Coretto cover project

[commanda]

I have an Ozito gun as a spare/backup unit. Out of curiosity, I cracked it open to have a look. This was not easy as you need a long shaft security type bit for one of the screws.

Anyway, the circuit board has a rotary potentiometer, an LM358 dual op-amp, an opto-coupler, and a triac. So it's basically a light dimmer.

Amanda

[saoye]

I have an Ozito gun as a spare/backup unit. Out of curiosity, I cracked it open to have a look. This was not easy as you need a long shaft security type bit for one of the screws.

Anyway, the circuit board has a rotary potentiometer, an LM358 dual op-amp, an opto-coupler, and a triac. So it's basically a light dimmer.

Amanda

I have no idea what all that means. Is a light dimmer good or bad in terms of control-ability? If it works anything like a light dimmer in the way a light dimmer delivers light I'd be happy with that. I'm going to try roasting with an understanding of the way the heat spikes at certain settings and not react to the graph while I roast to see what that does.

[Beanz.]

Anyway, the circuit board has a rotary potentiometer, an LM358 dual op-amp, an opto-coupler, and a triac. So it's basically a light dimmer.

Amanda

Thanks Amanda that is interesting, I am surprised, I would not have anticipated that looking at saoye's graph

[commanda]

The logarithmic curve is simply what you get when you do a step change of heating power into an enclosed space. The enclosed space does not step change in temperature, but takes time to warm up. The greater the differential between incoming air temperature and the current enclosed space temperature, the greater the rate of rise.

I think the other thing is, that the rotary control effectively changes the applied voltage, but since heat is power is voltage squared, you get ever increasing amounts of heat input vs degrees of rotation of the control knob.

I personally have an opinion that energy input (watts) is more important than actual air temperature as such in generating a profile.
(temperature is a function of watts vs air flow volume)
I had hoped to profile the Ozito in terms of watts vs control knob rotation, unfortunately my watt meter died. I will replace it soon.

Amanda

[saoye]

The logarithmic curve is simply what you get when you do a step change of heating power into an enclosed space. The enclosed space does not step change in temperature, but takes time to warm up. The greater the differential between incoming air temperature and the current enclosed space temperature, the greater the rate of rise.

I think the other thing is, that the rotary control effectively changes the applied voltage, but since heat is power is voltage squared, you get ever increasing amounts of heat input vs degrees of rotation of the control knob.

Hi Amanda, you obviously have a background in electrical engineering. What you said is correct, the curve is simply showing step changes and the time it took to reach temperature. I rotated the control in increments each time I thought the temperature had stabilised.

[sando]

This topic has changed to a "heat gun temp control" type topic.

"The low cost Heat Guns will have cheaper control systems, I would expect the higher cost guns may have better control systems."

I took my Bosch 630 HG and fitted an Aldi Power meter to it.
The meter shows the watts etc that an item that is plugged into it is drawing.

Started to profile my HG and check the watts drawing at the different temps.

Now I thought that as the heat went up the watts would go up at some similar rate....NOT SO.

There must be some type of PID as the watts jumped all over the place.
From as low as 125w to 1512w to 650w 157w etc not as expected....

If it has a PID in the control system that could explain the "metronome" consistancy of the HG.

Happy roasting

[commanda]

Sounds more like your Aldi power meter is responding to the electrical noise generated by the switching triac. 7650 watts, really ?
The Ozito has no filter inductor.

Amanda

[sando]

Sounds more like your Aldi power meter is responding to the electrical noise generated by the switching triac. 7650 watts, really ?
The Ozito has no filter inductor.

Amanda

Sorry that was a typo

the readings were jumping between +- 125 and 1800

I just used some numbers as a representation.
Not very accurate.

cheers

[commanda]

Sorry that was a typo

the readings were jumping between +- 125 and 1800

I just used some numbers as a representation.
Not very accurate.

cheers

OK. Those numbers I would believe. At least more than your previous set of numbers.

Amanda

[DesigningByCoffee]

What you said is correct, the curve is simply showing step changes and the time it took to reach temperature. I rotated the control in increments each time I thought the temperature had stabilised.

Hi Saoye
That's what I did with my gun profiling too. Upped the temp on the gun incrementally, waited till the DMM reading stabilised, then did it again, right through the range in 20deg increments. If I had logged mine using Andy's software, I probably would have got a similar looking stepped graph as yours - just with many more steps :-)
I suppose in layman's terms the real working difference between guns is that in an Ozito style gun you have 5-10 repeatable click stop type positions (with a range of 'hunt & hope' half-stop locations) with their set power outputs. With something like the Bosch, there are 55 very accurate stops (50-650 deg in 10deg increments).
In a covered corretto system (back to thread topic :-) ) sometimes only small adjustments are needed to control the roast. For example, just before 1C, and again at the end of rolling 1C, I drop my temp back 50deg on the gun (which in reality is only a drop of 30deg inlet temp into the pan) to coast through to 2C.
From your graph it seems like the steps on the Ozito are at least 50deg inlet temp - which would possibly stall the roast?
My brother has had a similar issue with a cheaper Makita adjustable (either too fast 1C-2C on one click, or stall on the next) and has solved it by a partial gun lift of about 2cm, giving a good result.
That's the great thing with the profiling - you can almost rehearse your roast with an empty pan, and see what'd happen :-)
Matt

[saoye]

Thought I'd add a video of a trial roast with the cover. It's in 5 parts but I just show the start middle and result.

Corretto Coffee Roaster Chimney cover coffee roasting part 1 - YouTube
Corretto Coffee Roaster Chimney cover coffee roasting part 2 - YouTube
Corretto Coffee Roaster Chimney cover coffee roasting part 4 - YouTube

[commanda]

Saoye,

Just received my replacement power meter, and had a short run with the Ozito gun. Biggest thing I noticed, was that there is a large amount of slack between moving the dial, and it actually doing anything.

That rotary knob has a shaft, which fits into a slot on the little potentiometer on the pcb. Obviously it is a very loose fit, because typically moving it one "click" forward does nothing, so you move it a bit more, and then (having used up all the free space in the loose fit) it jumps forward. Which I think is pretty much what you explained in your second video.

As a side note. Averaging out the jumpy readings due to electrical switching noise, and allowing for the dead band in turning the knob, very roughly it looks like power is linearly correlated with dial rotation. That is, dial at mid point is pretty close to half power.

What I did with my gun. I marked one dimple with some white tape. I then turned it fully counter clockwise, and marked the body with a texta at the white tape. I then advanced the knob so the next dimple was over the mark I just made, and made a new mark where the white dimple now is. Continue this to fully clockwise, and you'll have 13 marks. Lets call them 0-12 so we avoid the superstition number.

Now, using my printed roast sheet with the target profile marked out in one minute intervals, and recording the actual bean temperature and the gun setting by the minute, after several roasts you can dial in pretty close where you need to set the gun minute by minute. Have done several roasts now like this, and am getting progressively closer to designingbycoffee's recently published profile.

Amanda

[DesigningByCoffee]

Hi Saoye
Interesting vids :-) Those click-stops with no action must be frustrating. I wonder if you could file off the 'clicking' bit so there would be no slack/overrun caused by the clickstops, so you just have more of a rotary dimmer feel with all the slack taken up? Amanda is right - if you mark your own 'graduations that actually do something' and record the temp output they give (on the DMM when the reading is stable) - then you can repeat these setting when necessary.

In reality - I basically only use 4 temp setting out of my gun to get my profile - an initial warm, three temp steps up and the same steps back down. The nice thing with the Bosch is that I can make small adjustments to this to deal with ambient (eg add an extra 10 deg to everything if its cold).
But you could also make these finer adjustment another way if you needed - maybe 1/2 covering/uncovering the big chimney (trapping more heat inside) or moving the gun out a couple of cm? When I was using a fixed gun, I got a similar profile to my current one by lifting/lowering the gun a total of about 4-5cm (I had an extension nozzle on the gun) - which is not very much really!
If the ozito is linear like Amanda has said, there should be some way of pimping it for greater accuracy!

Matt

[commanda]

Matt,

There's no "click stops" as such. It just free turns through 270 degrees.

Amanda

[DesigningByCoffee]

Ahhhhh
That should make it easy then! You'll just need to deal with the backlash in the rotary dial :-)