As part of my tuning posts, I'm trying to find a free engine simulator- and was wondering if the MS system of products have that.
Anyone know?
Otherwise- is there a free simulator that one puts in some basic dimensions and have it output some basic flow estimates? That would be very, very handy when doing desktop calibrations prior to getting started.
I'm sort of surprised that you don't already have exactly that at work. Don't you guys build stuff with model based controls? We do.
In reply to tuna55:
We have them, but they are not exactly free. I'm trying to post something anyone here can get and come up with reasonable desk top numbers.
In other words- this question isn't for me, it's for the community.
alfadriver wrote: In reply to tuna55: We have them, but they are not exactly free. I'm trying to post something anyone here can get and come up with reasonable desk top numbers. In other words- this question isn't for me, it's for the community.
Wow.
In that case, thanks!
And so everyone understands, the models we have available for our stuff at work would literally be more valuable than every piece of inventory and capital we have.
I am sure your stuff at work is equally valuable, if not more so.
At first I thought you were thinking of something like one of these:
http://www.diyautotune.com/catalog/jimstim-15-megasquirt-stimulator-wheel-simulator-assembled-p-178.html
This unit can generate simulated crank / cam, coolant temperature, and other sensor outputs, and check ECU outputs such as fuel injectors and spark outputs. We have people use them to test a variety of non-MegaSquirt ECUs as well.
Airflow simulation - I'm not aware of a (non-pirated) free option. The closest thing we have commercially available is that the registered version of TunerStudio has a VE table approximator that can use a predicted torque curve and a few other engine parameters.
Other stuff out there - Desktop Dyno can do airflow calculations based on engine parameters. It won't do anything for estimating ignition timing, but the same company has a more expensive setup (around $400) that can pull that off.
MadScientistMatt wrote: Airflow simulation - I'm not aware of a (non-pirated) free option. The closest thing we have commercially available is that the registered version of TunerStudio has a VE table approximator that can use a predicted torque curve and a few other engine parameters.
A decent flow approximation is what I'm looking for. Not that interested in the estimated torque, as it's not all that useful.
I found a 1-D model from Ricardo this morning that they have for free. Not sure how good this may be- http://www.ricardo.com/en-GB/What-we-do/Software/Products/WAVE/
But a simple model like that can also be used to fill in a Speed-Map table that outputs PW. Would take a lot of intimidation out of getting a system going.
Could this all be done in Excel?
I've seen some generic VE table generators that took expected horsepower and torque and gave you a map that made you feel better about starting from scratch but was about as useful as starting from scratch. I tried one of those on my RX-7 and tried a couple different basemaps, when I did my first MS install. All equally worthless. I rolled my own the hard way: cell by cell and eyeball interpolating for the cells I couldn't reach.
I get to install a standalone on an engine for which NOBODY has ever installed a standalone that I'm aware of. I'm going to use it as an opportunity to try a table generation method, the name of which escapes me, where you start at the low load/RPM level and write that value to the whole map, then as you move higher in load/RPM you alter ALL cells past that point by the same percentage. As you iterate up the map, the error gets smaller and smaller, and the end result is that you automatically interpolate usefully correct values in the hard to get regions without having to actually go there.
In reply to Knurled:
The one I would want has some cylinder dimension info, valve side info, cam timing info, and some port info. Stuff that a builder could measure when putting the engine together. And while not perfect, it should deliver a decent start- better than nothing.
pres- I thought about that, and thought about looking it up. If I have some spare time, I'll look up some estimates in my ICE books that are decent.
What about something like the flow generator tables people use when pursuing turbo/sc sizes for forced induction? Seems like you plug in similar info, estimating bsfc and ve to generate a flow table. Let me see if I can dig up the most useful one I found a while back.
To be honest, with the use of a wideband it doesn't take that long to get started and put together a pretty good VE map. I find spark timing and cold start much more difficult to optimise. This based on experience with MS on a 3.5hp Briggs and Stratton, cammed chryco 2.4, and multiple turbo 2.4's. There were no base maps at the time - it was very early ms days.
As I recall you're in the Ann Arbor area. If it helps, you are more than welcome to fool around with the B3 dyno and it's MSII. I'm about 25 miles west of Ann Arbor.
The one you may be thinking of is here: http://www.bgsoflex.com/auto.html
Not as involved as what you want though from what I remember.
Vacation post from Florida. 
This isn't just for MS applications, although that is a very obvious place to use it.
With some basic math, you should be able to fill a speed-map to pulse width table. Kinda something I want to illustrate in my other discussion.
Paul, when I get back home I'll check some of those functions out.
Pulse width map? May be applicable to some standalones, but not all. I think the only ones I have used that use PW are Tec and 034 EFI (Pantera).
I think in VE now. Spoiled by MS. Make sure you can get to actual VE and I think it could be useful. Though I can tune a car with a completely unpopulated fuel map pretty quickly.
Now, find a good way to come up with an MBT ignition MAP for any setup and I will pay you money!
At the University of Wisconsin Madison, they had a pretty good engine dyno lab. I was looking for any information they had regarded the dyno and a thermodynamics program they use there called Engineer Equation Solver (EES; not free) and found some links. May or not be helpful.
https://powertrain.engr.wisc.edu/about.html
My thought is instead of using a spreadsheet program, use an equation solver program. I think there are some free ones in the Linux world... Wolfram's Mathematica comes free with Raspbian (Debian for the Raspberry Pi).
alfadriver wrote: A decent flow approximation is what I'm looking for. Not that interested in the estimated torque, as it's not all that useful. I found a 1-D model from Ricardo this morning that they have for free. Not sure how good this may be- http://www.ricardo.com/en-GB/What-we-do/Software/Products/WAVE/ But a simple model like that can also be used to fill in a Speed-Map table that outputs PW. Would take a lot of intimidation out of getting a system going.
There's a free version of that? I was trying to figure out where to download it and can't find a place to grab it. I could definitely find a couple uses for it.
In reply to MadScientistMatt:
I looked again later to see there is not a free one. Bummer.
I have an engine simulator program at home, not sure of the licensing of it, maybe I can "sell" it to you, CD and all?
In reply to Paul_VR6:
This isn't for me. The idea is for the community.
Yeah forgot that when I posted, I got excited when I remembered I still had the cd. 
Paul_VR6 wrote: Pulse width map? May be applicable to some standalones, but not all. I think the only ones I have used that use PW are Tec and 034 EFI (Pantera). I think in VE now. Spoiled by MS. Make sure you can get to actual VE and I think it could be useful. Though I can tune a car with a completely unpopulated fuel map pretty quickly.
I think the old Accel computer was tuned in pulsewidth. Everything else I've personally seen has been VE.
I've seen enough how-tos and how-do-Is on 034's forum that I have decided that I'm happy with never having tuned one
The zone method of building a VE table from scratch (I think it was in one of Greg Banish's books) is very compelling.
Now, find a good way to come up with an MBT ignition MAP for any setup and I will pay you money!
As will many other people! Assuming "good" means "rapid"...
Yes good = quickly and without a dyno.
Isn't this just parameterizin a basic pump fluid flow problem? Or are you going to get into non adiabatic losses, compressible flow, turbulence, and mechanical (sealing losses)? If you start getting fancy I assume you'll be accounting for flame front propigation accounting for fuel atomization characteristics and ignition timing impact? I loved thermo.
In reply to Raze:
Was most of that in English?
And Alfa, I gave some links in the other thread. Unless I can get a direct email address, I can't send the files im playing with to you.
You'll need to log in to post.
