That's not a 'light' task. It can absolutely be done but you can get into some nasty spots doing it. I'd have to ask what your experience with doing systems of that complexity?

There are standards or at least standard ways of running stuff like gas valves and burners. As you can imagine a lot of safety additions are needed. As for a panel shop that isn't a topic that you walk into a panel shop and declare you need 'a coffee roaster controller' in a couple of weeks. Most shops would not be capable of pulling it off.

Does it need to be a listed panel? While often that might be needed in a big city it might not be needed in a lot of places. If it needs to be listed because the insurance company said it does then you are doomed to needing the services of a panel shop.

Keith Cress
kcress - http://www.flaminsystems.com

Was the roasting not adequate with the original temperature control, presumably cycling on-off or low fire to high fire?

The firing rate temperature controller output would have to be changed to accommodate a modulating valve and the flame safety logic would have to account for purge with a modulating valve, probably a different flame safety control.

Keith's comments about a panel meeting standards like UL are to the point.

Is all this upgrade investment going to be profitable for a one-off?

Thanks for the quick feedback.

I have no experience with systems of this complexity.

After some online searching it looks like NRTL listing is required here in California, but I'll check with the city on Monday.

I reached out to a panel shop and he also mentioned this would be a challenging project and that the cost would probably be higher than I would expect.

I can't say if the investment will be worth it or not. I am enjoying the process, yet I will need to move toward commercial viability. The newer roasters have modulating gas values, data logging and control features. I guess if this turns into something too challenging/expensive, I can sell it and buy a newer one. Although, there is something special about these older rosters.

A panel was added to the roaster probably 10-15 years ago with a temperature controller to cut gas at a set point.
Temperature is controlled manually by turning vales on two rows of burners under the drum. Controlling bean temperature over the duration of the roast is less than exact. My objective is be able to have the bean temperature follow a predetermined roast curve and to do it in such a way to smooth the rate of rise.

The plan is to use an application called Artisan Scope to not only show roasting data while roasting but to also control the roaster (via Modbus). The application also has a built in PID controller. This is what I was thinking:



I really appreciate your help.

If you can use electricity this entire thing becomes almost trivial. Yes, the energy is more but the control is probably more than twice, heck, probably 5x more accurate than gas and you gain complete control of the air flow since it has nothing added in it like burnt hydrocarbons and a forced amount of air required to make combustion ratios happy. You can modulate electric heat to precise levels. No single mistake can level the place. And, as for 'modern and commercial' you can add solar to completely wipe the energy consumption and use that as a marketing tool that you can never use with gas. You can probably do your own control panel or a panel shop could probably handle that if you want to play that game.

Keith Cress
kcress - http://www.flaminsystems.com

Interesting. Just adding some NiCr coils where the burners would be? I'm not sure if that'll provide enough heat. Although, you're right, I could slow down the blower. I like where you're going with the marketing too =).

I was asked to do that for a San Francisco bean roaster (chocolate) and was about to embark on it but the guy was soooOOOo squirrelly I couldn't get him to focus on the job he wanted done and decided to walk on it.

Keith Cress
kcress - http://www.flaminsystems.com

some one,
Yes, anywhere in California you are going to need an NRTL label for it to be installed where a permit is going to be pulled. If the gas valves are not INSIDE of the panel, they don't count, but the valves themselves may need to be listed.

Keith,
Was that Tcho chocolate at Fishemen's Warf? He was fishing around for a long time trying to get someone to do what he wanted. What I finally got out of him was that he wanted to somehow (magically) measure the INSIDE temperature of his rotating copper roasting drum with no wired connections. That was way beyond me... I MAY have given him your name though, thinking of your little temperature transmitters and your proclivity for tackling near impossible tasks...

" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

Yep, that was he! They were trying to avoid $55k for PG&E to run a gas line under the sidewalk and thru THREE(!!) different government approval committees since it was on a wharf in the SF bay. Man, what a quagmire. I had no trouble getting the temp out of the 'ball' but I couldn't get any commitment or even focus.


I've actually had that happen several times and only with coffee and chocolate people.

Keith Cress
kcress - http://www.flaminsystems.com