Electronic Carburation

[Hugues]

This might end up to be a long post/thread, so please be patient as i develop the idea.
Electronic carburation or computer aided carburation is not a new concept, the aim is to help the carburetor fine tune the air/fuel ratio to better stay around the ideal value of 14.7 (stoichiometric ratio) with the use of sensors and actuators.
A stock carburetor without any electronic aid is designed to do just that. But if you ever opened up your carburetor you will understand how complex these mechanical systems have developed over the generations.
Maintaining and tuning a carburetor is an art and not everybody is able to do it perfectly, it takes time and patience, but even then, when conditions change (temperature, altitude, fuel quality,..) the air/fuel ratio (AFR) might drift away from the ideal value of 14.7, either the mix becomes too lean (above 14.7) or too rich in fuel (below 14.7). There is of course a range around 14.7 that is good enough.

Background:

You can skip this section if you're not interested in what i'm trying to achieve and why. But please come back to it before you shoot some tough questions.
I''m converting my 710M into an expedition truck to travel around the world when i retire in a couple of years, my build thread here: viewtopic.php?f=18&t=12644
I would like to have more information about how my engine is running (temperatures, air/fuel ratio, vacuum, fuel consumption, ignition advance..) and be able to optimize these parameters when conditions change.
But at the same time I need a reliable vehicle, one that i can diagnostic and repair myself, in the middle of nowhere with basic tools. Any system I add on top of the stock engine should fail safe, meaning that if any part of this system fails (sensor, ECU, actuator,..) I will still be able to drive my vehicle to the nearest village where i can diagnostic and repair/order potential new parts.

I have added a Megasquirt 3 Pro Mini ECU to my Pinzgauer to first better drive the ignition advance timing, details here: viewtopic.php?f=12&t=12789
This system has been running fine for the past few months and i have learned a lot in the process, especially having the vacuum and air fuel ratio in real time displayed on my tablet while driving.
I have kept my stock distributor in case my ECU fails, it would take me only 15 minutes to revert to mechanical advance only. This is very important.
This system has allowed my to see that my carburetors were not tuned properly for all driving modes.
After cleaning and synch them and changing a few parts, situation has improved but i can see the air/fuel ratio going out of range rather often. I think i can do better with the fuel i'm using and have a smoother running engine.

With an ECU like this, the next logical step would be to change from carburetors to electronic fuel injection (EFI), and I have researched the topic extensively.
Pretty much all modern cars are using fuel injectors and no longer carburetors. But fuel injectors can only be driven by the ECU. If it fails or if a critical sensor fails, the fuel injectors stop working and you cannot drive anymore. Could by a silly tiny cable or sensor, or the ECU itself. I don't want to risk this when traveling to remote places. So i'm trying to use my ECU and sensors to drive my carburetors and make a better use of fuel and have a smoother running engine. A bit like EFI, but with carburetors, hence the name electronic carburation. I've not invented this, you can find all sorts of systems on internet. We'll see where i end up, but even if it fails, i will certainly learn a lot in the process.

First step: can i use the carburetors idle solenoids to impact air fuel ratio ?

Our Pinz have 2 Zenith 36NDIX carburetors. Each have 2 idle solenoids cut off valves, so one per cylinder. Their only use is to cut off the fuel supply when the ignition key is switched off, so the engine does not continue to run on its own. They allow fuel to go through the idle port, when the engine is idling, also when the throttle is opened. So the idea here is can I use these solenoids to impact the air fuel ratio (AFR) ? The initial answer here is : Yes. As demonstrated in my short video here:
https://youtu.be/xCvJbPMWoiw?si=h3Dww5Cmwxw-s3YW
I have designed a simple electronic board using an Arduino Uno, a potentiometer, a mosfet and a relay, that can cycle on and off the 24V line going to the idle solenoid valves. And it works
In the video, after letting the engine warm up, we can see on my tablet that the AFR is too rich (~13) and the idle is around 740rpm. At this moment the idle solenoids are always opened, so fuel is going through.
As I slowly turn the potentiometer, i start to change the duty cycle of the solenoids from 100% on towards 20% on progressively. As i do this, you can see the AFR going up to around 14.5 as fuel supply is reduced, and the idle comes down to around 660 rpm. I'm currently cycling them at 10Hz, so they switch on and off 10 times per second, and the potentiometer adjust the time they are opened (duty cycle).
I know i can adjust my idle mixture screws to avoid having a too rich mixture, but i have done that many times and somehow the AFR is not always right, depends on air and fuel temperature and other things too.
So the next step will be to use my ECU to drive the idle solenoids based on AFR instead of me through the Arduino board and potentiometer. What i don't know currently is if these solenoids can tolerate being switched on and off like this, i could not find the solenoids specification. For now I will try to keep them at 100% duty cycle and only cycle them when my other method below is no longer able to correct the AFR.

Is there another way i can impact the air fuel ratio as well ?

The short answer is yes, and you can test it yourself. While your engine is running, simply unplug slowly the vacuum hose that comes out of the engine and towards the break booster, before the little white plastic valve. YOu will hear the engine rhythm change, as you let non carbureted air enter the cylinder and reduce the vacuum and thus the fuel that is extracted from the carburetor. I forgot to film it, but i could see my AFR go up when i did this, mixture going leaner. The idea here would be too always let a bit of air in, so the ECU can then use the valve to increase or decrease the AFR.

So in order for the ECU to be able to do this based on AFR, i have bought and Idle Air Valve. Most modern cars have this, it is used by the ECU to control the idle speed by letting more or less air in the engine. It's simply a valve controlled by a solenoid that opens or closes it. The one i bought is Pulse Width Modulated (PWM) and can be driven by my ECU.

So i'm going to experiment with these 2 ways of impacting the AFR and report on my progress/or failures here. Any constructive comments are welcomed, and yes, I know carburetors are meant to be maintained and cleaned and tuned to do just this. But please read my intro before going down this way. I'm just experimenting for now and learning in the process, no carburetors were hurt in the process. And i know fuel is required to cool the engine down, i have 2 temperature sensors to monitor it.

Below is the diagram of my test board, in Thinkercad, for those interested to follow me down this rabbit hole

Capture.JPG (91.74 KiB) Viewed 4292 times

[rmel]

Very creative -but- keep in mind this is an antique Air cooled engine with rather old metallurgy
especially W.R.T. the rings, valve seats, etc. Yes, you can get hardened seats from S-Tec but they
are far from what's in more modern engines today.

You really DO NOT want to run stoichiometric on these old girls -- been there done that and Valve
seat recession will be your end result At 14.7 you will run hotter and lower power. Best compromise
AFR for this engine IMHO is ~13 to avoid high Exhaust valve temperature and as a bonus you get a wee
bit more Power -- ya, perhaps a bit lower gas mileage.

I should add that the actual MIX into each cylinder will not be the same, your average may be 14.7
but one cylinder will be greater than that and one lower.

On my 710K it's EFI, I run OPEN-LOOP and worked hard to tune my Maps to stay W/I 12.5 to 13.5.
I have Pyrometers on each cylinder and my #3 is the hottest. I have a Rich/Lean knob and when going
up a very step grade this poor little engine can get hot especially on #3 -- when I increase the Richness
#3 temp drops almost immediately -- proof that theory works in practice richer/cooler leaner/hotter.

I have an AFR and Pyrometer on my 712 but it's Carb'd, so I don't have control over the fuel mix other
than fooling around with Jet sizing. However, with the 712 Jet sizes it tends to run ~13.5 which is damn
amazing to me.

Ron

[Hugues]

Thanks Ron for your valuable input, much appreciated.
Indeed I need to keep an eye on temperature.
I'm interested in your Pyrometers, which ones did you use and where did you install them ?

As for cooling especially in hot conditions and driving uphill at low speed for long time, I'm interested to add "water injection" in the air intake after the filter.
Again, nothing new, they sell kits out there, fairly easy to install and my MS3Pro Mini ECU would be able to set a PWM output to drive the water injection pump in function of engine temperature.
I also carry 100 liters of fresh water in the back. Back of the envelop calculation shows about 1 liter of water per hour when driving uphill at low speed but low gear so let's say 3000 rpm.

BTW, the Hylomar Advanced Formulation HV I've used to seal the base of my cylinders to the block is working great, i have no oil seeping through, i've used it also around the oil pan and the oil cooler. Thanks for the tip

[rmel]

Pyrometers are VDO, mounted 40 to 50mm from the Exhaust header flange.
Drilled and welded bungs in place (bungs are carbon steel, Pryo's are SS).
Typically #3 is the hottest. So if you add a Pyrometer #3 exhaust is the best choice.

If you want to add more cooling for "free" check to see how much Airflow leaks from
the back of the "Air Box" due to the 70mm exit for the Heat Exchanger. Depending
upon the air valve at the heat exchanger you could be venting up to 20% od cooling
to atmosphere.

[Hugues]

Thanks,
what sorts of temperature you usually see on your #3 ? and the highest you saw ?
To give me an idea what to expect,

thanks for the idea about the air box, i'll check it out.

[ChickenPinz]

Hugues, I would read up on how mixture is handled in aircraft engines, which have *some* similarities to the Pinzgauer engine.

As pilots, we constantly manage our fuel flow to optimize engine performance for our goals at the time. Under high load, we go *very* rich to keep CHTs under control. Fuel is a [relatively] cheap form of cooling, and in the climb-out high performance aircraft are frequently limited by the thermal limitations of the engine thus we flood it with fuel to keep it from melting down.

So....yes, you can use the "perfect" air/fuel ratio, but that may put you smack in The Red Box. Read about it here: https://www.savvyaviation.com/red-box-red-fin/ as I think this will give you some great insights into trade-offs on mixture selection. FWIW, most people are moving toward LOP operation given the reduced stresses on the engine, but there is a lot of old legends out there and traded anecdotes that cause people to fear the science, and rest comfortably in the ROP regine.

And, as rmel points out, unless you've taken the time to balance your engine, each cylinder gets a different air/fuel mixture, and you need to ensure that none of them are in the red box. You can run LOP or ROP with an unbalanced engine, but ROP is more forgiving whereas LOP can result in a rough-running (but very safe) engine when cylinders aren't well matched.

Another useful article is here: https://www.sciencedirect.com/topics/en ... 20mixture).

Note that in the aviation world we generally run a constant RPM and usually want to run the engine at the maximum safe continuous power output, so the operating conditions are different than a land vehicle going up and down through the gears. Still, and ICE is an ICE, and the article might help you in your ponderings about what mixture to run.

[eric82000]

Google translate
Hello Hugues
Bravo for your persistence in advancing your project.
Just one comment: I don't know whether the retarder solenoids are designed for oscillating operation or for on-off operation.
This could have an impact on their life!

For my part, I had thought of positioning injectors on the intake pipes (which are long enough), keeping them on the carburettors only.
Keeping carburetors only for gas shutters cut off from their petrol supply.

original version
Bonjour Hugues
Bravo pour ta persistance à l'avancement de ton projet.
Juste une remarque : Je ne sais pas si les solénoïde de ralentis sont prévues pour un fonctionnement oscillatoire ou plutôt pour un fonctionnement marche arrêt.
Cela pourrai avoir une incidence sur leur durées d'utilisation!

De mon coté j'avais pensé positionner des injecteurs sur les pipes d'admission (qui sont suffisamment longues)
En conservant les carburateurs uniquement pour les volet de gaz coupés de leur alimentation en essence

Eric

[Hugues]

Google translate
Hello Hugues
Bravo for your persistence in advancing your project.
Just one comment: I don't know whether the retarder solenoids are designed for oscillating operation or for on-off operation.
This could have an impact on their life!

For my part, I had thought of positioning injectors on the intake pipes (which are long enough), keeping them on the carburettors only.
Keeping carburetors only for gas shutters cut off from their petrol supply.
..
Eric

Salut Eric,
Indeed, i could not find the spec for the solenoids. But i have some ideas:
-I could cycle or simply switch off the solenoids when engine breaking going down a hill, to reduce or cut the fuel supply. Currently my carbs are giving a rich mixture when engine breaking. I believe modern cars switch off the the injectors when engine breaking. Of course the ECU will also monitor the temperature, but i would think this should not be a problem when engine breaking. If ever it is after a long time down hill and no fuel, water injection could kick in.
- I would then keep the fuel mixture control playing with the idle air valve, which could be set in a way to enrich or lean the mixture by letting less or more uncarburated air in
- I just thought also that the ECU could drive the stock Pinz starter to enrich or lean the idle mixture, i did just this on the highway this week, colder temperature in the morning and while cruising at 80 km/h, I noticed my AFR was too lean (16-18) and engine stumbling a bit, so i pulled the starter until i brought back the AFR closer to 14, engine was running fine, then when it got to working temperature, i pushed back the starter. This could be automated with my ECU and sensors.

Good idea for a hybrid Fuel injector/carburetor for air control. Some days ago i checked how to fit injectors in there, there is place, although tight, but don't forget the fuel rail and a way to hold the injectors in place. And we cannot keep the airbox open, the air must still be forced around the cylinders and heads.

I will experiment first with the few ideas i described above to see how far I go, then revisit fuel injectors if need be, but i really want to keep my carbs as explained in my first post.

[pinzinator]

General Motors installed computer controlled Rochester carbs on thier cars in the early to mid 80s with success. This was called CCC, or Computer Command Control. Rich and lean conditions were controlled by a solenoid that continuously raised and lowered the float level based on inputs by the computer.

[Hugues]

General Motors installed computer controlled Rochester carbs on thier cars in the early to mid 80s with success. This was called CCC, or Computer Command Control. Rich and lean conditions were controlled by a solenoid that continuously raised and lowered the float level based on inputs by the computer.

Interesting.
But how would changing the float level on our Zenith 36ndix impact air /fuel mix ratio?

BTW, I'll soon update on my progress, I have put a switch on my idle solenoids and shut them off manually when engine breaking, works smoother than I expected. Will automate it with my ECU and provide more details.

[pinzinator]

My mistake. The solenoid moved jet rods up and down, not the fuel inlet. The longer the jet was up, the richer the mixture. The longer it was down, the leaner. Google Computer Command Control, it was an interesting concept that was used before fuel injection.
There's no way you could adapt a Zenith to do this.
I have a 1983 Oldsmobile that has CCC, and it runs quite well. I have a Snap-On scanner that can communicate with it and give the currect conditions.

[Hugues]

Update on my project.
The first modification I've made is to use my Megasquirt ECU to interrupt the 24v going to the idle cutoff solenoids to switch off fuel supply when engine breaking. I've been using it for a month now and after a bit of tweaking it works nicely.
The ECU switches off a relay when RPM is above 2000 and MAP is below 40, which is only happening when foot off the accelerator and going downhill. It switches on the relay again if RPM goes below 1700 or MAP goes above 45, meaning I have pressed the accelerator.
Cool, I should be saving a bit of gas here in hilly Switzerland.
Next step I will be looking into driving a idle air valve to bring the AFR at the right place. Currently it's OK when cruising and idling but too lean when accelerating and too rich when decelerating.