I hate to be a total dumbarse but how do I access that my friend? Thanks for the reply

http://ww1.microchip.com/downloads/en/AppNotes/00958A.pdf

 

There is a method outlined in the picmicro AN958a based on burst firing a Triac.
Max.

Max thanks so much ... crazy question for you ... does anyone produce a finished unit like is described in the PDF? My limitations are great LOL. I am quite inspired by the Einstein quote in your sig... very true very true

 

Max thanks so much ... crazy question for you ... does anyone produce a finished unit like is described in the PDF? My limitations are great LOL. I am quite inspired by the Einstein quote in your sig... very true very true

Never come across a commercial version, I did put one together some while ago now, not quite the same but an adaptation that was equivalent.
Max.

 

I was going to write a comment amount the slow response time of big heating elements, which can be both bad and good, but thinking about those subjects got me wondering about your heating elements.

What kind of elements are they? Are you repurposing heating elements from other coffee equipment (meant to be submerged in water) or are these specifically made for heating air? My concern is that elements designed for use in water may not do a good job of transferring heat quickly to the air. Sorry, this is a bit of a tangent away from your original question.

 

I don't know about Max's circuit, but there are oodles of pre-made scr "dimmer" circuits out there. This one claims to be good for 10000W, so maybe a separate control for each element?

https://m.ebay.com/itm/10000W-AC110...?_trksid=p2385738.m4383.l4275.c10&_mwBanner=1

My old Probat had separate gas control valves for two banks of burners - it was a little weird at first, but with practice I felt like it was actually easier to make small, precise adjustments with two separate knobs than with one big one, although it was a little different there - two banks of burners on one knob (macro adjustments) and one bank on the other knob (fine tuning.)

I have no idea what the quality of straight-from-China unbranded eBay items is. I would assume similar products exist from better-known vendors (at much higher prices) but my Google search results are overrun with Amazon and eBay exclusively. You'd have to search shop by shop for high wattage scr dimmers if you want a more trusted source.

 

I don't know about Max's circuit, but there are oodles of pre-made scr "dimmer" circuits out there. .

The one quoted uses burst firing instead of the usual phase angle method, which seems to work best on heating element loads.
Max.

 

The one quoted uses burst firing instead of the usual phase angle method, which seems to work best on heating element loads.
Max.

Awesome, thanks for that! I had forgotten the significance of that terminology until you posted this (at which point I Googled it to refresh my memory.)

I'd agree that burst firing seems preferable in terms of electrical noise generated by the control system. It would be nice if there was a straightforward way to achieve it without requiring a microcontroller and programming (although if the thread starter is comfortable with those things, maybe it's no problem.)

I'm wondering if you could make a good approximation of burst firing control by making a pwm circuit (555. 4017, function generator plus comparator, whatever) with a long period, maybe somewhere between 0.5-2 seconds. If you used this simple pwm output to drive a zero crossing SSR, then your load switching transitions would be at zero crossings for minimum noise, just like with the burst firing. The simple pwm circuit wouldn't be sync'd with the zero crossings, so the relationship between the pwm output and the resulting burst firing duty cycle wouldn't be perfect, but I bet it would be fairly close as long as the pwm period wasn't too short.

Any thoughts on this approach? When I imagine it in my head, it seems legit, but I haven't tried to work out the details yet.

 

With the burst firing method it is necessary to use the zero crossing opto couplers as shown in the Fairchild app.
It could be possible maybe to modify the 555 phase angle switcher in fig 3. possibly with another 555?
Max.

 

@crutschow:
Any idea what happened to your simple PWM circuit over at electro tech?

Google finds a post referring to that circuit pretty easily, but the post says the article is being updated. Meanwhile, the schematic and most of the description are gone. There are a few links on the page, but they all redirect saying I need to log in. I'm not a member over there, I didn't need to be to see your circuit last time I looked.

Anyway, I was thinking of trying to adapt that circuit to this application. Any chance you could share it? Or are you really in the midst of changing it and want to finish before sharing it again?

 

@crutschow:
Any idea what happened to your simple PWM circuit over at electro tech?

Google finds a post referring to that circuit pretty easily, but the post says the article is being updated. Meanwhile, the schematic and most of the description are gone. There are a few links on the page, but they all redirect saying I need to log in. I'm not a member over there, I didn't need to be to see your circuit last time I looked.

Anyway, I was thinking of trying to adapt that circuit to this application. Any chance you could share it? Or are you really in the midst of changing it and want to finish before sharing it again?

@crutschow, sorry to be a pest, but would you mind sharing your simple PWM circuit? I could try to recreate it from what I remember of it, but I have a feeling it wouldn't go well!

 

I'll stop harassing @crutschow now - apparently you have to sign in over at Electro-Tech to see his circuit. That didn't used to be the case, but it seems to be now. Anyway, I created a user name and login just to get it, and now I've had some fun playing around with it.

I started with his basic PWM circuit, then adjusted values to get a period between 1-2 seconds. The PWM output is fed into the input of a simulated zero-crossing SSR (my simulation of zero-crossing is a little cludgy, but seems to get the job done.)

This system allows for a simple analog circuit (no microcontrollers or programming required) and doesn't require any circuit design/assembly on high-voltage, potentially dangerous components. All of the custom circuitry is on the low voltage side of an SSR input, and the actual high-power switching can all be done with readily available SSRs.

In the image above, a zoomed in view of a single burst, you'll see that, because of the zero-crossing SSR, each burst consists of entire cycles starting and ending near zero volts. The good side of this is that it creates minimal electrical noise (as opposed to phase-angle proportional control, which is legendarily noisy.) The bad side of this is that it limits the resolution of the heat control, creating a trade-off between response time and heat resolution.

With 120 zero-crossings per second (assuming 60Hz mains,) the heat can be on or off in ~8.3ms increments. This means that if you choose a 0.5 second period, you'll only be able to adjust the heat in 60 steps (1.67% of total heat per step.) If you increase this all the way to 2 seconds, you get 240 steps (0.42% per step,) but you won't get accurate step changes in heat for up to 2 seconds after making an adjustment. I suspect that these limitations aren't really a problem, but as long as I'm pitching an idea, I might as well acknowledge its potential weaknesses!

The simulation below isn't nearly as complicated as it looks - the giant block in the middle is just my messy attempt at simulating a zero-crossing SSR. In real life, I'd recommend building the modified version of Crutschow's circuit and feeding it into a "real" (commercial) SSR for both safety and simplicity reasons.

This whole idea is based entirely on @crustchow's simple PWM circuit (https://www.electro-tech-online.com...dimming-etc-improves-on-555-pwm-circuits.797/) He should get full credit for any awesomeness it provides, and I take full responsibility for anything I messed up when modifying it!

For what it's worth, this is conceptually just trying to do the exact same thing as the circuit Max suggested - the only difference is that this attempts to do it without a microcontroller.