On road Dyno

richyvrlimited · Post by **richyvrlimited** » Mon Dec 25, 2006 6:36 am

More Christmas thoughts ;)

Was wondering if this would be something that could be incorporated into MLV?

A mate who has the same car as me uses the LINK ECU which you datalog and tune with a program called DataLogLab. it's a nifty bit of software that supports multiple ECU's.

One cool feature it has is it can calculate the power output of your car. (you need to feed it certian information like the final drive ratio etc).

It seems to be pretty accurate. but the more important bit is it's very consistent, so much so that you can compare and actually log differences modifications have made, or if that 3 hours you spent tuning actually made any more power ;)

LT401Vette · Post by **LT401Vette** » Mon Dec 25, 2006 6:55 am

Check out 2.60
Under "Calculated fields" -> Optional Fields

There is now a HP and TQ.
These are probably not finalized but just added with the help of Dr Hess

richyvrlimited · Post by **richyvrlimited** » Mon Dec 25, 2006 6:56 am

Damnit!! how did I miss that!

Apologies:)

Peter Florance · Post by **Peter Florance** » Mon Dec 25, 2006 8:13 am

LT401Vette wrote:Check out 2.60
Under "Calculated fields" -> Optional Fields

There is now a HP and TQ.
These are probably not finalized but just added with the help of Dr Hess

Can I log this with one 3rd gear pull?

Thanks!

evolotion · Post by **evolotion** » Mon Dec 25, 2006 8:20 am

richyvrlimited · Post by **richyvrlimited** » Mon Dec 25, 2006 9:46 am

just been playing with it, and can't get any usable figures from it.

The data I'm feeding it is

weight 2244lb or 1070kg (not sure which is required)
tire diameter 22.6 (is that correct? the combined wheel and tyre diameter, so it's the whole lot in diameter)

195/50/15 therefore the profile is 97.5 converted to Inches is 7.6 inches *2 for the profile top and bottom of the wheel + the 15" of the wheel

converter stall speed = 0
Final drive is 4:3.1 but It won't accept the : so i've input 4.31
Gear ratio is the ratio of 3rd gear (as that's what I power run in) so 1.330

that look right?

Thanks!

evolotion · Post by **evolotion** » Mon Dec 25, 2006 10:29 am

i have not looked to much into the code, but going on the questions it asks, i can only assume it does not know frontal area, coefficient of friction etc (this will meen the graph will "drop off" earlier than the actual torque is). and also teh resultant figures will never bee meeningfull. however where this WILL come in usefull is finding the minimum advance required for peak torque. i.e. it is a tool for comparing one set of settings to another and tryign to see if there is a real world improvement.

i used an excell spreadsheet useing similar formula to those used above to analyse my "power runs" to work out my ignition timing.

richyvrlimited · Post by **richyvrlimited** » Mon Dec 25, 2006 10:45 am

yeah that's true, I'd not thought about adding CofE figures and whatnot. Shame as I know them for my car ;)

I get some strange spikes in the code too, at times I have 180hp (I wish ;-))

evolotion · Post by **evolotion** » Mon Dec 25, 2006 11:40 am

the spikes are jsu trev's rising sharply, like when you blip the throttle wiht the car in neutral etc. i also know the Frontal area and drog coeff for my car, but i dont care much for the power figure, just that the tool is useable for tuneing

LT401Vette · Post by **LT401Vette** » Wed Jan 03, 2007 11:53 am

Yeah the way this is currently implemented, just using an appended field, this calculation gets very heavy quick. You will notice an increase to file load time when you have HP or TQ on, so you probably don't want to have them on unless you are using them.
I would look at the current implementation as more of a proof of concept. If it is of value, I could add a new dialog that would just focus on a specific portion of the file and do much more refined calculations like aerodynamic drag along with far better averaging and smoothing of curves.

richyvrlimited · Post by **richyvrlimited** » Wed Jan 03, 2007 12:04 pm

LT401Vette wrote: I would look at the current implementation as more of a proof of concept. If it is of value, I could add a new dialog that would just focus on a specific portion of the file and do much more refined calculations like aerodynamic drag along with far better averaging and smoothing of curves.

I think it's of fantastic value, and if you could refine the calculations futher it'd be brilliant!

evolotion · Post by **evolotion** » Wed Jan 03, 2007 12:31 pm

jsut a thought, that may also speed up the "calculation" of the HP and TQ graphs for us on slower laptops, (well depending on how you implement the formulae i guess) It may be an idea to limit HP and torque calculations to TPS>90%? if the way you have the formaulas in your program allows for this?

LT401Vette · Post by **LT401Vette** » Wed Jan 03, 2007 1:37 pm

Sounds like a doable suggestion, but currently there is no support for conditional calculated fields, they are all on every record.

I'm thinking if I put in a separate dialog or tab, that focuses on one section there will be quite a few advantages.
1) the display can be layed out specifically for viewing hp & tq
2) calculations could be built into the program and compiled instead of interpolated by my math engine.
3) With the file fully loaded, averaging and smoothing could use future and past records. At load time, I can only use records that are already loaded.

RangingRover · Post by **RangingRover** » Thu Mar 01, 2007 3:59 am

do i have to do anything special to get this to work? I've got them turned on in the optional fields, but HP shows nothing in the log and TQ shows 0 the whole time, but somewhere has a maximum of 800....

Can someone give me a tip on how to get this data to come up? starting from basics would help, because I'm probably missing something blindingly obvious (i've been using megasquirt for about 3 weeks, and only just started to get the hang of tuning it decently!)

LT401Vette · Post by **LT401Vette** » Thu Mar 01, 2007 7:22 am

What version MLV are you running?
What parameters are you using?
Automatic or standard?
Can you post a sample log file?

RangingRover · Post by **RangingRover** » Fri Mar 02, 2007 12:23 am

Okay - version is 2.62....
Not quite sure what you mean by parameters, I've entered some rough values for it to calculate off (ie weight, gearing, etc) into MLV - I'll make them more accurate when I get any kind of reading out of it.
Manual transmission (i'm assuming thats what you were asking?)

And sample log.... should come up below....

BTW, is weight supposed to be in pounds or kilos?

LT401Vette · Post by **LT401Vette** » Fri Mar 02, 2007 7:56 am

For Parameters I was looking for weight, stall speed, gear ratio....

LT401Vette · Post by **LT401Vette** » Fri Mar 02, 2007 8:19 am

Ok, I just tried it on 2.62 and it isn't working....

Do an auto update for 2.63, that should fix it for you.
Let me know if that does it.

RangingRover · Post by **RangingRover** » Fri Mar 02, 2007 9:22 pm

that did the trick - cheers mate! Knew i was missing something simple...

Figures seem maybe a touch high, but reasonable enough. should be able to get some valuable tuning info out of them.

For reference sake, i'm showing 273 horsepower at 5,000 rpm - a stock 4.6 is rated at 210 hp at ? rpm. My motor has a 4wding cam which is supposed to be good from 1,000 to 3,900, but i think it works higher than that, as I suspect its been developed on a 3.5. My motor has an extra point of compression - aprox 9.25 as compared to 8.3. Also have extractors and a free flowing intake, and mildly ported heads + intake manifold. So that sounds like its probably not too far out. Torque showed at 295 (ft/lb?), with stock being 265 ft/lb - again, sounds pretty close, and feasible.

The Aspirator · Post by **The Aspirator** » Sun Apr 01, 2007 12:45 pm

Bump, this is cool! Is it purely a calculation or does it read off of the HP/TQ gauges in MegaTune? Because I'm looking at a recent log (didn't log hp/tq) and I turned on the HP/TQ fields in MLV and I get nothing. Also they say VSS HP, I don't have a vehicle speed sensor, would it be required to make them accurate?

Also if one were to dyno their car and compare it to the log, couldn't you modify the calculation to have MLV read pretty much exact?

Last question, fuel flow, what unit of measurement is that?

THANKS!

John

Edit, I'm dumb, for some reason I completely didn't notice the HP and TQ options under calculated fields, I just saw the VSS ones. Works great now! Now to do some testing