04, 2021, 08:27:51 PM »
Hi!

A little time passed but I found problem why I cannot did phase lead properly.
In low power arrangement where current transformer has one section there is necessary to add in serial to CT resistor. Without it phase lead is impossible. I did just like below:



Resistor has to be about 1K and 5W or  little more to keep safe level from overheat. I used 100uH choke as coil and 200R variable resistor. I got zero cross but strange thing happened.
First half wave seems to be fine, but in every next cycle phase lead get disappear and time delay is getting greater. I suppose that is caused by choke which I used and it’s not intended to this type operation. I added 1uF capacitor which makes work of AD790 more stable. At first cycle it likes to oscillate regardless from voltage offset at second input...
 
To set ZCS I just using variable resistor. Coil has 100uH and looks exactly like this:



I suppose I need something like this:



About FET’s. I used them but system will be feeded up to 24V just like ZVS induction heaters. It’s just did for educational purposes only and in further I want to use IGBT’s.

5V zener clamping diodes works fine just like signal diodes. I had never lost any AD790 at this moment:) Only keep in mind to not overcurrent them...

Scope screens:





At this moment I need to find why I got this strange delay...
I suppose wrong type of burden coil...






Offline davekni

  • High Voltage Expert
  • ******
  • Posts: 1296
  • Karma: +62/-0
  • Physicist, engineer (electronic), and hobbiest
    • View Profile
Re: Proper phase lead adjustment at the input
« Reply #7 on: September 05, 2021, 12:39:34 AM »
You are getting farther away from a workable circuit.  I encourage you to try circuit simulation for your experimenting.
David Knierim

Offline Chrisader

  • High Voltage Enthusiast
  • *
  • Posts: 11
  • Karma: +0/-0
    • View Profile
Re: Proper phase lead adjustment at the input
« Reply #8 on: September 20, 2021, 10:40:33 AM »
Hi!
I finally did it and find all mistakes as I think...







Schematic now looks that:



I need to put one more resistor between R24 and C38. It separates from distortions or trapezium waveform, caused by zener diodes. Phase lead I can smoothly regulate by R11 and it’s enough in my case. Values of all components are very high in my case but I wanted run it first time on 12V power supply to not destroy anything. It was done only for educational purposes, before I make power supply from grid and run my Tesla coil. I want to do phase lead tuning on input stage unlike Predikter where constant phase lead need to be delayed.
Thanks for attention:)

High Voltage Forum

Re: Proper phase lead adjustment at the input
« Reply #8 on: September 20, 2021, 10:40:33 AM »

 


* Recent Topics and Posts

post Re: Measuring TC secondary parameters
[General Chat]
davekni
Today at 09:51:49 PM
post Re: Tesla plate
[General Chat]
Duane B
Today at 09:18:43 PM
post Re: Measuring TC secondary parameters
[General Chat]
Duane B
Today at 09:16:36 PM
post Re: Idea for QCW DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
Today at 08:53:36 PM
post Re: Idea for QCW DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Uspring
Today at 06:49:45 PM
post Re: Idea for QCW DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
Today at 05:34:54 AM
post prodc9 interrupter plexi case
[Dual Resonant Solid State Tesla coils (DRSSTC)]
costas_p
September 24, 2021, 07:14:53 PM
post Re: Idea for QCW DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Uspring
September 24, 2021, 05:37:56 PM
post Re: Tesla plate
[General Chat]
Benbmw
September 24, 2021, 05:20:01 PM
post Re: Idea for QCW DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Netzpfuscher
September 24, 2021, 07:26:30 AM
post Tesla plate
[General Chat]
Duane B
September 24, 2021, 01:54:49 AM
post Re: Idea for QCW DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 23, 2021, 09:08:23 PM
post Re: Idea for QCW DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Uspring
September 23, 2021, 07:09:23 PM
post Re: Idea for QCW DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 23, 2021, 05:46:38 PM
post Re: Help with SSTC feedback circuits with 4046 IC and Schmitt Triggers
[Solid State Tesla Coils (SSTC)]
AFreshLad
September 23, 2021, 11:12:25 AM
post Re: Idea for QCW DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Netzpfuscher
September 23, 2021, 08:49:51 AM
post Re: Idea for QCW DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 23, 2021, 06:15:34 AM
post Re: Help with SSTC feedback circuits with 4046 IC and Schmitt Triggers
[Solid State Tesla Coils (SSTC)]
davekni
September 23, 2021, 05:48:40 AM
post Re: Idea for QCW DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Uspring
September 22, 2021, 07:21:12 PM
post Re: Help with SSTC feedback circuits with 4046 IC and Schmitt Triggers
[Solid State Tesla Coils (SSTC)]
AFreshLad
September 22, 2021, 10:41:20 AM
post Re: Help with SSTC feedback circuits with 4046 IC and Schmitt Triggers
[Solid State Tesla Coils (SSTC)]
davekni
September 22, 2021, 07:07:48 AM
post Re: Help with SSTC feedback circuits with 4046 IC and Schmitt Triggers
[Solid State Tesla Coils (SSTC)]
AFreshLad
September 22, 2021, 05:26:40 AM
post Re: Measuring TC secondary parameters
[General Chat]
MRMILSTAR
September 22, 2021, 05:14:30 AM
post Measuring TC secondary parameters
[General Chat]
Duane B
September 22, 2021, 05:01:22 AM
post Re: Load resistance and discharge terminal for my 14-stage CW multiplier
[Voltage Multipliers]
MRMILSTAR
September 21, 2021, 10:33:34 PM
post Re: Futurist's DRSSTC 0.5
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Mads Barnkob
September 21, 2021, 07:21:50 PM
post Re: Load resistance and discharge terminal for my 14-stage CW multiplier
[Voltage Multipliers]
klugesmith
September 21, 2021, 05:24:35 PM
post Re: BrOdin coil - High power Big sparks!
[Dual Resonant Solid State Tesla coils (DRSSTC)]
futurist
September 21, 2021, 04:18:47 PM
post Load resistance and discharge terminal for my 14-stage CW multiplier
[Voltage Multipliers]
MRMILSTAR
September 21, 2021, 06:30:44 AM
post Re: Help with SSTC feedback circuits with 4046 IC and Schmitt Triggers
[Solid State Tesla Coils (SSTC)]
davekni
September 21, 2021, 06:04:52 AM
post Re: BrOdin coil - High power Big sparks!
[Dual Resonant Solid State Tesla coils (DRSSTC)]
fh89
September 21, 2021, 03:56:28 AM
post Re: Idea for QCW DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
johnnyzoo
September 20, 2021, 09:24:10 PM
post Re: BrOdin coil - High power Big sparks!
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Uspring
September 20, 2021, 05:07:43 PM
post Re: Large Rectifier Tube
[General Chat]
Benbmw
September 20, 2021, 04:12:27 PM
post Re: Proper phase lead adjustment at the input
[Dual Resonant Solid State Tesla coils (DRSSTC)]
Chrisader
September 20, 2021, 10:40:33 AM
post Re: Help with SSTC feedback circuits with 4046 IC and Schmitt Triggers
[Solid State Tesla Coils (SSTC)]
AFreshLad
September 20, 2021, 10:39:59 AM
post Re: BrOdin coil - High power Big sparks!
[Dual Resonant Solid State Tesla coils (DRSSTC)]
futurist
September 20, 2021, 10:38:31 AM
post Re: Large Rectifier Tube
[General Chat]
klugesmith
September 20, 2021, 07:07:55 AM
post Re: Help with SSTC feedback circuits with 4046 IC and Schmitt Triggers
[Solid State Tesla Coils (SSTC)]
AFreshLad
September 20, 2021, 06:03:57 AM
post Re: Help with SSTC feedback circuits with 4046 IC and Schmitt Triggers
[Solid State Tesla Coils (SSTC)]
davekni
September 20, 2021, 05:31:57 AM
post Re: Help with SSTC feedback circuits with 4046 IC and Schmitt Triggers
[Solid State Tesla Coils (SSTC)]
AFreshLad
September 20, 2021, 04:59:17 AM
post Re: Help with SSTC feedback circuits with 4046 IC and Schmitt Triggers
[Solid State Tesla Coils (SSTC)]
AFreshLad
September 20, 2021, 04:28:03 AM
post Re: Idea for QCW DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 20, 2021, 12:53:39 AM
post Re: Idea for QCW DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
johnnyzoo
September 19, 2021, 11:09:37 PM
post Re: Idea for QCW DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
AstRii
September 19, 2021, 09:47:14 PM
post Re: Idea for QCW DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
davekni
September 19, 2021, 08:28:05 PM
post Re: Help with SSTC feedback circuits with 4046 IC and Schmitt Triggers
[Solid State Tesla Coils (SSTC)]
davekni
September 19, 2021, 08:19:01 PM
post Idea for QCW DRSSTC
[Dual Resonant Solid State Tesla coils (DRSSTC)]
AstRii
September 19, 2021, 07:58:09 PM
post Re: Help with SSTC feedback circuits with 4046 IC and Schmitt Triggers
[Solid State Tesla Coils (SSTC)]
AFreshLad
September 19, 2021, 03:16:40 PM
post Re: Driverless MOSFET SSTC
[Beginners]
AstRii
September 19, 2021, 01:32:47 PM

Sitemap 1 2 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 
SimplePortal 2.3.6 © 2008-2014, SimplePortal