OMG Induction Heater / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

OMG Induction Heater

first previous 11 12 13 14 next last

Move Thread

LAN_403

Steve Conner

Thu Mar 22 2012, 01:26PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

The LCLR has the same problem too, due to the impedance inversion.

Current-fed inverters driving parallel resonant tanks (eg: Royer, vacuum tube oscillator) do the opposite: the current draw decreases to a safe level when the workpiece is removed.

But the downside is that the power throughput also decreases when a ferromagnetic workpiece goes above its Curie temperature, and that's just when you need the power to stay high, if you want to melt it. Using more power to start with will fix it.

The circuit I used was just my Mk2 DRSSTC driver with the interrupter wired permanently on. It has a current limiting circuit that took care of the no-workpiece problem.

Kizmo

Thu Mar 22 2012, 07:16PM

Registered Member #599 Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 624

Steve Conner wrote ...

The circuit I used was just my Mk2 DRSSTC driver with the interrupter wired permanently on. It has a current limiting circuit that took care of the no-workpiece problem.

Thats almost same what im using. Normal drsstc driver, feedback from tank circuit current and OCD current transformer is measuring inverter output current.

Im also planning to use slightly modified interrupter for power control.

Gabriel35

Fri Mar 23 2012, 02:32AM

Registered Member #2310 Joined: Wed Aug 19 2009, 08:04PM
Location: Santa Catarina - Brazil
Posts: 169

But what about the L-match Inductor?
Isn't there to avoid high current surges when Load is removed from the workcoil?

Steve Conner

Fri Mar 23 2012, 06:39AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

No, it's not.

It's an impedance transformer that allows the work coil current to be higher than the inverter output current.

Gabriel35

Sat Mar 24 2012, 02:11PM

Registered Member #2310 Joined: Wed Aug 19 2009, 08:04PM
Location: Santa Catarina - Brazil
Posts: 169

Well, but when using a series tank topology, reducing the number os turns on the primary of the toroid, means more current draw on the inverter.
Right?

Kizmo

Sat Mar 24 2012, 06:15PM

Registered Member #599 Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 624

So far so good.

1332612374 599 FT36475 Inverter Voltage Vs Current

Yellow trace is half bridge output voltage and blue trace is inverter output current. Tank current would be sinusoidal and in phase with inverter output voltage.

I did heat 250mm long piece of 35mm dia steel shaft bright yellow at the middle. Power level was around 8kW (400V 20A DC input). Measured inverter output current was 120A rms. These are fairly slow igbts to be used at 44kHz so i had quite a bit of swithing losses and actually popped two igbts in the 8kW++ run (heatsink was too hot to touch.. oops). Sadly camera wasnt running.

Next step is to convert this thing to step down transformer fed series resonant circuit.

EDIT:

Lets add another one

GeordieBoy

Sat Mar 24 2012, 11:55PM

Registered Member #1232 Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881

Steve Conner is right. An LCLR induction heater is just as capable of blowing up with no load as one using a simple series resonant tank if it's not controlled appropriately.

With no load in the series resonant tank the Xl and Xc reactances of the work coil and tank cap are equal and opposite at resonance. They cancel-out leaving very little series resistance in a low-loss work coil arrangement with no workpiece. There's no "loss" in the system to limit the current rise due to high Q-factor. So if you don't back off the drive the inverter current will quickly go sky high, along with the tank current and tank component voltages and something has to give.

With the LCLR arrangement it's not as easy to see what happens with no workpiece present. Again there is very little loss in an efficiently designed tank system and the Q factor goes high. At a particular frequency the Xl and Xc of the work coil and tank cap don't quite cancel out, and leave just enough net capacitive reactance to perfectly cancel out the inductive reactance of the matching inductor. The end result is a very very small resistive load and large current rise again if you linger at this spot frequency for too long without backing off the drive.

Of course in practice, you would detect either the inverter output current rising, or the tank voltage/current getting too high and detune the drive frequency, reduce the inverter output voltage, or take some other measure to limit the ring-up of energy in the tank circuit before something blows up.

So in reality the series resonant tank and the LCLR arrangement don't behave that different. The LCLR just has a lot of practical implementation advantages like keeping the main induction heating current out of the inverter and confined to a small loop local to the heat station. It is also fault tolerant to a s/c of the work-coil or tank capacitor, both resulting in the inveter being presented with a light inductive load.

One good way to get your head round the LCLR tank arrangment is to split the physical tank capacitance into two halves C1 & C2. Then think of the LCLR arrangment as a series-resonant tank circuit (consisting of the matching inductor and C1) feeding a parallel-resonant tank circuit (C2 and the work coil) connected across the C1 of the first series resonant circuit. The first series-resonant LC circuit provides the impedance matching between the inverter and the load, and the second parallel-resonant tank circuit effectively provides power-factor-correction for the work-coil by parallel resonating it's inductance with a local "PFC" capacitance, therefore containing the large circulating current in the work-coil.

-Richie,

IamSmooth

Sun Mar 25 2012, 02:51AM

Registered Member #190 Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1567

Kizmo, how are you cooling your igbts?

Conundrum

Sun Mar 25 2012, 06:24AM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4061

Someone please do an "induction soldering iron" build

Hand held, runs off internal Li+ batteries and allows you to heat a bolt, screw etc to red heat from 2" away.

see

-A

Kizmo

Sun Mar 25 2012, 07:23AM

Registered Member #599 Joined: Thu Mar 22 2007, 07:40PM
Location: Northern Finland, Rovaniemi
Posts: 624

IamSmooth wrote ...

Kizmo, how are you cooling your igbts?

130x120x300mm heatsink and 120x120x38 fan

first previous 11 12 13 14 next last

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob