building an arduino vacuum pressure gauge/carburetor balancer

Thread Starter

byakk0

Joined Nov 29, 2015
57
My main question is finding the pinout for vacuum sensor MPS20N0040D-D (as found on ebay and aliexpress) or a similar vacuum sensor that would also do the job. So far various extensive searches yield a whole lot of nothing. Seems to be the norm with cheap Chinese parts. It's really not a big deal for this project.
(Digikey has one, but at 10 times the cost and has two more pins-I can go that route, but for now I want to keep it on the cheap.)

I would like to connect it to my arduino uno analog inputs. I should be able to slog through the workings, so at this time I am simply after the pinout, but if knowing what I am up to helps, read on!


The project:
Carburetor balancer.
I've got 2 side draft carbs on my car (see my avatar pic). Carb balancers are available, but they are mechanical. I've just discovered the Arduino and may as well make an electronic version, right? Why not?

I'll need 2 vacuum sensors with some sort of readout from the Arduino, whether the 16x2 LCD display I have, or connecting in two 4 x 7 segment displays. Probably need 1 or two pots for calibration and zeroing.
An additional feature I would like to add is bluetooth output to a custom android app.

how I see it working:
Hook the vacuum sensors to the carbs via tubing, have a kPa readout for each sensor.

Thanks for your help, ideas, or whatever you feel like slinging at me.
Hope I've included enough info.
~Hazen
 

Alec_t

Joined Sep 17, 2013
14,280
Does this help?
According to this image, the front (as seen) has a notch. If the sensor follows the markings for ICs, the first pin counter-clockwise from the notch would be pin 1.
Sheet 2 of this datasheet shows the notch and pin-out.
 

JohnInTX

Joined Jun 26, 2012
4,787
Are you sure that you can balance the carbs using vacuum ports? Mechanical balancers like Unisyn measure airflow through the carb using vacuum created by its venturi but that's different than manifold vacuum. But if you can do it with the ports, an arduino would do the job. Good luck and welcome aboard!
 

Thread Starter

byakk0

Joined Nov 29, 2015
57
John,
I plan on taking the vacuum reading at the air intake just like the unisyn balancer. I'll probably build a little holder that bolts on to the face of each carb.
I want to display both readings simultaneously and adjustments will show any changes.

Alec,
Thanks for the schematic, helps immensely. How did you dig that up?

To make sure I am reading this correctly, looks like I connect the V++ to pin 2 and ground to 5.
What are three outputs though 2 - and one +? I just need to send a voltage range (0v~5v) to an analog input pin on the arduino.
 

JohnInTX

Joined Jun 26, 2012
4,787
John,
I plan on taking the vacuum reading at the air intake just like the unisyn balancer. I'll probably build a little holder that bolts on to the face of each carb.
Hmmmm... not sure about that. The Unisyn measures airflow by passing the intake air through a venturi. That creates a vacuum but its just a measurement technique - it really is measuring airflow into the carb. The vacuum is incidental. It will be interesting to see what your final hardware looks like. You might want to consider a hot-wire anemometer technique like that used in MAF sensors.

Anyway, here are a couple of links to TI apnotes that describe how the sensor/amp stuff works that you may find interesting, even if you are using canned arduino modules.
Enjoy.
 

Thread Starter

byakk0

Joined Nov 29, 2015
57
Yeah, I guess that is how the unisyn works:rolleyes:. Haven't had a chance yet to play with one, been restoring my car for many years....but that's another story.
This balancer is all an idea right now anyway...so anything helps.
I got the idea for the vacuum sensors from an instructable. http://www.instructables.com/id/Arduino-Throttle-Body-Syncronization-Shield/
If an airflow meter is a better option, why not? I hadn't thought of that. Think I could just grab most any MAF from a junk yard and get it to work?
 

JohnInTX

Joined Jun 26, 2012
4,787
Yeah, I guess that is how the unisyn works:rolleyes:. Haven't had a chance yet to play with one, been restoring my car for many years....but that's another story.
This balancer is all an idea right now anyway...so anything helps.
I got the idea for the vacuum sensors from an instructable. http://www.instructables.com/id/Arduino-Throttle-Body-Syncronization-Shield/
If an airflow meter is a better option, why not? I hadn't thought of that. Think I could just grab most any MAF from a junk yard and get it to work?
Looked at the link. It looks like it works because each throttle body feeds only one cylinder? Not sure what your engine does. My multi-carb setups (Corvair 140hp 4x1V and 240Z 2xSUclone) fed common manifolds so the manifold vacuum was spread over a multiple carbs. If your carbs are not interconnected on the manifold side, vacuum may work. Before taking my word for it, I think first you should see if indeed you can do it with vacuum ports - maybe scab up a couple of vacuum gauges or manometers, run the engine at a constant speed (I used a turnbuckle on the gas pedal) and play with the linkage adjustments to see if you get something meaningful. If so, go with it. If not, try plan B.

I do like the concept, though.
What is that? Spitfire, GT6, GT250?
 

Thread Starter

byakk0

Joined Nov 29, 2015
57
yes, it's a Triumph GT6-1967.
2 side draft Zenith Stromberg Carbs, 1 carb feeds 3 cylinders.
The balancers go off of intake air flow at each carb, as does the 4 cylnder setup (I think). Those are Mikuni side drafts, so the concept is the same. Either way, the concept of measuring the airflow is likely the direction I need to head. I guess I've got to delve into the workings of my carbs and see if there are vacuum ports I can hook into. Some models of ZS carbs can, but I'm not sure on mine. If I understand correctly, Colortune balancers use vacuum.
 

JohnInTX

Joined Jun 26, 2012
4,787
yes, it's a Triumph GT6-1967.
2 side draft Zenith Stromberg Carbs, 1 carb feeds 3 cylinders.
The balancers go off of intake air flow at each carb, as does the 4 cylnder setup (I think). Those are Mikuni side drafts, so the concept is the same. Either way, the concept of measuring the airflow is likely the direction I need to head. I guess I've got to delve into the workings of my carbs and see if there are vacuum ports I can hook into. Some models of ZS carbs can, but I'm not sure on mine. If I understand correctly, Colortune balancers use vacuum.
Do you mean 'Carbtune'? That looks more like your original vacuum balance concept. I looked up 'Colortune' and got this plus the video. It replaces a spark plug so that you can see the color of combustion. Neat. That would help set the mixtures but I think the Carbtune vacuum/airflow approach is also necessary to get the linkage sync'ed up. The manifold vacuum / airflow through carb is a function of butterfly position and how the engine is pumping air. That's mechanical. The combustion color is a factor of that plus how much fuel you add and if the cylinders/valves are in good shape. That's chemistry.

Back to your question though.. whatever approach you pick, there is lots of help for you here at AAC.

I had TR6's and 7's way back when and put to use all of my talents such as they were into keeping them on the road. Fun to drive but wrench-intensive. A lovely young lady I worked with was kind enough to shuttle me to the repair shop/parts store on a regular basis. I bought here lunch each time. Eventually, she took pity on me and married me. 33 years on we're still an item.. But now I drive a Tahoe. I think the hood opens. Never had to really find out.

Looks like you are having fun.
 

Thread Starter

byakk0

Joined Nov 29, 2015
57
Yes, definitely having fun. The car has been an ongoing project for the past 20 years...off the road for 18 and getting close to roadworthy again...You know Triumphs. ;)

Thanks for the help. Definitely has been welcome. I may try some junkyard MAF's just see. As long as the have a 3 wire connection I should be able to make them work. Just need some sort of analog signal I can connect to the arduino.

One other question, though unrelated. How do I add a signature for my posts? I couldn't find it in the profile area.
 

JohnInTX

Joined Jun 26, 2012
4,787
One other question, though unrelated. How do I add a signature for my posts? I couldn't find it in the profile area.
If you are using the orange theme, mouse over your avatar on the right end of the title bar then select User Panel from the drop down. 'Signature' is on the left. Sign away.
If you are using the blue theme, basically the same but you can select 'Signature' directly from the mouse-over drop down.
 

Thread Starter

byakk0

Joined Nov 29, 2015
57
Yeah, that's is where I looked, not there-even using 'find' function (ctrl+f). (I've got the orange theme)
not a big deal. If I get bugged enough I'll go through the 'contact us' link at the bottom there.
Thanks!
 

Thread Starter

byakk0

Joined Nov 29, 2015
57
Just to keep anyone interested in my build up to speed, and just to say I haven't abandoned the project.
Digging for info on something else I realized MAP sensors may just be what I need, so I just bought a pair of inexpensive GM MAP sensors. As they have a 0-5V output, that will work perfect for the analog inputs on the Arduino. The immediate plan is to simply display their output on a 16x2 LCD screen.
Later I may attempt an android app and send the data via bluetooth.
Still waiting for the sensors to show up in the mail.

As for the car, I've made much progress on it recently. I still have not yet balanced the carbs, but as they are close enough for now I have been able to take it for a couple of joyrides around the neighborhood.
 

Thread Starter

byakk0

Joined Nov 29, 2015
57
Completed the prototype.
If I suck or blow into the sensors I get a little change on the output, but when I connect it to my car there is very little change unless I rev it.
It could be the cheap $6 eBay sensors, but I need to either redesign how the sensors sample the air flow, or I just need some that are much more sensitive.


As you can see, I've got the sensors connected to the tubes, which extend to the center of the intakes.
Any ideas?
My next step is to visit a junk yard and pick up some other sensors on the cheap.
 

Sensacell

Joined Jun 19, 2012
3,432
Your vacuum pickup venturi is not going to give you a meaningful or accurate anything.

Google up "venturi" - your setup lacks the correct geometry to work.
 

JohnInTX

Joined Jun 26, 2012
4,787
At this point, you can measure the voltage output of the sensors vs. airflow. If you can discern a good relationship between airflow and voltage output, you can scale the output to fit the desired range. If the actual voltage range is small compared to the input range of the ADC you can use an op-amp circuit to make the signal swing over more of the ADC's input range for better resolution. If the signal has enough swing for the ADC but the numbers don't the map to the range you want, you fix that in the program's math. Usually, it's a combination of both. You amplify to get a good signal range to the ADC then you apply some arithmetic to scale the readings to your desired range.

Good luck!
 

DNA Robotics

Joined Jun 13, 2014
647
On a Uni-syn you adjust the center section to restrict the air flow, increasing the vacuum to the sight tube, to keep the float in range for the throttle position you are adjusting. Mostly closed at idle and more open at 1500 RPM. I'm not sure if venturi effect is needed. When you add adjustable restrictors, you can use them to calibrate your gauges.

Uni Syn.JPG
 

Thread Starter

byakk0

Joined Nov 29, 2015
57
I'll see what I can come up with.

In the mean time, is there a way to burn an arduino program to a chip? If I wanted to make this as compact and stable as possible, I'd like to get it in chip form, so if nothing else the reset button does't get pressed and lose all the data in the arduino. I do have a nano, but that only makes it a small form factor.
 

Thread Starter

byakk0

Joined Nov 29, 2015
57
Are you suggesting a venturi tube type measuring setup like this? (Just took a class covering this exact thing. Not sure why I didn't think of it ;) )
Then I can just take the change in pressure from high to low and output that. Would require 2 more sensors, but I can test the idea out with what I have...if I can find--or figure out how to make--a suitable venturi for this application. I don't have access to a lathe, so the cost of milling one would kill the project.
 
Top