Air to air split heat pumps for UK spring and autumn heating

[markocosic]

Hitting max frequency perhaps?

https://efiling.energy.ca.gov/GetDocument.aspx?tn=234914&DocumentContentId=67776

 

Doesn't explain why the PWM is so effing low frequency though. Boiling water is fine. The buggerance occurs when trying to do things with milk etc that respond differently to running 100% for 1 sec then nothing for 2 secs. Heavy pans would fix is but SWMBOs don't like them.

[Radian]

1 hour ago, markocosic said:

 

There's something more subtle going on.

 

For the same make/model of hob and the same pot the behaviour differs between rings of differing dimensions.

 

At high power all the rings PWM "quickly" (no discernible change in heat input to pot)

 

At lower power (say <50%) some rings switch to a low frequency PWM (of the order 3 second cycles) whereas others continue to do high frequency PWM all the way to the minimum setting.

 

This image from a Quora post seems to describe the behaviour. 

 

 

https://www.avdweb.nl/Article_files/Tech-tips/Panasonic-HV-inverter/Technical-Guide-Microwave-Ovens-with-Inverters.pdf

 

The change in frequency must be above resonance rather than below as the inductor's impedance = 2πƒL  otherwise the impedance would drop and the whole thing would draw more power at the same time as reducing the cooking power.

 

2 hours ago, markocosic said:

What's the technical limiting factor that forces a ring to switch from "FM" to "PWM" though?

 

The switching losses in the IGBT switches increase with increases in drive signal frequency. This means there's a sweet-spot for the operating frequency. I'm a little surprised they even bother with FM - it can't maintain optimum power transfer efficiency over a very wide range.

 

I guess the problem with low frequency PWM is the effect of visible flicker from lighting circuits as the 10's of Amps being gated might produce a significant volt-drop. So it's a trade off between efficiency and controllability. Lower power induction loops might be resorting to low frequency PWM if their switching currents are low enough.

 

Mark-spacing the HF drive waveform with PWM can maintain resonance but still entails an efficiency hit as the switching rise-times become a greater proportion of the dwell time. I guess I'm so used to designing just for maximum power transfer that I'm not that familiar with techniques used to modulate it for cooking.

 

[markocosic]

2 hours ago, Radian said:

 

The switching losses in the IGBT switches increase with increases in drive signal frequency. This means there's a sweet-spot for the operating frequency. I'm a little surprised they even bother with FM - it can't maintain optimum power transfer efficiency over a very wide range.

 

Spendier IGBTs = faster IGBTs = more modulation?

 

Efficiency of induction hobs is in 70-80% range AFAIK. Quite a bit pishes off out the bottom.

 

 

2 hours ago, Radian said:

 

I guess the problem with low frequency PWM is the effect of visible flicker from lighting circuits as the 10's of Amps being gated might produce a significant volt-drop. So it's a trade off between efficiency and controllability. Lower power induction loops might be resorting to low frequency PWM if their switching currents are low enough.

 

Limited by regs?

 

https://www.nutwooduk.co.uk/archive/old_archive/990619.html

 

The slower you go the bigger the flicker you're allowed. 20 changes a minute for 1% voltage change?

 

Does switching a couple 00 watts really have such an impact?

 

 

If only we could tell which hobs allowed you to warm milk without incinerating it before buying them...

[markocosic]

Perhaps limits on high frequency field strength force the switch? Can't crank up the frequency because fail field strength >90 kHz?

 

https://www.aceee.org/files/proceedings/2014/data/papers/9-702.pdf

http://www.emfservices.co.nz/resources/emf-measurement-examples/induction-cooktop

 

 

 

This one seems to say it's flicker EMC vs the milk and supports ~3 sec cycles:

 

https://www.infineon.com/dgdl/Infineon-AN2014_01_Reverse_Conducting_IGBT-ApplicationNotes-v03_01-EN.pdf?fileId=db3a30434441da19014445a1e7560135

 

"The minimum operation frequency of the burst mode has to be set high enough to avoid any non-uniformities of heat distribution in the vessel, and at the same time to avoid noticeable fluctuactions of the temperature in the food, which would end in unefficient cooking processes. On the other hand, the maximum operating frequency is usually limited by the EMC requirements. Usually, a burst frequency of 0.2-0.3 Hz is used."

 

Clashing frequencies explain some of the squeaky harmonics with multiple pots on the go too. Deliciously complicated things are these!

 

 

[SteamyTea]

9 minutes ago, markocosic said:

unefficient cooking processes

I have worked with chefs that can't spell, actually I am one of them.

[Radian]

37 minutes ago, markocosic said:

Clashing frequencies explain some of the squeaky harmonics with multiple pots on the go too. Deliciously complicated things are these!

 

Well at least you know to look for a hob that uses a half-bridge series-resonant converter rather than a single ended one. Had fun reading the datasheets thanks!