Engine dyno coversion to miles 1

[rlchv70]

Does anyone have a source or article describing how to translate hours on a dynomometer to miles per hour. I know that you can use a "N/V" value of 35 to convert RPM to MPH, but I need a source to reference and document the derivation of this number. I am arguing with some coleagues about the proper way to translate hours into miles.

Thank you!

Randy

 

[Warpspeed]

I don't think you really can ? Maybe fuel usage might be a guide to wear and tear, but even that is not going to be a fair comparison.

Half a million miles life out of a taxi or truck engine is not unknown. A racing sports car is not going to be quite the same. Life on a dyno can be as easy or as harsh as you wish to make it.

 

[rlchv70]

The engine in question would be on a pre-defined cycle.

 

[ivymike]

if you want to convert engine rpm to vehicle mph, then you need to multiply (engine rpm) by (wheel revs)/(engine rev), which depends on the transmission, transfer case, differential, etc., and then multiply by (miles traveled)/(wheel rev), and then multiply by (minutes)/(hour). That would give you an equivalent speedometer figure... what the number means for practical purposes is another question altogether.

 

[patprimmer]

Isaac is spot on.

Also the engine might not be able to pull those revs when in a car in a higher gear.

There are formula's to convert co-efficient of drag and frontal area to power required for particular speeds. These can be compared to engine output at the wheels at full throttle at a similar speed. Obviously if the power output is less than required, the speed will not be reached.

Dyno hours do not translate well to engine life unless you have a lot of data about conditions of use, and an ability to translate those conditions to data to control the dyno load and engine speed.

Regards
pat pprimmer@acay.com.au
eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[GregLocock]

Rule of thumb: if the engine will survive 100 hours at full power on the dyno then it can probably be made to survive in service, but specific issues at lesser speeds and torques may need to be solved.

If it can't sustain 100 hours then don't bother looking at other speeds and so on.





Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[rlchv70]

Thanks for replies. However, an SAE document or EPA document would be extremely useful. Some background info: We are trying to calculate oil consumption. Our specification is in miles per quart. Converting the hours to miles using ivymike's method meets the specification. However, I need something to document that this method is valid. Using the method that my coleagues does not produce results that meet the specification.

Thank you,
Randy

 

[ivymike]

if you want to convert dyno hours to vehicle miles in a way that gives you the right oil consumption, than you'll usually need to run a specifically tailored oil consumption test cycle which has been correlated to some degree with actual vehicle testing. Usually in such an arrangement then engine would be operated at steady state in about 8 or so different operating regimes, and it would undergo brief transient operation periods between each of these points. Transient oil consumption is much different from any nearby steady-state points, and the various combinations of load&speed can give widely differing results between steady state points.

 

[franzh]

Southwest Research Institute in San Antonio Texas has developed a method of calculating oil consumption using radioactive tracers (low level) that has been pretty accurate at short terms.

http://www.swri.org/4org/d03/vehsys/filfine/rtoil.htm

Franz

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[ivymike]

That method gives you short term engine oil consumption, which doesn't necessarily tell you anything about long-term oil consumption in a vehicle. Engine oil consumption measurement by this method or any other is a useful tool when improving the oil consumption of the engine, but at best lets you make an educated guess about what will happen in a real vehicle.

 

[rlchv70]

We already have an oil consumption cycle defined. We know how much oil the engine uses. We just need to translate lb/hr into miles/qt.

 

[ivymike]

Then it sounds like you just need to translate your oil consumption cycle into a drive cycle (may already be done, depending on your cycle). If your OC cycle was representative of vehicle use (in terms of the load/speed histogram & amount of transient operation), then the simple conversion at the top of the thread should be as good as anything else.

I'm a bit surprised that a company with a miles/qt bogey doesn't have experience relating oil consumption measurement in test to typical results in the field. I wouldn't expect to find an EPA document on the subject, since they don't directly regulate vehicle oil consumption, and I would be surprised if an OEM gave away so much info about the behavior of their engines & vehicles in an SAE paper (but someone might have). You could try calling an engineering consulting firm with automotive experience - Southwest, Ricardo, et al. may have an answer for you.

 

[rlchv70]

Our company supplies engines to one of the "big 3". Our company converts to miles based upon fuel usage. Our ring supplier converts to miles based on the rpm to mph conversion method above. These 2 methods produce similar results. Our customer converts it using an average MPH from the fleet that the cycle is based upon. Since the cycle is not a 1:1 correlation to the fleet data, we feel that the average fleet MPH conversion is not accurate.

 

[ivymike]

I'd tend to agree. I'm a little unclear on the details, though-
* Is the OC cycle in question based on this particular fleet's operating characteristics? (I assume that's what you meant)
* If it is, then in what way doesn't it correlate, and why?
* If the cycle is based on the operating characteristics of the fleet, then shouldn't the cycle weighting factors give you an effective average MPH that is the same as the fleet average MPH?
* Does the average MPH calculation exclude any key operating points? Does the cycle overemphasize certain load points, to the detriment of the average speed calculation (too much idle, not enough low speed low load, maybe?)

When it comes right down to it, it sounds to me like you need to talk to someone at the customer to find out why your methods don't align with theirs, and see whether some sort of conversion or adjustment can be agreed upon. At the end of the day, whatever requirement they have is going to be the one you're stuck with, right?

 

[rlchv70]

Our company supplies engines to one of the "big 3". Our company converts to miles based upon fuel usage. Our ring supplier converts to miles based on the rpm to mph conversion method above. These 2 methods produce similar results. Our customer converts it using an average MPH from the fleet that the cycle is based upon. Since the cycle is not a 1:1 correlation to the fleet data, we feel that the average fleet MPH conversion is not accurate.

 

[rlchv70]

The cycle is based upon the fleet data, but the points were chosen using "engineering judgement". Therefore, the cycle excludes some of the lower speed, lower load points.

 

[dgallup]

When I worked at Mack Trucks we used our own dyno test cycles to evaluate oil consumption, durability, etc. We also ran a large fleet of test vehicles to correlate the dyno tests to actual vehicle data. Some tests were designed to be very brutal to produce failures quickly, others were less severe. We never had a straight conversion factor between dyno hours and vehicle miles. It's much too complicated to be reduced to a general purpose rule of thumb.