Any ideas on how to accurately measure mpg’s on an older car with a carburetor?
 

So I couldn’t find any recent posts on this topic so I thought I would ask – does anyone know of an accurate means to measure fuel flow? I can’t use any of the modern Scangauges as I have a ’66 with a carburetor and an analog tach. I have been making a lot of mods and I just don’t think the pump gauge is cutting it unless I take a real long drive and get the same pump. The less expensive boat fuel meters seem to be discontinued and I can’t justify buying on of the $400 versions. Thanks for your help!

You might find an old zemco unit. We have discussed inexpensive flow meters here and how to monitor them and report mpg but nothing solid. There are too many fuel injected cars to make it a priority IMHO, and converting to fuel injection can probably be done for less than $400 with some boneyard parts, and would provide better economy/control/monitoring.

Have a look at fuel flow meters for marine engines ($$$, though, such as FLOSCAN). The trio of driver gauges would be:

Fuel Flow
Vacuum
RPM

Additional would be:

Airspeed


Depending on the car there are (or were) spark timing adjustment boxes (MSD, and others) where one can advance or retard ignition timing. With the above gauges (and maybe a knock sensor) can can "lean out" the engine for best cruise parameters.

Some of the "feedback carbs" worked in a similar manner.

What kind of car is it?

EDIT: I see you are aware of marine fuel flow meters.

.

I had a scangauge TYPE of product back in 1979 that I ran on my 1966 1bbl carb 200ci Mustang.
I think i alsoways had to tell it how many gallons I put in.
It worked great.
I used to get 27mpg on the road.

You can calculate mileage with pencil & paper, or a simple calculator. An electronic gadget isn't really necessary.

My very first car, given to me by a wealthy cousin when I was in college, was a jalopy, and a real 'land yacht': A 1963 Caddy convertible. It even had tail fins and real chrome everywhere in it (which was totally out of date by the time I got it.) :cool:

Sure, it was a gas guzzler - but what wasn't back then? - except maybe a VW?

It sure was fun to drive though: No pollution controls. No seat belts. Kettering ignition points. Rotted exhaust system. Distressed paint. Bald tires. Leaking radiator. But the leather seats, the radio and the convertible top all worked great until its dying day, despite it being on its last legs. Fond memories... gas only cost about 69 cents per gallon then.

As far as instant/current information goes (extremely valuable when dialing in various techniques and mods), aside from the zemco unit, there was one case where someone sourced a flow meter and hooked it up to a bike computer (fuel consumed is measured as a "distance). Now you can do a lot better than a bike computer, i.e. arduino or mpguino for a starting point, but you get the idea.

http://ecomodder.com/forum/showthrea...html#post12516

:eek:REALLY!!!!!!!

:eek:
:eek:
:eek:
:eek:
:eek:
:eek:

:eek:

And you think I'm soooooooo stupid that I don't know this!:rolleyes:

Please, help me (help you) .......explain the purpose of that sentence attached to my quote.

ps. and a car isn't neccessary either......WALK!!!!!
(oh gee that almost sounds as stupid as your pencil/paper comment.....:cool:)

Wow! You certainly do read into things, taking great offense at a simple observation. Go back and read it again. Nothing offensive was intended, stated or implied.

Maybe calling yourself "stupid" is appropriate after making such an emotional spectacle of yourself with your incredibly HOSTILE reply.

No, maybe YOU need to reread your passive/agressive post.

The op wants to know about " xxxx"
not about "yyyyyyy"

you decide to quote MY post and say that "geee you don't need "XXXX" silly boy, just do "yyyyyy".

Maybe, you are 'tonedeaf' and don't 'hear' what you are saying anf how you are saying it.


So you:
1. quote me and not the op. WRONG
2. Offer up a non solution. not related to the op's request. WRONG
3. Don't get it. WRONG
4. Come back with passive/aggressive. "gee why are you picking on me????" WRONG

conclusion: Your incredibaly out of touch, clueless, tonedeaf or stupid.

Guys: you need to calm down, seriously.

Keep the personal attacks / insults off the forum, please.

KEEP IT CIVIL or take it elsewhere. Just drop it.

1966 model would have virtually no inspection restrictions, at least here in Virginia.

You might consider using a burette and a t connection into you existing fuel line at the point where the rubber hose connects the metal supply line to the fuel pump (assuming it is mechanical). You would also need a means to disconnect the main fuel supply and switch to the burette (mechanical cable to a switching valve).

This would give you very precise measurement of fuel delivered to the engine, but in a limited quantity depending on the volume of the burette. You would cap the burette with a vent fitting that was directed to outside air, or just place the burette outside of the passenger compartment in one of several places. The burette should be of a plastic type, not glass, and be impervious to your fuel, with no open top that could spill the fuel.

regards
Mech

A compromise between finding & installing a fuel flow based meter (instant feedback) vs. tank-to-tank MPG calculations (feedback may be too slow for those trying to evaluate mods) is to calibrate your fuel tank sending unit and effectively create a "trip gauge".

- With your tank nearly empty, connect a digital multimeter to the fuel tank sending unit (measuring resistance)

- Add a small measured quantity of fuel (say .25 gal)

- Note the change in the multimeter reading

- Repeat until tank is full (multimeter reading stops changing with added fuel)

Now you have a graph with decent resolution of fuel level vs. sending unit resistance, which you can use to figure out fuel consumption. (Assumes you park in the same spot as when you calibrated the sending unit... and the sending unit isn't sticky etc. etc.)

Metro, I did just that with my brothers 91 Civic hatch to do some testing on a practically deserted road near his house. Although the OP would not have that problem the loop return fuel system added heat to the tank as the return fuel was warmed by engine heat. On one trip I went 9 miles and when I got back the reading had increased instead of decreased!

Once I figured it out and learned to compensate I was getting close to 90 MPG doing 15 to 35 P&G with engine off glides. No lights or traffic made it easy.

regards
Mech

Those old Zemco units (I have one) can be inaccurate in certain situations.

The fuel flow sensor has an optical sensor that counts the circulation of a small ball in fuel flow. It also counts the bubbles in fuel and if you have a mechanical fuel pump it typically creates bubbles when hot.
Especially when the engine is hot and you're idling at traffic lights the mpg reading goes nuts. But during cruising it can be quite accurate.

Sometimes I have played with a thought that how the MPGuino could be used in carbureted vehicles. At least it would be nice to see the differences in A-B-A testings when you have done some mods.

So, my idea is not to measure the fuel flow, but instead we measure the load of the engine. That could be done with a Manifold Absolute Pressure sensor, which typically give 0-5 volts according to the pressure.

If we can tell the MPGuino that "Manifold vacuum x rpm = engine load" then I think we should be able to see if certain mods will have less load on the engine. But of course we cannot see the real fuel consumption, since the carburettor's fuel delivery is not absolutely linear with engine load.

And to make MPGuino understand the vacuum signal, it would require changing the code or changing the 0-5Volts to PWM (pulse width modulation) and also we should input also the rpm info.

One other option would be using a MAF sensor, which measures the air entering to the engine. Then the rpm information would not be needed, I guess.

Thanks for all great feedback. I am going to try to find a burette for my idle / regen / vacuum adjustments that I can do while parked in my driveway. Depending on the final setup, I may or may not use it for actual trips. I’ll post a pic if I get it working.
I also like the ideal of calibrating my fuel gauge by adding small increments of gas and measuring the sending unit’s resistance.

Has anyone tried mounting only the bottom of a burette to the fuel line to use it as a visual fuel gauge (like you might see on older water tanks)? Although MetroMPG’s method sure seems safer – and thank you Old Mechanic for the tip to this method.

Slowmover – it is a 66 Mustang with a straight 6 (200ci) engine and a brushless DC electric motor.

Thanks again everyone!

NCWA

Depending on the car there are (or were) spark timing adjustment boxes (MSD, and others) where one can advance or retard ignition timing. With the above gauges (and maybe a knock sensor) can can "lean out" the engine for best cruise parameters.

Hey guys - after playing around with a few options I found an interesting data point. I burn 1 oz. of fuel each minute at idle! That is 0.47 gallons per hour. One way to look at it is that I need to be rolling along at ~14 mph while idling if my target goal is 30 mpg, which is a challenge in my 1966! I have no idea how some of you are getting close to 60 mpg’s!

Greetings Gentlemen
I have seen this & other posts about flow sensors for carbs with the Mpguino ,I have been tesing various devices over the last few months(talking cheap here),the current device I am testing is showing some promise.This sensor (2-30 litres/hour flow rate) is available from Futurlec.com at US$13.00 but needs modding to make it's signal compatable .
So far the results are looking ok , but more mods are needed to the sensor to at least stabilise the readings
I will try to add some photos & better explanation tommorrow.

Regards
Tp

Welcome, and thank you, hope to see more. The "trend" of agricultural or other small engines into vehicles would benefit by this also. Was discussing this kind of problem with a power equipment repairman the other day.

.

Thanks slowmover
here are some photos of the flow sensor ,it uses a single magnet (black dot) on the internal paddle wheel ,the other black dot is steel counter weight.
Externally the magnet sensor is hall effect device,in in original form the gph reading is extremely erratic , so far I have added another hall effect device externally & this improves the gph stability quite a lot ,I want to add a third device on top & see what we get.
From my bench testing of the mpguino it nees at least three magnet sensors to show smooth gph readings.

Tp

Nice!

FYI, my saturn idles at about 0.3 gph. This looks to have a lower limit of about 0.5gph. Not a big deal if you hardly idle, but just fyi. It would be good to see how it tracks on a tank to tank basis.

Carb flow sensor & speed sensor
 

the flowsensor has a 1.3mm orifice at the inlet to the paddle wheel ,even though it has a 5/16" hose fitting ,I am hoping adding more sensors will allow it to work with lower flow rates.

As for the speed sensor ,see attatched picture ,it uses the same hall effect sensor I have used to mod the flow sensor .
I have left it unpotted & exposed until I have tested it further.
So far the speed sensor works ok & it's simple (three electronic components & two magnets).

speed sensor for carb engines
 

a better photo of the speed sensor

...hey, TestPilot, there's some info about the Flo-Scan™/Space-Kom meter is this earlier posting:

http://ecomodder.com/forum/showthrea...eup-18198.html

Thanks Old Tele Man
Yes I did see that post. My motivation for the flow sensor was to replace my old Zemco sensor ,that dies about 5 years ago. . I spent enormous amounts of time trying to stop it reading the vapour ,about 15 years ago,I gave up.
At The moment this sensor looks good as far as vapour goes ,it's summer here ,but low temps ,about 28 c ,so still testing.

I've had problems with the vapor too. I've tried many different things to prevent the Zemco counting the bubbles. Last summer I took the sensor off the car, as the hot weather + 2hr at highway resulted in a total vapour lock and the engine did not get any fuel.
Then I decided to get an electric fuel pump, Hoping it would solve vapour problems. Maybe I get it installed for next summer.

Are you using this new flow (and speed) sensor with MPGuino?
I would be interested, how have you connected the hall sensors to it?
I quess you need a few electrical components to adapt the signal for MPGino?
Any schematics?:rolleyes:

...have you considered installing one of the 1970's MoPar "vapor-seporator" fuel filter cans just before the flow-meter unit?

...I'll try to find a picture (below), but it basically was a regular fuel-filter assembly with three pipes: an inlet on bottom and two outlets: one for FUEL to carburetor and one for returning VAPORS to gas tank.

...here's a picture of a HASTINGS GF84 replacement unit:

http://autoparts.junkyardlocator.com...tings/GF84.jpg

I quess you meant this question for me?

Well, the answer is no, since I never had seen or heard of this kind of separator before.
But I created a system of my own.:rolleyes: Since the mechanical fuel pump attached to engine block gets hot, it will make the fuel boil. Especially when the flow is slow. (and I have tried insulating gaskets in between) And the bubbles will expand in both ways.
So I installed the flow sensor before the fuel pump and also put 1 meter of fuel hose and a big fuel filter in front of the sensor to catch the bubbles.
This setup helped to get more accurate readings, but there was also quite big delay in the readings.
But I hope that the electrical fuel pump will solve the vapour things, since it's not connected to the hot engine.

Hows this one?
Kind of cheap with a high lower limit read out.

NavMan Fuel Gauge Model F2100

The Flow & speed sensor Photos I have posted ,are being used with the mpguino currently,my old zemco was retired when I bought the mpguino .
As for vapour separators ,I used one from a 70's 430 mercury wagon ,it worked ok but the zemco flow sensor could have been though out better ,the optical sensor is just too sensitive .
Later today ,I will post the circuit for the speed /flow sensor here.It's relatively simple so everyone should be able to coble one together.

...so, you're converting "pulses-per-unit·time" into "pulse-width modulation" (PWM), correct? Is that what the display uses?

you get pulse width variation by default on my circuit ,(so you need to count the 0 volts "on" time of the output) , which is the same as a fuel injector signal but a lot less pulses & that is where my problem lay ,with the sensors I have previously tested .
The speed sensor I assume just counts the change from 12 volts to 0 volts & not pulse width.

Still no hot weather to test vapour issues ,February is traditionally the hotest month here ,mostly high 30's so here hoping.

Greetings from the opposite side of the globe!:)
February is usually our coldest month here. Now we have -15*C and plenty of snow.
I took this picture once I was coming from work by bicycle.
http://ecomodder.com/forum/member-su...492-winter.jpg

Thanks for the schematic! I might use that for my speed sensor.

I think you are right that you can get pwm from your fuel sensor since paddlewheel speed will wary. But I would think it would be more accurate, if you could modify the MPGuino code for counting the amount of pulses.

Hi Superturnier
Yes , I agree with altering the software to suit the sensor ,but from a previous post ,I have read , that isn't going to happen anytime soon ,unless of course , I learn something about writing software ,I would like to , if I had some more free time.
I was hoping ,to come up with a simple solution , so you can swap the mpguino between Fi & Carb vehicles .
The weather here is similiar to Los Angeles, -15 too cold for me ,even in car.

I have done some small modifications to the MPGuino code to suit better for my needs, but nothing big. Mainly copy +paste things.
Although I do automation software for my living I'm not so familiar with the MPGuino language. I guess it's some kind of C+ or something. I wish I could learn it more, but my free time is also too limited:(

I agree, many times it's best to keep things simple. It would be nice to have a solution also for the carburetted vehicles.

This is the latest mod to my fuel sensor ,it now has 4 hall effect sensors .
I am hoping to improve the output at idle ,for a steady gph reading .
The 2 sensor version seemed to be consistent as far as final tank readings go ,but the instant display was all over the place.
will do some tank testing over the next week.

Update for the previous mod to my flow sensor.
This sensor , I am now sure ,that they have got the rating confused & it should be rated at 2-30 litres a minute not 2-30 litres an hour as the data sheet states .
After much modding of this sensor ,I am going to call this one not suitable at this time as we need something that is simple .
I am looking at other methods to measure the fuel flow,including the sensor from Conrad.de .
Tp