Slant Six Forum https://slantsix.org/forum/ |
|
A long range goal of this forum. https://slantsix.org/forum/viewtopic.php?t=20396 |
Page 1 of 2 |
Author: | Sam Powell [ Sat Oct 21, 2006 5:39 am ] |
Post subject: | A long range goal of this forum. |
I have heard that OEM companies spend millions on R&D for their computer programing. I can believe it now. I think ONE goal of this forum could be to establish, through questions and discussions a set of guidelines, or parameters for calibrating our own ECMs, or chips. The advice you get from books is usually pretty vague, not very specific, and not at all a routine, or protocol for this task. They do say, get your VE tables set up first, in pretty much every book, but then they kind of leave it up to you. Often they say, take it to a dyno, but I wasn't all that thrilled with my experience. It would be nice if we could generate a "sticky" thread here to keep posted we add to as we accumulate knowledge, or advice. I don;t know what the process is for getting sticky established that stays at the top, and can be added to. And I don;t know who can add to a sticky, but I think if we are aware of this as a purpose, than when we see a good guide line of a general nature come up, we can transfer it to that sticky. It might be nice if there were one person who was responsible for taking a thread and summarizing it and posting tht summary to the sticky thread. A good example of an idea that should be in a sticky is the advice about keeping the thing in open loop for months, perhaps, until you really get the thing dialed in. The rule about getting your VE tables set up first is a good one, but then how do you do that? What guidelines should you follow? Those tips could show up in a sticky for this purpose. I know this would be hard, but we could try to do it. Of course there are many other functions, and purposes this forum serves, but this could be one to add to the list. How do we get this idea started? Does anyone else out there support this kind of effort? I don't know who would pass off, or aprove of an idea as a generally accepted guideline coming from the slant six EFI forum. I also know it would take time. It would evolve as our projects evolve. It just seems as if this kind of focused effort would be far more beneficial tht simply saying "do a search" to new comers. It was Bob DeBiase's article that gave me the courage to do this, even though I ultimately went in another direction. Sam |
Author: | sick6 [ Sat Oct 21, 2006 9:11 am ] |
Post subject: | |
I support the idea, I have much to learn about EFI and would find it greatly useful, especially if as you said it was "sifted through" and brought down to just the facts. I am not a fan of the search function on this board. each time I have used it I have found it to be quite useless. quotation marks do not help, and even the slightest reference to a word comes up. just typing "spark plug" will get you body plugs, plugging of holes in the body, distributor spark curves, etc. Its very frustrating. |
Author: | Reed [ Sat Oct 21, 2006 10:34 am ] |
Post subject: | |
I like that idea. In my opinion, that is the whole point of boards like this- benefitting everyone with group efforts and a commonly accessible knowledge base. The only problem I would see is we would have to agree on what EFI system to work with so the information was applicable to everyone. We also would have to agree on a baseline engine configuration if we want to get figures for chip burniong. At least I think so. I don't know much about EFI but I want to learn, but I would think that different cams, compression ratios, headers, etc... would make a diference as to how the chips would be calibrated. |
Author: | Joshie225 [ Sat Oct 21, 2006 12:33 pm ] |
Post subject: | |
Wow, this is a large undertaking if it's done right. With speed density any modifications that affect volumetric efficiency would require changes to the ECM programming. This would mean you'd have a different calibration for a 1970 model 225 and a 1971 model 225 due to the cam change. I know the programming differences in my example would small, but this serves my point. One of the first things we need to do is make up a laundry list of engine and vehicle specific items that would affect the programming. Of course most everyone knows about camshaft specs, head flow, compression ratio, etc, but I think we need to cover smaller details too. Where did you mount your inlet air temp sensor? Coolant temp sensor? Using EGR? And on and on. Unless we focus on one ECU with one set of sensors and injectors all we're going to have is individual case studies which is fine, but not as useful. For many of us I think Mega Squirt is a very good choice as it's inexpensive relative to most systems and allows engine changes without burning a new chip as with a hacked GM ECM. Mega Squirt is well supported and flexible. I could use it to run throttle body injection on my bone stock '67 225 or port EFI with a fabricated intake on a turbo 170. Once a car is up and running with Mega Squirt (or any other stem really) we could share calibrations, but everyone that shares will have to state all the specifics of their system for the calibration to be meaningful to others. A lot of folks drive cars with engines that are mechanically stock and that would be a good place to start. As you can see I agree with Sam and Reed on most points. I think we need to develop a spreadsheet or list to account for as many variables as possible like the injectors, sensors, engine build, etc. I think I'll start today and open another thread once I have something so everyone can comment and hopefully catch all the things I missed. |
Author: | Matt Cramer [ Sat Oct 21, 2006 2:16 pm ] |
Post subject: | |
Well, I'd be glad to post my MSQ files (for you people who aren't into Megasquirt, that's a dump of the entire set of parameters) in this forum once I get my Dart dialed in. And a VE table, which may be a bit more useful to people with other ECU's. But this would, like Josh noted, only really be applicable to my specific slant six, not to every injected slant six out there - not even to every turbo slant six out there. |
Author: | Sam Powell [ Sat Oct 21, 2006 7:33 pm ] |
Post subject: | |
What I am suggesting, is not a set of calibrations or table for any one EFI set up, or a parts list, or formula, but I am suggesting we try to generate, in the process of answering each other's questions, a general set of underlying prinicpals for getting where you are trying to go. What steps come first? And then what is next? How do you proceded from phase to phase in the programing? OK, so assuming your hardware is all together, and it sort of runs, the first thing is you get a set of VE tables. Well, how do you arrive at that? I know there are multipal ways , but there are some underlying principals that can be applied in a universal manor to any set up. For instance, in my case, what I thought I needed to do right now, is improve cold start drivability. Well, now, I have come to the conclusion that it just might be too early for that. I have decided patience is the order of the day, and am still driving around trying to get readings on each cell in the fuel map, and making corrections for that. If I try to apply VE modifiers for warm up for an incorrect VE table, then I am going to be chasing my tail forever. So there is one under lying principal. Plus, I realized I had turned the closed loop feature back on way too early. This made it pretty much impossible to pinpoint troubles when they happened. There is a second underlying principal. Then yesterday, Al T suggested getting the intake air from outside the engine compartment. Undelying prinicipal number three. In the begining I did not save the original calibrations encoded by Rance before he sent it to me. There is another underlying principal: save each calibration as a file after you have tuned for awhile, so you can go back to the old one if the new one turns out worse. I am just starting to look at data logs. AT this point, to me, they are a somewhat confusing set of graphs. What are we looking for when we read these? Now, I know some things, but the answer to this question is going to be full of underlying principals for people trying to get an EFI system calibrated. If we want to share calibration files, that is probably very usefull, and my guess is we can most likely use them as a starting point for our own slants if things are similar enough from engine to engine, but there is probably an underlying principal that would govern how useful these calibrations for other engines would be when used on another engine. This is just an idea in germination right now, and the things others have suggested are also possible. Let's try to think of the most useful guiding truths here when we answer each other's questions. In fact, the questions I will be asking for awhile will be more or less brainstorming kinds of questions aimed at generating some kind of generalities that would be useful to anyone interested in putting EFI on a slant. I'll start tomorrow iwth a general question about VE tables. Let's all chime in here with what we understand about them. Sam |
Author: | Pierre [ Sat Oct 21, 2006 10:41 pm ] |
Post subject: | |
One place to start is the megamanual - its specific enough to get you going in the right directon but broad enough to be universal. Also the MSNS-E version. Sam, I know how frustrating it can be to get going in the right direction. There is no step 1, step 2, step 2a, step 3, etc type of direction for aftermarket efi. Think of it this way - how much R&D time and money go into modern engine managment systems before they are released? No way you can condense that to X pages. Thats part of the fun (or frustration) of the project! And about VE tables - Megatune w/autotune activated + wideband = tunes itself. Gotta love that. |
Author: | emsvitil [ Sat Oct 21, 2006 11:16 pm ] |
Post subject: | |
Quote:
And about VE tables - Megatune w/autotune activated + wideband = tunes itself. Gotta love that.
That's what I would consider ideal........... Do you just give it an A/F ratio to strive for (under various conditions (idle, cruise, WOT...)) and let it do the rest????? |
Author: | Pierre [ Sun Oct 22, 2006 12:54 am ] |
Post subject: | |
More or less. You have your main VE map (MAP, directly translates to vaccum on the Y axis, RPM on the X axis, and the data in the center is VE, a unit-less percentage). Then you have your Target AFR map - same axis, but data is whatever AFR you want at that point. Higher vaccum, lower rpm areas are cruise, and higher rpm lower vaccum areas are power/WOT areas. Start with the auto generated VE map (or add an arbitrary %, say 10 or 20, to make sure its rich, overly rich wastes fuel but overly lean = uhoh!), tailor the AFR map to your desire, and drive around. The algorithim measures AFR, compares to target AFR, then scales the VE entry appropriately. You can set limits as to how much it can scale and how far away on the map a single scaling can spread. If you have only narrow band you have to limit the algorithim to cruise areas. If you have wideband, you can set the WOT/power areas richer and let the algorithim do its thing. After a few days/weeks (sooner if your starting map was closer) in theory you'll have whats ideal for your engine with the settings given. You can also do datalogs to view the changes, and if you notice a general pattern you can edit the table manually to help it along. Would be wise to start driving gently don't do WOT pulls right off... It doesn't change the variable permanently - it changes them in RAM (temporary mem of processor, erases itself upon power off/on), allows you to view the changes, then if you agree you hit the "burn calibration" button. This only works on fuel I believe, not spark. Its also smart enough to turn itself off when certain enrichments come into play, like cold start or accelerator so it won't get thrown off. This translates to no variable load dyno needed, in an environment thats more realistic. Whew, sorry, longwinded. Check http://www.megamanual.com/megatune.htm#autotune for more info, and the rest of the megamanual for more MS info. |
Author: | Charrlie_S [ Sun Oct 22, 2006 5:26 am ] |
Post subject: | |
May I suggest as a real basic starting point, a glossary of terminoligy. Hopefully there will be people reading this forum, to learn about EFI, with an eye towards installing one. I am familiar with OEM EFI but some terminalogy used here is not familiar, and I would it is even harder for non auto thech to follow some threads. Is VE Table, Volumetric effiency, or something else? How about wide band and narrow band O2 sensor? Megatune, Megasquirt, Etc? |
Author: | Reed [ Sun Oct 22, 2006 8:19 am ] |
Post subject: | |
Quote: May I suggest as a real basic starting point, a glossary of terminoligy. Hopefully there will be people reading this forum, to learn about EFI, with an eye towards installing one. I am familiar with OEM EFI but some terminalogy used here is not familiar, and I would it is even harder for non auto thech to follow some threads. Is VE Table, Volumetric effiency, or something else? How about wide band and narrow band O2 sensor? Megatune, Megasquirt, Etc?
Seconded!
|
Author: | Pierre [ Sun Oct 22, 2006 10:31 am ] |
Post subject: | |
http://www.megamanual.com/v22manual/mapp.htm#gloss for terms www.megamanual.com is the main start page. We can probably get permission from the MS creators if we want to copy and paste the glossary instead of putting it link form. |
Author: | Charrlie_S [ Sun Oct 22, 2006 2:16 pm ] |
Post subject: | |
I think that would be a good idea. It could be put up as a "sticky", so it will always be easy to find for new members. Could you check on getting permission? |
Author: | Sam Powell [ Sun Oct 22, 2006 6:09 pm ] |
Post subject: | |
Now this is what I am talking about. These last half dozen posts are underlying prinicpals. I am talking about really basic things such as the terminology, and perhaps a really basic discussion of what a VE table is, and how it works, etc. Pierre, your description of Mega Squirt's Auto tune activated plus wide band is a good start at underlying principals. This type of tuning is available on many of the after market programs in one way or another, and brief discussion of this feature in Meag Squirt allows averyone, whether they are in the middle of EFI are just speculating, to grasp one of the major underlying prinicpals in the EFI project. Now, we could go on from there with a few more details about how it works, and what to watch out for when running this feature. What can get you into trouble, when does it work best? etc.etc. I am not saying we should do this now, but this type of thread would then be useful in a broad way to a lot of people; maybe especially by getting those still unaquainted with EFI used to the concepts and language of the technology. |
Author: | Pierre [ Sun Oct 22, 2006 6:44 pm ] |
Post subject: | |
Sam, a lot of those questions are answerd in the megamanual. It is quite a read within itself. I'll find the email to B&G and ask about the glossary - I think they also post on the msefi forums so I can pm them there. Will let you know as soon as I get anything. |
Page 1 of 2 | All times are UTC-08:00 |
Powered by phpBB® Forum Software © phpBB Limited https://www.phpbb.com/ |