Induction Cooking Table : IGBT keeping to short !

[Phil Allison]

<[email protected]

"An intermittent fault ( ie internal short) in the load. "

Of course it would. But not always the same transistor.

( snip loads of puerile shit )


** Bollocks.

You have thoroughly worn out you welcome here - pal.

Fuckwit GG posters are tolerable in small doses only.

Sooooo - YOU are now fair game.

Watch out.



..... Phil

 

[William Sommerwerck]

You have thoroughly worn out you welcome here - pal.
Fuckwit GG posters are tolerable in small doses only.
Sooooo - YOU are now fair game.
Watch out.

For what? You're going to throw a tub of spicy hummis at him?

 

[Phil Allison]

"William Sommerwerck"

You have thoroughly worn out you welcome here - pal.
Fuckwit GG posters are tolerable in small doses only.
Sooooo - YOU are now fair game.
Watch out.

For what? You're going to throw a tub of spicy hummis at him?

** That joke is tad obscure.

My remark is fair warning that the GG poster has crossed the line and needs
to pull his head in.

Or I will bite it off .....



.... Phil

 

[William Sommerwerck]

You have thoroughly worn out you welcome here - pal.

Fuckwit GG posters are tolerable in small doses only.
Sooooo - YOU are now fair game. Watch out.

For what? You're going to throw a tub of spicy hummis at him?

** That joke is tad obscure.

It is obscure. It's a line from a movie I haven't seen, "Paranorman".

My remark is fair warning that the GG poster has crossed the line
and needs to pull his head in. Or I will bite it off ...

If you mean figuratively... You won't be doing much actual damage.

 

[[email protected]]

All Phil will ever do, is bore people to tears with his juvenile whining.

Internet Turet's syndrome maybe. But I am not going to dwell on that. The only real power Phil has is that every once in a while he has something to say so people don't completely ignore him YET. Whatever.

"Wouldn't increasing the capacitance from 0.68uF to 1uF result in a 50%

increase in cooking energy? "

No, that is a common misconception in the business. Do you remember the formula for capacitive reactance ? Apply it and just guess the frequency is over an octave above the sonic range.

The fact is that those caps are not being used as reactive components like in a tuned system, they are being used as coupling caps.

In that circuit they are effectively in parallel. Ground and the power supply rail are effectively at the same AC potential, so it's not 0.68, it's aready 1.36 uF. That is almost a piece of wire at 20 Khz. Almost, but we are dealing with a quite higher frequency here.

I got some pretty beefy 1 uF/400 V here but they are old and I can't be sure they can really handle the current. Actually I would like to use them in speaker crossovers someday.

Anyway, people also make this mistake working on SMPSes. Some use a coupling cap to keep DC off the transformer, OK, but that is a coupler. Think about it a sec., it LOWERS frequency to produce more output. Letting the capacitive reactance curve into that would fight against what you want to accomplish.

"Now i'm stuck at finding the polypropylene caps. "

Digikey doesn't have anything ?

What are you dealing with here, rectified 240 volts ? At 320, 17 amps is 5,440 watts. Four burners would add up to 68 amps. What size breaker does this thing take ?

It is possible that they are special caps of course. Manufacturers love using special parts.

Either way, lots of that type of cap fail by exhibiting leakage under stress. Same with the snubbers.

When you get a circuit like that, treat is like a bridge rectifier. Balance.. You change all four diodes in a bridge right ? You change both outputs inan audio amp right ? Same deal here.

Also everybody think about why SMPSes with the "totem pole" configuration usually use full wave rectification on the secondaries. Because balance is as important as it is in an audio amp. It may not seem so, but it is. If notthey would save money on diodes.

 

[Franc Zabkar]

increase in cooking energy? "

No, that is a common misconception in the business. Do you remember the formula for capacitive reactance ? Apply it and just guess the frequency is over an octave above the sonic range.

The fact is that those caps are not being used as reactive components like in a tuned system, they are being used as coupling caps.

In that circuit they are effectively in parallel. Ground and the power supply rail are effectively at the same AC potential, so it's not 0.68, it's aready 1.36 uF. That is almost a piece of wire at 20 Khz. Almost, but we are dealing with a quite higher frequency here.

Here is an induction cooker design where the capacitors and coil are
said to form a "resonant tank":
http://beaversource.oregonstate.edu/projects/44x201103/wiki/IGBT+Coil

The second IGBT's body diode allows the coil current to decay
gracefully when the first IGBT switches off. I forgot about that.

- Franc Zabkar

 

[Franc Zabkar]

Here is an induction cooker design where the capacitors and coil are
said to form a "resonant tank":

http://beaversource.oregonstate.edu/projects/44x201103/wiki/IGBT+Coil

The above design references the following document:

Induction Heating System Topology Review:
http://www.fairchildsemi.com/an/AN/AN-9012.pdf

- Franc Zabkar

 

[[email protected]]

"The second IGBT's body diode allows the coil current to decay
gracefully when the first IGBT switches off."

No. Don't you understand that ground is a human thing, not an electronic thing ? There is no up and down.

The only possible difference is if one side of the coil has more capacitance to ground, which would make an inbalance. This COULD happen, but if it did, the bottom Xsistr failing would be such a common failure mode it would be on Fox fucking news. Switching the leads to the inductor could prove it, because then the top Xsistr would fail first.

We are talking about a ground fault condition here, without that, no anomality in the load could be imbalanced after it is running. Failing on startupis a different story, and this ain't it.

Get a grip, or a firm base of theory. Somehting. I'm surprised people can tie their shoes,,,, oh wait, they can't.

So much for that.

 

[cLx]

Wouldn't increasing the capacitance from 0.68uF to 1uF result in a 50%
increase in cooking energy?

Also, wouldn't each capacitor take longer to charge, and if the
capacitor wasn't fully charged when the IGBT switched off, wouldn't
this result in an interruption of the current in the coils, with a
potentially damaging back-EMF? Or am I completely misunderstanding how
this appliance works?

It's a serie LC resonnance driven by a half bridge switcher, if I'm correct.

 

[William Sommerwerck]

I've been following this with curiosity and amusement.

I suspect this product is a marginal design that fails more-often than it
should, but not often enough to be considered outright "defective". (The
NuWave product advertised on late-night TV seems to be plagued with similar
problems.) There is probably no way to "fix" it, short (hmmm) of a complete
bottom-up redesign. It's a shame, because a counter-top induction "burner"
is a good idea. (I almost ordered the NuWave until I learned how unreliable
it -- and its seller -- are.)

The Infinity "SWAMP" switching amplifier from the late '70s is an example of
such a marginal design. It blew its output transistors far too often. The
designer later found the problem, and admitted that adding two cheap diodes
per channel would have prevented it.

 

[Franc Zabkar]

It's a serie LC resonnance driven by a half bridge switcher, if I'm correct.

Yes, I see that now. I should have researched the topic instead of
relying on the OP's rough circuit diagram. The absence of the flyback
diodes from the drawing left me wondering how the coil current would
decay after the IGBTs switched off, so I assumed that the current had
to be zero when this happened.

- Franc Zabkar

 

[Franc Zabkar]

No. Don't you understand that ground is a human thing, not an electronic thing ? There is no up and down.

The only possible difference is if one side of the coil has more capacitance to ground, which would make an inbalance. This COULD happen, but if it did, the bottom Xsistr failing would be such a common failure mode it would be on Fox fucking news. Switching the leads to the inductor could prove it, because then the top Xsistr would fail first.

We are talking about a ground fault condition here, without that, no anomality in the load could be imbalanced after it is running. Failing on startup is a different story, and this ain't it.

Get a grip, or a firm base of theory. Somehting. I'm surprised people can tie their shoes,,,, oh wait, they can't.

So much for that.

I don't understand what triggered your "ground" rant. Of course the
circuit is balanced, and of course there is no connection to terra
firma. Did I suggest anything else?

Are you perhaps misunderstanding what I meant by "second" and "first"?
All I'm saying is that after the first IGBT turns off, the second
IGBT's flyback diode allows the coil current to decay. Then the second
IGBT turns on. After the second IGBT turns off, then the first IGBT's
diode allows the coil current to decay, and so on.

- Franc Zabkar

 

[Franc Zabkar]

No, that is a common misconception in the business. Do you remember the formula for capacitive reactance ? Apply it and just guess the frequency is over an octave above the sonic range.

The fact is that those caps are not being used as reactive components like in a tuned system, they are being used as coupling caps.

XC = 1/wC = 1 /(2 x pi x 20000 x 1.36 E-6) ~ 6 ohms

My research would suggest that a typical inductance value for the coil
would be of the order of 50uH.

XL = wL = 2 x pi x 20000 x 50 E-6 ~ 6 ohms

So XL = XC, ie resonance.

In that circuit they are effectively in parallel. Ground and the power supply rail are effectively at the same AC potential, so it's not 0.68, it's aready 1.36 uF. That is almost a piece of wire at 20 Khz. Almost, but we are dealing with a quite higher frequency here.

If those 0.68uF capacitors were coupling capacitors, then it wouldn't
matter how large they were. In fact the larger, the better. So let's
assume for the sake of analysis that they were infinitely large. This
means that the voltage at their junction would be constant (Vsupply /
2), irrespective of the induction coil current.

So when the upper IGBT is on, the voltage across the coil would be
Vsupply - Vsupply/2 = Vsupply/2. Similarly, when the bottom IGBT is
on, the coil voltage would be -Vsupply/2. This results in a linearly
increasing coil current, first in one direction and then in the
reverse direction, ie a symmetrical triangular current.

Vsupply/2 = L . dI/dt

I would think that a triangular current would be undesirable.

- Franc Zabkar

 

[Franc Zabkar]

Now i'm stuck at finding the polypropylene caps. They are expensives,
and what i found can handle only 17 amps. Seem a bit underrated for such
that case.

Doing measurements is not easy as, without load or transistors, the
drive shuts itselfs (not detecting anything), just trying shorts pulses,
and i would not wasting my replacements parts.

I would measure the voltage at the junction of the two 0.68uF
capacitors, with the coil disconnected, and confirm that it sits at
half the DC supply voltage. This might identify any imbalance in the
capacitor characteristics, ie leakage, reduced capacitance.

- Franc Zabkar

 

[[email protected]]

"I don't understand what triggered your "ground" rant..."

I didn't mean it like it sounded I guess.

Anyway I understand what you say about the Xc, but I am assuming that the frequency is much higher than 20 Khz. I'd bet it's somewhere around 75 Khz which takes the tuned circuit out of the picture. The Xc of the caps would be much lower and the Xl of the coil much higher. If the whole shebang was 12 ohms total, one burner would pull almost 27 amps providing the supply is 320 DC. The numbers fall into line at around 75 Khz or so.

At that point, there is nothing tuned about it. I didn't mean to offend, I just know that so many have a few misconceptions about these things, actually not even these things pty se, SMPSes that use a similar configuration. If you don't have those misconceptions great.

I try to think from the engineer's standpoint. Why would I waste energy on Xc ? There is a reason to do it for example in the horizontal sweep ciscuitof a TV but usually it's all or nothing. Block the DC and pass the AC.

J

 

[Franc Zabkar]

Anyway I understand what you say about the Xc, but I am assuming that the frequency is much higher than 20 Khz. I'd bet it's somewhere around 75 Khz which takes the tuned circuit out of the picture.

That Fairchild design note I alluded to in an earlier post discusses
several topologies (including the OP's), all of them based on
resonance principles. The resonant frequency was chosen as 24kHz while
the operating frequency "of the system is set at 28kHz, which is
higher than the resonance frequency, in order to avoid noise generated
within the audio frequency band".

Wikipedia also suggests that 24kHz is a common design target:
http://en.wikipedia.org/wiki/Induction_cooking

The above article tabulates the skin depth of various materials at
24kHz.

However, Wikipedia also states that "Panasonic Corporation in 2009
developed a consumer induction cooker that uses a higher-frequency
magnetic field, and a different oscillator circuit design, to allow
use with non-ferrous metals."

As for some real examples, here are two service manuals for induction
cookers:
http://www.garland-group.com/docs/uploaded/gar/products/g_ec_sm_inducrtcs_4521635.pdf
http://www.garland-group.com/docs/u...atalog Files/Service manual Fajita Heater.pdf

Both manuals refer to a diagnostic mode whereby the microprocessor is
accessible via an RS232 interface. There are several sample outputs
which indicate that the operating frequency is 20080 Hz.


- Franc Zabkar

 

[Franc Zabkar]

I found this interesting application note:
http://www.st.com/internet/com/TECH...AL_LITERATURE/APPLICATION_NOTE/CD00115561.pdf

"Induction cooking functions based on the principle of the series L-C
resonant circuit, where the inductance L is the cooking element
itself. By changing the switching frequency of the high voltage
half-bridge driver, the alternating current flowing through the
cooking element changes its value. The intensity of the magnetic field
and therefore the heating energy can be controlled this way."

It describes the OP's circuit as a resonant circuit consisting of an
L-C resonant tank. It states that "the IGBTs are driven by high
frequency complementary square waves with 50% duty cycle", and that
the frequency of "the PWM signal applied to the driver input pin ...
varies in a range between 19 kHz and 60 kHz". The two capacitors are
680nF 600V types, and the IGBTs are STGY40NC60VD.

The design has 9 power levels, with a frequency of 60kHz for lowest
power, and 25kHz for highest power. At 60kHz the plate current appears
to be roughly triangular, while at 25kHz it appears sinusoidal.

- Franc Zabkar

 

[Franc Zabkar]

It describes the OP's circuit as a resonant circuit consisting of an
L-C resonant tank. It states that "the IGBTs are driven by high
frequency complementary square waves with 50% duty cycle", and that
the frequency of "the PWM signal applied to the driver input pin ...
varies in a range between 19 kHz and 60 kHz". The two capacitors are
680nF 600V types, and the IGBTs are STGY40NC60VD.

The datasheet for the STGY40NC60VD states that the IGBT is good for
"high frequency operation up to 50kHz". Hmmm ...

http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/DATASHEET/CD00003462.pdf

- Franc Zabkar

 

[[email protected]]

Very interesting Frank. You do good research. However there is no real contradiction. The inductance must be higher than we assumed because if the Xl is only 6 ohms, it would pull a hell of alot more current. It had to be oneor the other.

Even more interesting is how they're making it work with non-ferrous vessels. Technically it should work with anything that conducts electricity, coils in a transformer are not ferrous and you can induce current in them right? It actually does work on non-ferrous I guess, just extremely inefficiently. They aimed to increase the efficiency. It says reduced efficiency and Isaw a chart, but it wasn't boiled down to just HOW inefficient a given unit will be on specific cooking materials - as if the layman would usually even know. It would still be trial and error.

Now if they can make it work on glass cookware, they get my vote for the Nobel prize.


Very interesting, thanks for bringing that in.

 

[josephkk]

Very interesting Frank. You do good research. However there is no real contradiction. The inductance must be higher than we assumed because if the Xl is only 6 ohms, it would pull a hell of alot more current. It had to be one or the other.

Even more interesting is how they're making it work with non-ferrous vessels. Technically it should work with anything that conducts electricity, coils in a transformer are not ferrous and you can induce current in them right ? It actually does work on non-ferrous I guess, just extremely inefficiently. They aimed to increase the efficiency. It says reduced efficiency and I saw a chart, but it wasn't boiled down to just HOW inefficient a given unit will be on specific cooking materials -as if the layman would usually even know. It would still be trial and error.

Now if they can make it work on glass cookware, they get my vote for theNobel prize.


Very interesting, thanks for bringing that in.

Making it work with glass would impress must anybody. However i wonder if
they can be made to work on slightly salty water (saline{?}).

Just a thought.

?-)