Oven design/baking performance

[mattl]

My hidden oven element failed and it got me thinking about modern oven design.  With cheap microprocessors easily available I'm wondering if there would be any advantage to having a constant oven temp rather than the saw tooth wave we have today.  It would be very easy to calculate the rate of heat loss and either pulse or partially power on the various elements at a rate to overcome the heat loss. Perhaps it's been done and I'm just not aware of it. I'd imagine the cost would be minimal.

 

Does the over temp portion of the cycle contribute to browning? Any benefits to the temperature swing?

 

 

 

[henene4]

There aren't any advantages to the temp swing.

And calculation isn't the issue.

Thing is that you can never just design an element with a specific power that would be a temp hold element.

Thing is that heat loss depends on temp setting and installation.
So it varies widely.

If so you'd need at least 2 of those 2 wattage elements (bake and convection - at least for most modern ovens).

But even then you'd need rapid switching or some way of high power variable voltage conversion.

With switching, you would basically create pulse width modulation for the elements.
That would require something that could switch extremely high currents pretty fast.
We are talking 3kW or more in the sub second rate.
Physical relays would break very quickly due to arcing and solid state components that could handle such a load are pretty inefficient and expensive.

Or you'd need a variable transformer that could adjust the voltage very precisely.
That would be huge and again add losses.

While possible, it won't do much.

While there are certainly ovens that sway to much, most modern ovens are pretty good at keeping the swings in check.

And a +- 10° swing won't harm most dishes much.
At 375, a 10 swing is like 4% swing.

That just doesn't change results much, certainly not enough to justify the losses and the work needed for that little improvement.

 

[kb0nes]

Even with a microprocessor there will always be some set point hysteresis. Of course with a microprocessor you could have more adaptive algorithm if they bothered to deal with overshoot and tailoring element wattage to cavity heat loss.

There would be no need for different elements as power to the element could be controlled to an almost infinite degree with a simple Triac. The electronics would be quite cheap and could easily be bulletproof reliable too, this isn't the 60's anymore. Since the processor already exists, the electronic control would likely be cheaper to implement than using a mechanical thermostat too.

But the real question is does it matter? Throttling the elements would result in lower element temperatures so less infrared energy which might hamper browning. This is likely a case of the perfect being the enemy of the just fine.

 

[henene4]

A normal triac can switch normal loads quite fine, yes.

Even switching loads for motors is doable by now with triacs.

Thing is most motors have a load of 500W? Maybe a bit more?

A bake element has 6-7 times the wattage of a motor.
And even for motors, heat sinking is already an issue to consider.

With the significantly higher load and the significantly higher temperatures these electronics have to work in, that will be a significant problem.

Not saying it is impossible.

But a lot of development for little reward.

I also don't think infrared radiation is a big concern.

US ovens - as far as I understand - only run the baking element for baking.
That is below the food and more often then not covered.

And further, convection browns just as well - maybe even better degree for degree - without direct infrared radiation from the element.

 

[norgeway]

A perfectly constant

Oven temperature was achieved many decades ago, All older pre 1960s gas ovens used a modulating flame oven control, these kept a perfect temperature, Todays junk will not, and can not keep as even a temperature.

 

[henene4]

Really?

Gas ovens are extremely rare over here.
I heard of the modulating gas dryer burner.

Any model examples?

Would love to read up a little...

 

[kb0nes]

Henrik, respectfully I think your experience with modern solid state high power electronics is behind the times.

Back in the 80's I was working on vapor-phase surface mount soldering machines and the solid state relay modules had no durability issues switching 40Kw resistive heating arrays. Today things are vastly more durable and much less expensive. We are only talking about controlling 10 amps here...

A Triac switches quite quickly and while either on or off there really isn't a lot of resistive losses so not much heat. Design it with modules with zero crossing switching and than there is even less heating loss.

As for the development, it is dead simple technology that is in common use today. Yes some software would have to be coded up but that is just software, write it once and it can be used in scads of units for no cost.

Most of the ovens I am familiar with feed some amount power to the broil element to enhance browning. The ability to control the element balance more or less infinitely with a advanced controller could actually make the idea more appealing to tailor oven performance!

I don't disagree that smoothing out the temperature variation may offer very little reward. Hence my comment on the perfect being the enemy of the very good. But it could be easily done

 

[henene4]

I think you are underestimating the load involved

Looking around on the internet, you find things like these:

https://www.electrodragon.com/product/4000w-high-power-scr-electronic-volt-regulator/

They are ment for resistive loads, so spot on!
Maybe not quite the type of equipment that would be used, but hey.

The product page lists an efficency of 90%.

That is a heatoutput of over 300W for just the switching.
Means for example for 3500W max heating power you'd have to go to about 3900W supply power.

The rest? Just blown into the kitchen.

Even going by the way more optimistic figures in the data sheet related to that item, we would still have up to about 20W of heat to dissipate.

Sounds awesome, right?

That is still the power of a modern laptop CPU.
Looking at computers like Intel's NUCs, you know what it takes to cool these with ambient air and the limited space.

Then you add the fact that you already have an insane heatsource just below your electronics, and it won't go well given your Tmax is 125C for basically all triacs.

So on the only way out is to design some really good cooling solution.

Which again means high development cost and high production cost while still having medium return.

There is a reason that there has been a number of instances where peoples ovens just died on self clean.

Electronics hate heat.

And one look at basicly any high power switching application shows the heat involved.

Most inverter boards for washer motors (I know these use most likely MOSFETs, but that really dosen't matter much in regards to their efficency profile) have heatsinks on them.
Yes, most of them are cooled passivley, but they only have to operate in ambient temperature.

Now again multiply the required output by 6 and raise the temperature around the device by 60C or so while maintaining a slim construction and you either have to compete with one hell of a cooling solution or just don't do it at all.

 

[eronie]

Reply #7

I hate a no it all that doesn't know what they are talking about, nuff said.

 

[kb0nes]

Google searches are a poor replacement for actual electronics knowledge.

Henrik I try to not be an ageist, but I graduated from electronics school 12 years before you were born. I have continued to work with electronics every day since and strive to stay current. Controlling an oven element would be easy with a few dollars worth of parts. If you switch at zero crossing the solid state relay efficiency is over 99%. Again it would be easy, cheap and reliable seeing as the microprocessor is already there.

Still not sure it is worth the effort, but it would be a fun experiment. It would be neat to actually experiment and learn, vs guessing and Googling...