I wondered if there is a need to install a fuel/air ratio kit, or something like that, to help evaluate the performance??
cz
Yes, Wayne has a narrowband O2 sensor with an AFR gauge in the dashboard. Even better would be a wideband sensor with gauge, they are more precise and read accurately at ratios other than 14:1.
Premium members, check out the new instrumentation page over here: http://driveonwood.com/premium/instrumentation
I got motivated to think on the temp probes today,⊠I guess I need some more info hereâ
You are using probes rated for 2100F, I first found some rated up to 2300F, but I thought i read somewhere that the internals of the unit (combustion zone specifically) reached higher temps. Do we know what the highest expected temp reading in a WK style gasifier is?
It looks like the absolute highest thermocouples available are rated for 1600C which is 2900F, this should definitely hold up. I found this while searching around;
http://thermocouple-online.com/thermocouple-T4.html
What is interesting about these is that they have a 4-20ma transmitter option - meaning they will connect to the much less expensive analog input cards as opposed to thermocouple input cards (GE PLC). - Sean, can you get your units with a 4-20ma transmitter?
Hereâs their ordering guide for this thermocouple;
What are your thoughts Sean?
Hi Henry,
So far this looks really good. I havenât tried Alumina yet currently working with Inconel and having trouble free results. The two probes on my f-150 were on my old red truck as well so over 20,000miles total on them now.
I would be interested to see what there pricing looks like. Current temps at the nozzle field at this time are not known. If I were going to measure this I would make sure the probe could handle at least 4000 f I can get these but your probably not going to like the price. Toss a nail in the gasifier and it will come back through the grate. toss a bottle cap or aluminum can in there and your never see it again. I will ask on Monday about the transmitters. Regards Sean
I was re-reading Chrisâs post about the AFR gauge Wayne uses. I found this wideband sensor/gauge kit;
Autometer 4378 on ebay at:
What I like about this kit isnât the price ($289.76 - havenât price checked elsewhere yet) but this;
âGauge has a 0-4 volt output data port that can be used for data acquisition, to connect to an ECU, or for custom tuning on your laptop.â
If this is a straight up linear 0-4V output, this could also be routed to the PLCâs analog input card. This would allow us to create a Fuel ratio subroutine that controls the air/fuel mixture for us on the fly!
Thoughts?
I like the product but we can do much better on the price. Try searching. Aem wideband gauge on eBay I use the analog dial version price should be 100 125 dollars at most
I think the wideband sensor output a linear voltage from 0-5 volts, you could get a simple voltmeter and translate the voltage into the desired ratio.
Agreed, that price is too high. I found several alternatives using that search. I am now interested in this one, and at $199.80 it is not only less expensive, it also has slightly better output resolution (0-5V instead of 0-4V); Great tip man!
As I begin to think about the instrumentation package, I am a bit torn conceptually on what is best. One of our core philosophies in the switchgear industry is to always have a backup. When you automate something, you always include a manual backup in case the cpu, plc, or whatever other component you are using dies. Many of the sensors, transducers, etc I am considering can be directly wired into the PLC without the need for further gauges, and in some cases it is more cost effective to then have the plc drive an analog gauge when one is desired, however I like the idea of also having the actual gauge to look at, just in case the plc does fail (they rarely fail, but it has been known to happen). I think my design concept will be to have (unless it is absolutely impossible) all measurement devices feed the PLC, with the ability to function independently if the PLC fails, or where this is not possible to have a second external device provide redundant indication. The automated bells and whistles are nice, really nice - BUT at the end of the day you must be able to use your ride, even if the toys arenât functioning properly.
Also, being sorta old school I just really like an analog gauge, itâs a lot easier for your mind to interpret the needle moving across a dial scale than it is for it to interpret digital numbers on a screen this was the first thing I learned when trying to parallel multiple generators.
Believe it or not this is exactly what a large portion of our industrial metering was! You would take a voltmeter or an ammeter and then provide a scale plate labeled for whatever you were metering. Old school but it definitely works, and it is very simple.
Unfortunately both of the above kits drive the voltage output from within the gauge. It looks like the wideband sensors provide a 0.45V RAW output (from sensor to gauge), meaning if we just wanted to save some $$$ by buying the sensor for $45 we would have to develop a home-brew transducer to generate a usable voltage or current output (this voltage is too low for the PLC and most voltmeters to read). Or we could buy one kit, open up the gauge and see if we can reverse engineer it, itâs probably a pretty simple PCB. This could become a huge cost saver if we all start using themâŠThen I found this;
0-1V to 4-20ma converter schematic; (Feeds signal to PLC or other micro-controller)
http://schematics.dapj.com/2005/10/0-1v-to-4-20-ma-converter.html
Couple it with something like this; (provides a gauge for us humans to look at)
This site has several useful schematics for those with the skill/knowledge to build their own PCBâs. If we devoted the time to design the PCB there are many companies that will manufacture them for you at very low cost, sometimes less than $10 / PCB though they almost all have minimum runs. If there was significant interest in going this route, I might be willing to devote the time necessary to develop a master PCB for woodgas use - after all required and desired metering is determined (no sense in making multiple small PCBâs). I wouldnât mind one that connected to all the sensors, probes, transducers etc and provided voltage outputs for gauges and 4-20ma outputs for automation, that would be pretty slick!
Although at $199 for the gauge, the sensor, the wiring harness, and the ability to drive a PLC (if desired) the AEM looks like a pretty decent value. Laziness is likely to get the best of me here!
OK, I spent a little time reading up on wideband sensors and narrowband sensors, and why thye always have controllers for wideband. This is what I came up with:
The narrow band sensors give off a voltage potential between 0 and 1 volts. Basically flipflop from one voltage to the other, to tell the computer ârich/leanâ. Computer is constantly flip flopping the sensor to try and ride the edge.
The wideband sensors take this a step further. They add an oxygen pump to the narrowband sensor, and add a small amount of oxygen to make the narrowband part stay right on the line. The current required to drive this oxygen pump varies linearly with the AFR of the exhaust gas. So we have gone from interpreting voltage to measuring current. This is where you need a controller. It has to run the oxygen pump at the right speed to balance the narrowband part, while converting the current measurement into a usable signal for the gauge/ECU. These controllers are set up to output 0-5 volts. You could then display the voltage result on any analog meter or gauge.
Summary: narrowband measures voltage 0-1v, wideband measures current and the controller outputs voltage 0-5v.
Those expensive gauges have a controller built in, there are other controllers that are inline between the gauge and sensor. Innovate makes the LC-1 and LM-1 which are inline. Pretty pricey even there.
I am familiar with the Arduino PLC platform, it may be possible to write a program to run the whole shebang and output to a cheap display or voltmeter. The chip is $4, depending on what you need with it development kits run $70 or so.
1 Like
Fuel/air
If one has loud mufflers he can set the fuel/air ratio by the sound of the motor but if the vehicle is quite the fuel/air ratio gauge really helps.
Once the gasifier has warmed up you should be able to drive out a hopper of fuel without readjusting.
I filmed this short clip just at daylight this morning.
Wayne,
You got it going on! I did notice that even as slowly as you were adjusting it, things jumped from lean all the way to rich very fast. I think you just demonstrated the problem with narrowband sensors, they are very sensitive in a narrow range, but not in the range that we need. Not that they canât be used, but thatâs why manufacturers are moving to wideband and why we should do the same.
Henry,
Check out this link: Precision Wideband Controller by Bowling & Grippo * * ©2004
Excellent Wayne. What kind of O2 âkitâ did you install? Iâm assuming you poked the probe into the exhaust, (I wonder where), and ran some wire. 2 wire, 3 wire or 4??
cz
This is great you all are making some real headway on this subject. Henry thatâs the same gauge I use keep searching you might be able to find it even cheaper. I will look when I get a min. I canât get away with loud mufflers in my neighborhood so these have been a real help. Make sure you install the sesnor pre cat converter or you wonât get good readings Regards Sean
P.S. This is the link to the same setup I have and it is also the one I would recommend notice the number scale goes from 8-18 8 rich 18 lean make sure your gauge has a wide scale range on it stay away from the ones marked e85 the price has gone up a bit since my last build
I find nothing wrong with the narrow band. The only range they display is the range you want he engine to be running in anyway. I donât need to know how far out of adjustment the mixture is.
T
Narrow band gauges use LED lights and if the mixture moves fast one way or the other and the driver misses the move, then confusion sets in. Once the extremes are hit the LED lights no longer light up. The analog gauge never confuses. I have tried both and would not use an LED gauge again. On the narrow band gauge the lean mark is no where the true lean mixture the truck can run on.
I agree Woody, LEDâs would be harder to follow. I have a dial face narrow band, no problem reading it.
Who is the gauge made by? What O2 sensor are you using?
Hello Carl,
I have been using the cheaper brands (about $30).
There is a hot wire, ground wire and a third wire that connects to the wire coming out of the computer that runs to the 02 censer.
Here are all my monitoring gauges Woody.
2 manometers, temp and AFR.(piggy back stock O2)
By the way, the thermocouple is 1" down from the restriction. It runs through the hopper where it shorted out from the moisture. When it was running, it would have the dial pegged. So of course I no longer use it. Someday I may hook it up to the grate.
Terry
