How to measure fuel consumption (flow) with a small carburetor 2-stroke engine

[razortje]

Hello,

We are going to test a small 2-stroke engine (150cc) on a eddy current brake installation with data aqcisition system connected to it.
This simple engine is still using a carburetor with no fuel pressure from the tank.

So we are looking for a device or a way to measure the fuel consumption during a cycle test (NEDC for example)
Difficulties: very low fuel flow, no real fuel pressure, accuracy must be under 1%, fuel is petrol mixed with lubrication oil.

- Are coreolis flowmeters the only option for us?

I hope there are also other options the might be cheaper.
Hope to hear some tips and tricks for people over here.

Thanks in advance!

 

[patprimmer]

Mount the fuel tank on an accurate set of scales.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[DRWeig]

And get a good stopwatch.

Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[DRWeig]

Oops, forgot to mention -- the stopwatch or a suitable weight datalogger is only needed if you need flow rates, not just consumption.

Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[dgallup]

Scales and timer are the tried and true method, the down side is you need to run steady state for a long period of time to get accurate numbers. No good for transient conditions. Companies like Max Machinery make precision zero pressure drop flow meters.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[HowDidYouBreakThat]

A burette would give you a very accurate measure of consumption. Combine that with a stopwatch and you can calculate flow.

http://en.wikipedia.org/wiki/Burette

Terry

 

[razortje]

Thanks for the fast answers allready!!

The method with the weight scale, burret and stopwatch I knew. But for a transient measurement, connected to other measurements (temperature, pressure, exhaust emission), it is not the best method because it's hard to combine those measurements afterwards in x,y graph.
So I want to measure it electronic and also transient, so I can measure and record it with my data acquisation system.

- With turbine flowmeasurement devices you have most of the time 3% inaccuracy.
- Don't know much about electric or magnetic measurements, maybe they are good, but also maybe just as expensive as a coreolis measurement flow-device..?
- precision zero pressure drop flow meters are new to me, tried to search more information about it on google but couldn't not find much about the principle and prices

Maybe other persones have experience with this type of fuel flow measurements on carburetor engines?

 

[dgallup]

Google Max Machinery or Pierburg, they both make them. Probably others but those are the two I have used.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[IRstuff]

Just bear in mind that there is often unaccounted for lags in transient measurements, e.g., in days of yore, measuring fuel input into a carburetor would result in a lag due to the reservoir within the carburetor itself, so simultaneity in measurement may still not be enough.

TTFN
faq731-376

 

[MikeHalloran]

When EFI was new, the SAE house rag carried ads for small gear pump type flowmeters that were alleged to require _relatively_ little DP. You can probably still find something like that.

I suppose you could automate the burette method with a couple of transparent tubes, four or so solenoid valves, and a couple of meniscus sensors. You'd still get discretized data, e.g. x seconds to drain from high sensor to low sensor, and it would still be delayed by the carb float as noted, but it would be pseudo-real time.

If the fuel is dirty enough (real reference fuel may not be, but retail fuel might), you could put a laser anemometer in the fuel line, I guess. Still subject to float delay. ... which should be treatable as a phase shift when you are screwing with the data.



Mike Halloran
Pembroke Pines, FL, USA

 

[amorrison]

How about a thin flexable plastic gasoline feed tube (aquariun bubbler or smaller)and a video camera with a builtin clock or a seperate clock as the timer.

 

[patprimmer]

Can't you feed data from a load cell of the scales directly into a data acquisition system and get real time weight loss from the tank.

As to delays in the carby, you would need to monitor float level and therefore changes of fuel in the float chamber.

Real time weight loss from the tank and real time float level gives enough data to calculate real time usage I would think.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[Tekton]

"Solo Wings" microlights utilise Rotax twin cylinder gravity feed carburetted two stroke engines with a very accurate electronic real time fuel consumption readout - - can't recall the manufacturer's name but they must be commercially available

Tekton

 

[BigClive]

Razor - I realize this is a bit low tech for what you want but this has proved to be a very consistent method of measurement of fuel usage:

http://helicalcamshaft.com/index.php?option=com_content&view=article&id=8&Itemid=8

No attempt was made to measure the amount of fuel used - the test was a comparison of the time the engine would idle on the same volume of fuel. But clearly the amount of fuel used could be measured easily enough.
Sometimes the simpler methods are more effective without recourse to electronic gizmos.

 

[razortje]

Thanks for the reply's!
Meanwhile I have also had contact with several suppliers and received some interesting options:

http://www.massflow-online.com/cms/wp-content/uploads/downloads/2012/03/FUEL-VIEW_leaflet.pdf

DFM-50A-K

http://www.massflow-online.com/shop/index.php?main_page=product_info&cPath=24&products_id=86

Model: DFM-50A-K
Min: 1 l/h
Nom: 25 l/h
Max: 50 l/h
Engine: 80 (kW)

http://www.bronkhorst-cori-tech.com/files/br_coritech/downloads/brochures-en/mini_cori-flow_m15.pdf

Coriolis flowmeter:

http://www.bronkhorst-cori-tech.com/files/br_coritech/downloads/brochures-en/mini_coriflow.pdf

Only, the problem for me is to determine what fuel flows will I reach and with which pressures I have to deal with (carburetor, 150cc 2-stroke engine).

I have done a calculation and based upon 1 hour of doing this test cycles the engine 'drives' virtually 22 km, with lines up with 1L of Petrol usage. So fuel flow is based upon this calculation 1L = 711 gr/hr

Can someone else confirm this? Or do they of other numbers? (just to make a comparisson)

But what pressures should I think of with a carburetor? When the fueltank is 5L = 3,5kg
The diameter of the fuelhose is 3~4mm (= 0.137 inch)

 

[icentacontrols]

For an economical option, if your flow rate is no lower than approx 1.0 l/hr then you could consider using a micro oval gear flow meter with a pulse output. An example of such a meter can be found at

http://www.icenta.co.uk/product/Low-Flow-Meter-OVAL-Flowmate.

Lower flow rates are possible but the price goes up and the accuracy falls. Hope this is of interest

 

[razortje]

Interesting option, but unfortunately the accuracy is not suitable for this kind of testing, it's ± 3% and that is to much inaccuracy.
Under certain conditions it is only 1% but I planning to test also a lot with lower flows then 1L/hour.

 

[benta]

You could also measure air mass at the intake (there's much more air to measure than fuel).
EFI systems do this, and the air-mass sensor is a standard and relatively cheap component.
To compensate for the fact that the engine doesn't burn 100% stoichiometric, measuring the CO-percentage at the same time would give you a correction factor.

Benta.

 

[razortje]

@ benta: I know this type of measurement, but to measure accurately the air-mass intake AND the CO in the exhaust, is also quite expensive, and still not as accurate within 1%.
Besides that, I think this is more suitable for a 4-stroke, a 2-stroke engine has stil a lot of oxygen in the exhaust, and CO, HC, etc.

 

[benta]

You're probably right about the 1%.
Expensive? Don't think so, this is standard automotive technology used in millions of cars: air-mass sensor and Lambda probe.

Benta.