Instrumented gasifier burn - analysis / insights?

[Nofossil]

Some crack for the graph addicts.....

I finally got my thermocouples installed. I didn't get them in until part way into the fire, but it's better than nothing.

The first one is in the secondary combustion chamber. It's near the outer wall, and shielded from the actual combustion by a baffle. Combustion gases have traveled through my labyrinth before they get to it.

The second is in the flue, just outside the boiler, about 3" past the bypass damper.

I calibrated them as best I could. I wrapped them both together with a calibrated thermocouple and stuck them in the combustion gas path at about 900 degrees. I then calculated the gain and offset values required so that my readings matched the calibrated unit at room temp and at 900 degrees.

In the graph, the thermocouples are the two lower analog signals. Multiply the readings by 10 - that would give you a peak secondary combustion chamber temp of just under 1200 degrees, for instance.

The digital I/O is as follows, top to bottom:

- Baseboard demand. Introduces cold water, as you can see towards the right side
- Wood boiler recirc valve. Opens to keep inlet temp above 140.
- Tank zone valve. Opens to dump heat to tank
- Hot tub zone valve

I was surprised by a few things. The flue temp was WAY higher than I expected. I have a magnetic flue thermometer about 3' downstream from the thermocouple, and it never broke 300 degrees during the fire.

The dips in the flue and secondary temps are from the EKO controller idling momentarily as water jacket temps reach 80 C.

I plan on doing enough burns to get a good baseline, then tinkering. I might even try something as radical as cleaning the HX tubes! One problem, though: the temperatures in the secondary chamber vary wildly from spot to spot, and I can't be sure that any changes in secondary temps are due to anything more than changing the flow pattrens so that the probe is in a hotter or cooler spot.

I'm interested in any thoughts, insights, questions, or analysis. Does this all make sense? Does this suggest that there are things I could / should be doing differently?

 

[Eric Johnson]

That's fascinating, nofossil. It would be interesting to see results from different labyrinth styles. Specifically, I'd like to see how the stock u-shaped brick assembly compares to your design. With this information, you can tinker around with the design to try to optimize its performance. I wonder if they do a sophisticated analysis at Orlan's engineering department, or if they have the old "heck, it's just wood boiler" design approach.

I have an internal stack probe thermometer at roughly the same position as your magnetic one. I get flue temps over 500 from time to time, but it usually hangs at 300 to 350 through most of the cycle. But it's just a $25 Condor calibrated for use with a catalytic combustor that's been used for a couple seasons and banged around, so it's calibrated for much higher temps and may not be very accurate. At a minimum it gives me relative temps, which is better than nothing.

 

[Jim Post]

NoFo- I am completely envious of your ability to monitor/record your operation. What sort of equipment do you use to do this? I am assuming the wood in and wood out lines are water temp into and out of the boiler.... I don't have storage and I think it would be nice to know how much the draft fan is running during any burn operation...then I could tell how close I am to achieving a perfect load for a given weather situation. What sort of device would allow me to moitor my draft fan on/off cycle?

 

[Nofossil]

NoFo- I am completely envious of your ability to monitor/record your operation. What sort of equipment do you use to do this? I am assuming the wood in and wood out lines are water temp into and out of the boiler....

I use a single-board computer with homebrew software. Sensors are thermistors, thermocouples, and relays. There's some description on my site - link is in my signature below. You're exactly right about the wood in and wood out.

I don't have storage and I think it would be nice to know how much the draft fan is running during any burn operation...then I could tell how close I am to achieving a perfect load for a given weather situation. What sort of device would allow me to moitor my draft fan on/off cycle?

If the draft fan is single-speed 115vac, you could perhaps get an electric elapsed time clock. They look like odometers, and count up whenever power is applied. Just wire it in parallel with the device you're timing. People often use them on oil burners.

Anything more detailed gets into the use of a computer of some sort - a dark pit of despair that's not for the faint-hearted. Not to suggest that you're faint-hearted, just know that it's driven many great men (and not a few great women) to lives burdened with software and sleepless nights.

 

[wdc1160]

I am speechless...

Crack indeed

Also very surprised. I swear you have a staff of 5 people working on this. Really nice addition to what you already had.

I don't know what to think of the stack diffence. Maybe just a difference in location of measurment??
I expected different things at different times, and my conclusions... well too early.

 

[barnartist]

NoFo,
I always wanted to get a temp in the lower, then a temp after the exchnger tubes as it sounds like you are doing.
You must have a heck-ov-a WAF.
Mine grows tires of my laptop open. I can barely read it faster than you cpu it.

 

[TCaldwell]

nofossil, I am following most of your thread entries, it takes some time to keep up, much less find the time i suspect to research and pull all together and still have time to make a living, my wife calls the garn "the other woman". All kidding aside I am espicially interested to see how the thermocouple and the associated factors with regards to cfm modulation and combustion analysis develop.I have had a thermocouple and monitor in use for awhile , the garn secondary burn chamber is just a unobstructed fiberfrax tube 2ft.long with a inside diaameter of 8 in. I have found the sweet spot to be at the far rear sticking up about 3 inches 2260deg+, with the thermocouple at the halfway distance 2050deg+, in the front 1800deg+. this difference goes along way in trying to calibrate with flue temps. The garn is basically a 5 pass flue gas exchanger, there is a thermometer at the beginning of the last pass. deoending on boiler water temps it usally stays below 450deg, and a thermocouple that i had in the flue pipe 3 ft behind the appliance stayed between 250-325deg. these are typical in mid burn. I understand the eko gassification chamber configuration is different, but i bet experimenting with the thermocouple placement will give differences, also keep track of your boiler temps with regards to correlation of flue temps, alot more heat transfer into a cold boiler than a hot one. As eric stated"you garn guys have religion",the eko's are closing in fast!

 

[Bartman]

Nofossil,
Is there any preference between using thermocouples vs. thermistors?

 

[Nofossil]

The garn is basically a 5 pass flue gas exchanger, there is a thermometer at the beginning of the last pass. deoending on boiler water temps it usally stays below 450deg, and a thermocouple that i had in the flue pipe 3 ft behind the appliance stayed between 250-325deg. these are typical in mid burn. I understand the eko gassification chamber configuration is different, but i bet experimenting with the thermocouple placement will give differences, also keep track of your boiler temps with regards to correlation of flue temps, alot more heat transfer into a cold boiler than a hot one. As eric stated"you garn guys have religion”,the eko’s are closing in fast!

Much experimenting to be done. I'm always looking for a better way. Not that I have a competitive bone in my body, mind you ;-)

Nofossil,
Is there any preference between using thermocouples vs. thermistors?

Thermistors: cheap, very nonlinear, big signal, easy to interface to, delicate.

Thermocouple: relatively expensive, tiny signal, more difficult to interface to, can withstand much higher temps.

 

[Bartman]

How much do you pay for thermocouples? It seems that they're pretty rugged, I picked one up that has a 1/4" thread with a hex on it, I was going to use these by drilling and tapping my water jackets on the boilers and eliminate aquastats.

 

[Nofossil]

How much do you pay for thermocouples? It seems that they're pretty rugged, I picked one up that has a 1/4" thread with a hex on it, I was going to use these by drilling and tapping my water jackets on the boilers and eliminate aquastats.

They seem to go for $30-$40, and they require additional electronics - about $12 per channel the way I'm doing it right now. I'm sure that thermocouples can be had cheaper off of eBay, but I need ungrounded jacketed probes. They do seem to work really well.

 

[TCaldwell]

another thermocouple supply,along with other interesting stuff,omega engineering @ omega.com

 

[Nofossil]

another thermocouple supply,along with other interesting stuff,omega engineering @ omega.com

That's where mine come from. Really helpful, very fast shipment.

 

[SteveJ]

nofossil,

I am not sure what your daytime job is but your system design and engineering is very impressive.

Are you up for a consulting job? To design controls similar to yours - list of components, vendors, etc.

I know most of the stuff is available on your site but I think you should be compensated for your experience and knowledge...

Let me know,
Steve

 

[wdc1160]

Do you know which type of thermocoup you choose?? r?S?c?gb? something different?? What was the sensitivity range

 

[Nofossil]

Do you know which type of thermocoup you choose?? r?S?c?gb? something different?? What was the sensitivity range

Flue: KMTSS-125U

Secondary Combustion: TJ36-CAXL-18U

Both type K, I think the TJ36 is rated for service to 2200 degrees. I use them with the Analog Devices AD595 thermocouple amplifier chip.

If it ever cools down, I'll build another fire and have more data to show.

 

[Bartman]

Nofossil,
Can you post a pic of the flue TC? I've had some ideas how to mount mine, I would like to see how you have your's mounted.

 

[wdc1160]

last night was it too much wood / not enough storage?

 

[Nofossil]

Nofossil,
Can you post a pic of the flue TC? I’ve had some ideas how to mount mine, I would like to see how you have your’s mounted.

When I get a chance. I basically drilled a 1/8" hole in the thimble just behind the boiler and inserted the probe about 4".



For innocent bystander's benefit, ABGWD4U is looking at my real-time data from my site. You can get to it by clicking the 'Display' button on the 'Orlan Boiler' line in the data section of this page. I've also attached a snapshot below.

I've also caved to popular demand and added colors and legends for the digital I/O. Comments on the broken-line format for digital signals (right image)? Trying tyo decide if I like that format better.

last night was it too much wood / not enough storage?

I assume you're referring to the fact that the EKO is periodically idling, even early in the burn. My analysis is that I was burning smaller chunks of wood, therefore more surface area and a hotter fire. When there's baseboard demand (top digital line on the graph below) I cycle non-baseboard loads in an attempt to keep the boiler near but not quite at the point where the EKO fan shuts down. Last night, I wasn't adding heat loads early enough. The anticipation logic is not sophisticated yet. It takes a couple of minutes for the effect of a load change to make it through to the outlet.

The other digital lines, in order top to bottom:

Domestic hot water force
Recirc zone valve (inlet temperature protection)
Storage tank zone valve
Hot tub zone valve

Remember that the two thermocouple lines (the lower magenta and gray lines) need to be multiplied by 10.

You can also easily see when I added wood. The initial load was about 3/4 full but very loosely stacked and mostly smaller pieces.

At about 7:45, I added wood filling it a bit less than half full with medium chunks.

At about 10:15, I filled it half full with larger chunks.

Yes, I could have refilled it once at 7:45. I held off because I want it to be burned down a bit when reloading, and I like to reload it last thing at night.

I guess everyone knows my bedtime now :gulp:

 

[wdc1160]

I like the change you made graphically to show demand. Very nice touch. I know the other graph groupies think your spoiling us.