Welcome
Username:

Password:


Remember me

[ ]
[ ]
Online
Guests: 11, Members: 0 ...

most ever online: 333
(Members: 3, Guests: 330) on 06 Jun : 15:15

Members: 2586
Newest member: Ray1214
Members Birthdays:
All today's birthdays', congrats!
Desmogod (38)
Alex Smith (21)


Next birthdays
04/25 Desmogod (38)
04/25 Alex Smith (21)
04/26 Bead (31)
Support 4hv.org!
Donate:
4hv.org is hosted on a dedicated server. Unfortunately, this server costs and we rely on the help of site members to keep 4hv.org running. Please consider donating. We will place your name on the thanks list and you'll be helping to keep 4hv.org alive and free for everyone. Members whose names appear in red bold have donated within the last three months. Green bold denotes those who have donated to keep the server carbon neutral.


Special Thanks To:
  • Aaron Holmes
  • Aaron Wheeler
  • Adam Horden
  • Alan Scrimgeour
  • Andre
  • asabase
  • Austin Weil
  • barney
  • Barry
  • Bert Hickman
  • Bill Kukowski
  • Brandon Paradelas
  • Bruce Bowling
  • BubeeMike
  • Byong Park
  • Cesiumsponge
  • Chris F.
  • Chris Hooper
  • Corey Worthington
  • Derek Woodroffe
  • Dalus
  • Dan Strother
  • Daniel Uhrenholt
  • datasheetarchive
  • Dave Billington
  • Dave Marshall
  • David F.
  • Dennis Rogers
  • drelectrix
  • Dr. John Gudenas
  • Dr. Spark
  • eastvoltresearch
  • Eirik Taylor
  • Erik Dyakov
  • Erlend^SE
  • Finn Hammer
  • Firebug24k
  • GalliumMan
  • Gary Peterson
  • George Slade
  • GhostNull
  • Grant
  • GreySoul
  • Henry H
  • IamSmooth
  • In memory of Leo Powning
  • Jacob Cash
  • James Howells
  • James Pawson
  • Jeff Greenfield
  • Jesse Frost
  • Jim Mitchell
  • jlr134
  • Joe Mastroianni
  • John Forcina
  • John Oberg
  • John Willcutt
  • Jon Newcomb
  • klugesmith
  • Leslie Wright
  • Lutz Hoffman
  • Mads Barnkob
  • Martin King
  • Mats Karlsson
  • Matt Gibson
  • Matthew Guidry
  • Michael D'Angelo
  • Mikkel
  • mileswaldron
  • mister_rf
  • Neil Foster
  • Nick de Smith
  • Nick Soroka
  • Nik
  • Norman Stanley
  • Patrick Coleman
  • Paul Brodie
  • Paul Jordan
  • Paul Montgomery
  • Ped
  • Peter Krogen
  • Peter Terren
  • PhilGood
  • Richard Feldman
  • Robert Bush
  • Royce Bailey
  • Scott Fusare
  • Scott Newman
  • Stella
  • Steven Busic
  • Steve Conner
  • Steve Jones
  • Steve Ward
  • Sulaiman
  • Thomas Coyle
  • Thomas A. Wallace
  • Timo
  • Torch
  • Ulf Jonsson
  • Vaxian
  • Weston
  • William Kim
  • William N.
  • William Stehl
  • Wesley Venis
The aforementioned have contributed financially to the continuing triumph of 4hv.org. They are deserving of my most heartfelt thanks.
Forums
4hv.org :: Forums :: General Science and Electronics
<< Previous thread | Next thread >>   

OMG Induction Heater

Go to page  1 2 3 4 [5] 6
Author Post
Steve Conner
Sun Mar 25 2012, 07:25AM
Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6455
Kizmo, from your waveforms you look a bit out of tune. The inverter current is supposed to be in phase with the voltage so you can have ZCS. That's more important than ZVS, especially when pushing the limits of slow IGBTs.
Back to top
Kizmo
Sun Mar 25 2012, 07:37AM
Registered Member #599
Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 510
How can it be out of tune with self oscillating driver? :)

This is just drsstc driver and im using tank current to generate feedback signal which is in phase with tank current.

EDIT:

I was told earlier that poor inverter power factor is known "feature" of simple LCLR heaters and there is not much one can do about it.
Back to top
Steve Conner
Sun Mar 25 2012, 08:38AM
Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6455
Well, surely you should be using the inverter output current as the feedback source, not the tank current.

With the series resonant heater, the two currents are in phase because they're the primary and secondary currents of the transformer. But in the LCLR they're out of phase, because a phase shift is necessary to transfer real power through an inductor.
Back to top
Kizmo
Sun Mar 25 2012, 08:42AM
Registered Member #599
Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 510
I tried taking feedback from inverter current and guess what happened..

I found out that the resonant frequency of my matching inductor and DC block capacitor is about 6kHz. Couldnt get it working at all.
Back to top
Steve Conner
Sun Mar 25 2012, 11:58AM
Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6455
Oh, I forgot you don't have a proper driver with a PLL.

A simple feedback circuit will oscillate at whichever resonant frequency has the highest loop gain. So, you could include a highpass filter in the feedback circuit to lower the loop gain at 6kHz, and make it prefer the frequency you want.

If you can't easily design a filter that blocks 6kHz without unwanted phase shift at the operating frequency, then design the best filter you can and make the DC block cap bigger to lower the frequency of the unwanted resonance. Although, the phase shift at the operating frequency might even be beneficial as it would work like a "Prediktor".
Back to top
Kizmo
Sun Mar 25 2012, 12:06PM
Registered Member #599
Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 510
I think im done with this topology. Already got 2 kilos of power ferrite and some copper tubing for series resonant transformer fed tank circuit :)
Back to top
Steve Conner
Sun Mar 25 2012, 01:30PM
Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6455
I suppose I have to agree with you, since I've only ever built series resonant heaters!

I chose series resonant because:

The inverter output current is in a fixed ratio to the work coil current, so a single overcurrent detector can measure and limit both. If you put a limit on the PLL tracking range, it can also protect the tank cap against overvoltage. (Tank cap voltage is proportional to current and 1/frequency.)

The feed transformer isn't that much bigger/harder to build than the matching inductor of a LCLR. The only problem is the massive water-cooled secondary winding, but we 4hvers enjoy building things like that

The feed transformer isolates the work circuit from the inverter, which is usually not isolated from the mains.

But Richie says the LCLR is theoretically better. I can't remember why, maybe he can comment?
Back to top
Kizmo
Sun Mar 25 2012, 01:51PM
Registered Member #599
Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 510
The transformer gets rather large when its made from ferrite, drive voltage is high (320V p-p) and highest possible frequency is 32kHz (limited by tank capacitor)
Back to top
Kizmo
Sun Mar 25 2012, 04:47PM
Registered Member #599
Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 510
Almost done. One question: With transformer fed series resonant topology i have to include transformer high current side winding inductance with work coil and wiring inductance when figuring out the resonant frequency?

Wonder how low it will drop :)
Back to top
Steve Conner
Sun Mar 25 2012, 05:47PM
Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6455
Yes, you do.

To avoid excessive inductance, I built my transformer right into the heat station as an extension of the capacitor's water cooling plates. That also got the secondary water cooled as a bonus. And I wound the primary inside the secondary, on the same core limb, again to minimise leakage inductance. (Since the inverter is a voltage source, it's the leakage inductance that's relevant.)

But a little extra inductance may not be a problem, it will reduce the change in operating frequency with workpiece type.
Back to top
Kizmo
Sun Mar 25 2012, 06:00PM
Registered Member #599
Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 510
Then its back to the drawing board. I cant have much more than 10uH of extra inductance :)
Back to top
IamSmooth
Sun Mar 25 2012, 06:06PM
Registered Member #190
Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1435
Kizmo, at your power levels I would suggest liquid cooling. You will loose IGBTs from overheating.
Back to top
Steve Conner
Sun Mar 25 2012, 06:06PM
Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6455
Well, here's mine :)

I sold it, as far as I know it's now being used for a student project somewhere.







Back to top
Kizmo
Sun Mar 25 2012, 06:27PM
Registered Member #599
Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 510
IamSmooth wrote ...

Kizmo, at your power levels I would suggest liquid cooling. You will loose IGBTs from overheating.

Yeps that is of course another option but i would like to fix the main cause of heating which is switching losses during turn off. And that is of course caused by poor power factor of this type of feedback method.
Back to top
Kizmo
Sun Mar 25 2012, 06:56PM
Registered Member #599
Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 510
This whole transformer idea is starting to feel stupid. I cant have more than 5uH of inductance from transformer or this thing will turn into ear wrecking machine. And there is no way i can fit enough windigs on any of my smaller ferrites :(
Back to top
Steve Conner
Sun Mar 25 2012, 07:12PM
Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6455
How are you calculating the inductance? It's the leakage inductance, so to a first approximation, it's the inductance of the secondary by itself with the primary and core removed. In practice slightly less, if you wind the primary close to the secondary as I did.

That should be much less than 1uH. To keep a sense of scale, I remember that a 10 inch diameter loop has an inductance of about 600nH.
Back to top
Kizmo
Sun Mar 25 2012, 07:22PM
Registered Member #599
Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 510
Now you made me feel like ultra stupid

I tought its the inductance that one can calculate from nH/turn^2 value...


Leakage inductance as you described it is no problem at all. I was just about to tear whole days worth of work apart and call it epic fail :p
Back to top
IamSmooth
Mon Mar 26 2012, 12:23AM
Registered Member #190
Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1435
Steve, the coil you showed...that looks like a levitation coil. What have you suspended?
Back to top
IamSmooth
Mon Mar 26 2012, 12:25AM
Registered Member #190
Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1435
Kizmo wrote ...

IamSmooth wrote ...

Kizmo, at your power levels I would suggest liquid cooling. You will loose IGBTs from overheating.

Yeps that is of course another option but i would like to fix the main cause of heating which is switching losses during turn off. And that is of course caused by poor power factor of this type of feedback method.


What is your frequency?
Back to top
Kizmo
Mon Mar 26 2012, 05:28AM
Registered Member #599
Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 510
IamSmooth wrote ...

Kizmo wrote ...

IamSmooth wrote ...

Kizmo, at your power levels I would suggest liquid cooling. You will loose IGBTs from overheating.

Yeps that is of course another option but i would like to fix the main cause of heating which is switching losses during turn off. And that is of course caused by poor power factor of this type of feedback method.


What is your frequency?


Limited by capacitor between 16 and 32kHz. This first prototype ran at 44kHz and at this frequecy calculated turn-off losses (=hard switching) are almost 800W/transistor.
Back to top
Steve Conner
Mon Mar 26 2012, 06:59AM
Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6455
Iamsmooth: I've levitated a small piece of aluminium tube, but I had to push the setup to the absolute limit. I turned the current limit up too high, and when the piece melted, it got unstable and fell out of the coil, and my IGBTs went boom.

Levitating a solid sphere of metal would take lots more power.
Back to top
Dago
Mon Mar 26 2012, 10:17AM
Registered Member #538
Joined: Sun Feb 18 2007, 08:33PM
Location: Finland
Posts: 154
I wonder why (all?) commercial heaters use the LCLR-topology, cause the poor power factor does make it sound pretty bad? Can you compensate the poor power factor in any way?
Back to top
Steve Conner
Mon Mar 26 2012, 10:23AM
Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6455
Yes, if you drive it at the correct frequency, you get unity power factor just the same as the series resonant.

The correct frequency is not the resonant frequency of the tank circuit, but slightly above, where the tank circuit appears as a capacitive reactance of the right magnitude to cancel the inductive reactance of the matching inductor.

Ameritherm's small heaters use series resonant. I got the idea from some gut shots that were shared in the 4hv chat years ago.
Back to top
IamSmooth
Mon Mar 26 2012, 01:25PM
Registered Member #190
Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1435
As I mentioned earlier, I used series resonant, got it up to 12kw of power and did all kinds of crazy stuff. At these power levels you need to make sure you stay slightly above as Steve mentioned. If you start get to the true resonance or below the switches will go boom. I put a 60A fast-blow going to the inverter to protect my board from secondary frying if the switches shorted.

I minimized my switching loses with lower frequencies and fast turnon/turnoffs. One issue with faster transitions is the ringing, which has to get snubbed, or else you will have a worse situation. As far as the frequency: while I used lower frequencie for melting bricks of iron, I used frequencies close to 120 kHz, but was still able to get rid of the heat without compromising the mosfets.

You might want to consider using mosfets if you are using frequencies above 20-30khz.
Back to top
Gabriel35
Mon Mar 26 2012, 01:41PM
Registered Member #2310
Joined: Wed Aug 19 2009, 08:04PM
Location: Santa Catarina - Brazil
Posts: 147
The discussion Is getting Interesting!
Unfortunatly I've never succeded with a PLL driven Induction Heater, But I'll not stop trying...

Currently I'm in doubt of what topology to use...
I'll try some of them and see what happens...
Back to top
Kizmo
Mon Mar 26 2012, 03:40PM
Registered Member #599
Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 510
IamSmooth wrote ...

As I mentioned earlier, I used series resonant, got it up to 12kw of power and did all kinds of crazy stuff. At these power levels you need to make sure you stay slightly above as Steve mentioned. If you start get to the true resonance or below the switches will go boom. I put a 60A fast-blow going to the inverter to protect my board from secondary frying if the switches shorted.



DRSSTCs are switched at perfect resonance at current zero crossings and they dont blow up...?

EDIT:

I guess i have to explain my idea of how im going to do this heater.

Series resonant tank circuit. 10µF 1kA celem capacitor, about 3µH work coil + stray inductance. Resonates roughly at 29kHz. Coupling transformer is just large ferrite transformer with 1:10..1:20 ratio. Feedback is taken from tank circuit current with measuring capacitor (150nF cap in parallel with celem and small ferrite CT). Inverter is igbt half bridge powered by rectified 3 phase and im not planning to use any filter capacitance on bridge, just couple µF worth of film caps. Driver is normal DRSSTC driver with interrupter.

Interrupter would be outputting pulses with 0-99% ON time. Idea behind this is to be able to have at least some sort of power control by varying on time. Drsstc has of course zero crossings synchronized over current protection which in this case would be measuring inverter output current. I guess if needed i could add another overcurrent protection circuit that measures tank current via same measuring capacitor. That would serve as tank capacitor over current/voltage protection.

Due to my limited knowledge, i dont see why this kind of contraption wouldnt work. At least it has proven itself in drsstc world. If you see potential problems, please stop me before i waste my time ;)

Mcu controlled PLL driver is *far* too complicated for my knowledge

Back to top
GeordieBoy
Mon Apr 02 2012, 11:48AM
Registered Member #1232
Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 863
> But Richie says the LCLR is theoretically better. I can't remember why, maybe he can comment?

The work-coil circuit is really just an impedance matching network to convert the unfeasibly low impedance of a lump of solid metal up to a figure that is realistic to present to a solid-state inverter running directly off the mains line. I don't think it really matters how you achieve this impedance transformation as long as it is done efficiently because the currents and voltages at work in IH are large.

I prefer the LCLR network for a number of reasons:

1. It keeps the enormous induction-heating current confined to the work-coil / tank capacitor loop. This loop can be made very small and may be located remotely from the inverter.

2. It does away with the need for a high-power high-frequency transformer requiring careful design to minimise core/copper losses and water cooling to carry away the heat from whatever dissipation remains.

3. It is better to alter the impedance matching by varying the air gap in the matching inductor than to change taps on a transformer. A multi-tap transformer implies poor winding window utilisation for certain tap choices resulting in an overall bigger transformer unless series/parallel winding combinations are used.

4. Stray inductance in the wiring between the inverter and the heat station merely adds to the matching inductance, rather than adding to the work coil inductance and reducing coupling to the workpiece.

5. The LCLR network presents a light inductive load to the inverter in the case of a tank capacitor or work-coil short circuit making the system s/c tolerant to faults at the heat station.

6. If the matching inductor is located close to the inverter it reduces HF switching noise leaving the enclosure improving EMI performance.

7. In high-power systems the tank circuit can be fed from several inverters in parallel through several matching inductors. The "ballasting action" of the matching inductors provides inherent current sharing. It also limits the "shoot-between" currents resulting from imperfect syncronisation of the inverters' switching transitions. (This way solid-state systems in the MW and 100's of kHz range can be realised with a high level of redundancy.)

The downsides of the LCLR matching network are:

1. It is more complex to understand the resonance transformation than impedance transformation with a transformer.

2. Leakage fields from the matching inductor can heat nearby metal enclosure!

3. No galvanic isolation of the work-coil from the mains line.

-Richie,
Back to top
Website
IamSmooth
Mon Apr 02 2012, 09:30PM
Registered Member #190
Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1435
GeordieBoy wrote ...


I prefer the LCLR network for a number of reasons:

-Richie,


Richie, you must have seen my schematic for the work coil. It is a coupling transformer from the switches (10-20:1) around the wire going to a series LC tank. If I were to improve the tank circuit, what would I need to do? A different topology or is this one reasonable? I was able to abuse it without failure, but I'm always interested in other and better ways of doing things.
Back to top
GeordieBoy
Tue Apr 03 2012, 11:50AM
Registered Member #1232
Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 863
hey, if it aint broke then don't fix it! If it meets your requirements and even survived abuse without catching fire then it doesn't sound like you have anything defficient to improve.

-Richie,
Back to top
Website
Kizmo
Sun Apr 08 2012, 07:10AM
Registered Member #599
Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 510
Series resonant version has been great success.






Some youtube clips

Boiling water
Melting car motor parts
Aluminum heating fail :)

Back to top
Go to page  1 2 3 4 [5] 6

Moderators: Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Bjørn, Dave Billington, Steve Conner, Wolfram, Mads Barnkob

Jump:     Back to top

Powered by e107 Forum System
 
Legal Information
This site is powered by e107, which is released under the GNU GPL License. All work on this site, except where otherwise noted, is licensed under a Creative Commons Attribution-ShareAlike 2.5 License. By submitting any information to this site, you agree that anything submitted will be so licensed. Please read our Disclaimer and Policies page for information on your rights and responsibilities regarding this site.