Learn Me Standalone Engine Management

Hey everyone!

Currently in the process of doing a 2JZ-GE swap into an RX8 - but the engine will be receiving boost as well as TT rods/pistons.

The initial plan was to run a JDM Aristo TT Ecu - which can be done quite inexpensively - however, I'm also looking into running a standalone as there are various little things I need to buy to use the JDM ECU. It ends up being about 500-600 total - and will safely run 440cc injectors and about 15ish pounds of boost - which should be enough for my power goals of 350-420whp.

Anyhow, with the cost being 500-600 and a Megasquirt 3X being about 750...  I'm considering "stepping up" but I'm also concerned it'll lead to a snowball effect where my goal posts keep getting further away as I build the car.

While something PnP would be nice, I really don't want to spend 1.5k on a standalone because that's just not grassroots enough for me.

I spoke with a couple professionals - and of course, they all think a brand new 1k dollar harness, 1.5k on engine management, and another 600-1200 for tuning is "the way to go".

Even though the car was 700, the 2JZ was 200, and the BMW transmission was 300.... 

Anyhow, I'm sure many of you have been in this situation before and I like how cost-conscious and creative everyone here is... 

I'm building my car as a "learning project" but don't really have any idea how to set-up a MS3x... which of course, can be a recipe for disaster but I wanna be able to build my own car... 

Tuning nowadays seems like spending 3k on stuff and loading up a basemap.

Should I just stick with the OEM ECU? Or should a try the MS3x? Or is there a new "budget" standalone now that Megasquirt is almost roughly the same price as LINK, ECUmaster, AEM systems?

Tuners back in the day seemed to be able to do so much and build cars so well running piggybacks... Should I try that instead?

Thanks everyone!
 

IMO the MS3x is a powerful ECU with plenty of features.  It is pretty much the default option for turbo Miatae so there's tons of experience and support for that application online, and base maps aren't hard to find.  For your application, I think you'll be doing more of a build from scratch thing.  Matt Cramer from DIYAutoTune wrote a book called Performance Fuel Injection Systems (or something like that).  It's a very good starting point.

I recently put an Ms3 Gold Box (EFI Source) into my project mustang. The basic harness is reasonably priced and I recommend it, at the least. Heck, knowing what I know now I might well have sprung for a PnP solution. 

Do you have a throttle plan? MS doesn't do throttle-by-electron. 

Do you have a gauge plan? There are various solutions there, depending on budget, but most aren't cheap. 

With MS you will also need TunerStudio which is also affordable for the full version but also an add-on to your number. 

I'm planning on running a throttle cable and as fair as gauges go, this little box from Russia seems to be able to take care of that.

https://all4swap.ru/product/rx8

It's essentially a CANverter box that would allow me to retain the factory dash.

So essentially, I'm debating between running a TT ecu or a PnP standalone like Ecumaster.

However, I think I might be leaning towards the later as the TT ECU, TT wiring harness, that converter box, and a used "old-school" piggy back ECU brings me close to 600-800 depending on how well I can bargain.

With the Ecumaster only an addtional 500 or so... I might be willing to "splurge".

The problem with this, however, is now that I know I'm not capped at 440cc injectors or 15ish lbs of boost - the build might end up snowballing into something that 1) wouldn't be the car I wanted to build in the first place 2) stuck in project car purgatory. 

Realistically, I can't outdrive a 400whp 3000lb car yet... so maybe it's better to be cost conscious so that I can actually use that on all the miscellaneous things we all "forget" to add to our shopping list. 

Someone please talk me DOWN so I can stay grassroots. 

As much as I wanted MS3X to work for me... it really doesn't at this point in time. The TT Ecu will run the car essentially like stock within certain parameters. Seems like there's too much that can go wrong as far as user error is concerned with the Megasquirt... 

And price wise, it's much closer to the PnP standalone.
Not so sure it's the budget option it once was before. (Still great for people who know what they're doing)

You want grassroots just put a microsquirt on it and be done. 
 

The last Toyota friends and I mangled we were able to double the power and more by adding an old aem fic piggyback. It works ok but has its limits.  Trouble is it's the same work or cost to get either one tuned by a pro.

The cost to build and develop a decent engine management package doesn't change as the rest of the car depreciates. So yeah, put a $1k ECU on a car with a $300 transmission. 

The most important thing is support. Get an ECU that can be tuned by someone who knows what they're doing. This doesn't mean a forum full of 300 people that have each tuned one car, you need one guy who has tuned 300. There is no substitute for that depth of knowledge.

Also, really consider if you want one. You will be responsible for all the hard parts of making a car run. Not full power, that's easy. Stuff like cold start and idle control and AC compensation. All the stuff that most amateur tuners just gloss over and accept will be crap.

Have you looked at what Holly has to offer?  I have seen there systems on several car tv shows and they look way to easy. Probably geared towards the LS motor swap crowd but still looked interesting. 

Keith Tanner said:

The cost to build and develop a decent engine management package doesn't change as the rest of the car depreciates. So yeah, put a $1k ECU on a car with a $300 transmission. 

The most important thing is support. Get an ECU that can be tuned by someone who knows what they're doing. This doesn't mean a forum full of 300 people that have each tuned one car, you need one guy who has tuned 300. There is no substitute for that depth of knowledge.

Also, really consider if you want one. You will be responsible for all the hard parts of making a car run. Not full power, that's easy. Stuff like cold start and idle control and AC compensation. All the stuff that most amateur tuners just gloss over and accept will be crap.

To follow up on this post- if you want to really DIY it, you should spend a lot of time reading about how the control system works, and very much understaning how to tune it.  For instance, the features you would get in the MS3x can be incredibly powerful if you know how they work and how to tune them.  

If two controllers do the same thing, but one is more powerful in a manner you will never use, well....  

And that's why a Microsquirt may be a really good option.

Anyway, what features does the engine *require*?  Backtrak from there.  And include the evaluation of who is doing the tuning.  If you are, and don't have the time to really learn how it works in depth- KISS.

Keith Tanner said:

The cost to build and develop a decent engine management package doesn't change as the rest of the car depreciates. So yeah, put a $1k ECU on a car with a $300 transmission. 

The most important thing is support. Get an ECU that can be tuned by someone who knows what they're doing. This doesn't mean a forum full of 300 people that have each tuned one car, you need one guy who has tuned 300. There is no substitute for that depth of knowledge.

Also, really consider if you want one. You will be responsible for all the hard parts of making a car run. Not full power, that's easy. Stuff like cold start and idle control and AC compensation. All the stuff that most amateur tuners just gloss over and accept will be crap.

Fair and good point. While everything else can be bought used, that normally doesn't really apply to engine management. I've spoken with Matt Cramer and he was extremely helpful with trying to get me started on MS but I don't think I'm actually ready for that just quite yet.

Most of the tuners I've spoken to locally have not given MS a chance, or used it and didn't like it... This puts me in a rough spot because I've been aware of MS since their first iteration and I'm very much aware of how far it has come.

I think at this point, I may indeed just use a JDM TT ecu, would have been nice to use a USDM Supra ecu but that doesn't control VVTI. Or, I may just go with the Ecumaster Black with the PNP adapter - literally plug in two connectors and I'm "done", and can load a basemap that isn't terrible.

In reply to dean1484 :

Yeah, Holley is geared more towards domestic motors and they tend to make great products... although a stock LS engine harness can easily be converted to "standalone" for very little work or money - I can see the appeal of the Holley EMS.

In reply to PHAN :

I'll add another viewpoint, as I've just recently tumbled down this rabbit hole(though I fortunately have a teammate doing the tuning). Obviously once you add a turbo you really need to control fuel & timing well enough to prevent knock and lean conditions under boost. The higher the boost(relative to what the engine can handle) the more serious of a problem it becomes.

So I'd say you need to view it as a sliding scale: you can get away with minimal or somewhat less precise engine management if you run very low boost(or remain NA), but as you start cranking it up you're definitely going to need to spend more money and gain more knowledge - even if that knowledge is just figuring out who will help you tune it & what you need to understand to work successfully with them.

For example, I've gained a reasonable understanding of how the systems, sensors, and data inputs work and relate to each other - but there's no way I can interpret a data log well enough to make decisions regarding what adjustments should be made. 

Pete Gossett (Forum Supporter) said:

For example, I've gained a reasonable understanding of how the systems, sensors, and data inputs work and relate to each other - but there's no way I can interpret a data log well enough to make decisions regarding what adjustments should be made. 

 

I sometimes think we should cover that in a live program sometime.  It seems like a mystery, but I think it can be simplified- especially when you look at it as "what do I want" and "what am I getting"- then you figure out how to make the two the same.  Looking at the data is just doing that.

It is easy to add features to an ECU, so the deciding factor is not what it can do, it is how well it can do it. Resolution (both temporal and digital) is huge. High end stuff will use floating point math and angle-synchonous calculations while cheaper hardware will use 8bit and fixed timing. I'd have to google what MS uses, so I'll let you do that ;) Then you've got the interface - if you or your tuner can't get it to do what you want, who cares if it can. Along those lines, an excellent ECU is an expensive paperweight if your tuner can't use it. I'd say start backwards - who is the best tuner in the area for the 2JZ, will he accept your project, and what ECU is he comfortable with. MS really seems like a great grassroots kind of option, but if he won't touch it then you'd be better off with a good tune on a less good ECU, to some reasonable extent. 

Also what Keith said above, tuning for WOT is great for YouTube views, but there's a whole lot more cells (and temperatures) to be calibrated than just that one curve. Cold start is particularly expensive because you have to wait for it to cool if you start building heat. That down time is pricey. If you want to drive well, plan a few days unless you can get a very good base map. This is where the ECU choice also comes in. If ECU #1 comes with no map or community support with one but ECU #5 does, that alone could save a ton of headache and cash. 

In reply to alfadriver (Forum Supporter) :

That's an awesome idea!

I'd flip this around. Set your goals, then select the hardware to get there. You say you want 350-420whp. That's a fair range, 20%, and they may cross a threshold. For example, the difference in effort and expense between 250 and 300 rwhp in a Miata is considerable even though it's only 20% more power. Did these numbers come from anywhere in particular?

What's your expectation with regards to reliability? How about driveability? How much time do you want to put into this, and how much do you want to learn?

Standalones are 100% required to get past certain power levels, and the quality of the tune is more important than the specification of the unit. But the quality of OE tuning is a big step up from aftermarket standalones, so if you can retain a stock ECU you'll be able to take advantage of that and that will depend on power level. Sometimes the interaction between an OE ECU and whatever magic boxes you have to add to make it do what you want to do can also cause driveability problems, usually in transitional areas. Depends on what your priorities are again.

I am that same step with a similar project, my NA-T 2JZ-GE powered hot rod. I can make a JDM TT ECU work under $1k with some major re-wiring. Or buy the $1k+ AEM that all the 2JZ folks recommend, and still have a lot of work to do, but tuners know that system well. Or spend under $500 on MS2/Microsquirt and DIY everything. Luckily I live in Atlanta, land of DIYautotune, who happen to have a 2JZ shop car, so I could probably get it tuned professionally. But maybe I'd like to learn how to tune on my own? Or will that cause further frustration? I just don't know.

If you live near DIY Autotune, the logical answer is to get them to set it up for you. Then, if you want to learn, you can continue to modify. It's not a one-way street, as long as you have a "known good" map you can always revert if something doesn't work.

In reply to alfadriver (Forum Supporter) :

Would love to see a live program focused on how to interpret logs or even how to tune timing. AFR seems to be much easier for most folks.

Contact Paul_vr6 on here... he builds, sells, and tunes mega squirt setups.

If you're on a tight budget, want a full-featured EMS and you're and not already a tuning expert, the only systems you should consider are some of the higher-end Megasquirts or the cheapest AEMs that could do the job. The Corolla in my avatar runs an MS3X. If you're on a budget, you probably don't want to work with the tuners who turn up their noses at Megasquirt systems.

Saron81 said:

Contact Paul_vr6 on here... he builds, sells, and tunes mega squirt setups.

I do, and am a happy distributor of all things DIY-Autotune as well. I do build harnesses and tune a few other systems as well and it's really hard to beat the cost/performance of all the MS3 setups (3/3x and Pro) for drive by cable cars. Once you want to add DBW or require (for whatever inane reason) onboard WB then there are others that can be better. The EMU is cheap because not all the features work the way anyone who has tuned things before should work (the boost control is awful, and I based on that don't trust the DBW) and the connectors are off brand copies of the "real" connectors. It's cheap though.

I'm with Keith on pick the features that you NEED, the features you WANT and then consider budget in there. 

cyow5 said:

It is easy to add features to an ECU, so the deciding factor is not what it can do, it is how well it can do it. Resolution (both temporal and digital) is huge. High end stuff will use floating point math and angle-synchonous calculations while cheaper hardware will use 8bit and fixed timing. I'd have to google what MS uses, so I'll let you do that ;) Then you've got the interface - if you or your tuner can't get it to do what you want, who cares if it can. Along those lines, an excellent ECU is an expensive paperweight if your tuner can't use it. I'd say start backwards - who is the best tuner in the area for the 2JZ, will he accept your project, and what ECU is he comfortable with. MS really seems like a great grassroots kind of option, but if he won't touch it then you'd be better off with a good tune on a less good ECU, to some reasonable extent. 

Also what Keith said above, tuning for WOT is great for YouTube views, but there's a whole lot more cells (and temperatures) to be calibrated than just that one curve. Cold start is particularly expensive because you have to wait for it to cool if you start building heat. That down time is pricey. If you want to drive well, plan a few days unless you can get a very good base map. This is where the ECU choice also comes in. If ECU #1 comes with no map or community support with one but ECU #5 does, that alone could save a ton of headache and cash. 

This is all true.  However, it's entirely feasible to have a pro dyno tuner do the map for WOT and high load conditions in a couple hours, and do the idle/cold start tuning yourself over a few weeks/months.  Yes, there's a learning curve and it's a PITA but it's also very hard to blow up a motor at idle or startup.  Idle and startup tuning takes many iterations, small tweaks and repeated testing under different conditions, there's no way I would pay someone else to do that.  If you can figure out how PID controllers work and you have enough time on your hands, you can get it right.

I will freely admit that I'm the one forum guy who has tuned one car that Keith describes.  However, it was an aftermarket turbo car with different injectors and a bunch of homebrew stuff, not a simple project.  I started with a generic base map and had it running well/safely on day one.  In another couple months I had idle and startup running better than OEM, and making good power with no combustion.  When I finally took it to the pro tuner, he barely changed anything.  He was comfortable pushing the limits with timing more than I was and he got me maybe 5% more WHP. 

Now if you're not a computer guy and don't have the time/patience to program your own car, I totally get it and I would recommend having someone else do it all.

Also, digital resolution is the defining difference between MS2 and MS3x.  Big leap forward in terms of processing power.

Haltech Elite. 

stop screwing around with megasquirt, its too technical if you just are interesting racing instead of engineering. 

fidelity101 (Forum Supporter) said:

Haltech Elite. 

 

stop screwing around with megasquirt, its too technical if you just are interesting racing instead of engineering. 

Unless you go through the totally DIY route, I don't think Haltech (or any of the other higher end standalone systems) are much easier, if at all. Most have their own caveats and tricks that are hard won (like getting non single pulse cam sensors working on Holley other than the LS 58x and other similar oddities!)

The MS3 Pro and Microsquirt options, bought through a reseller that can help you configure things, much like any other system. It's why we exist, and it seems to work.

Keith Tanner said:

Also, really consider if you want one. You will be responsible for all the hard parts of making a car run. Not full power, that's easy. Stuff like cold start and idle control and AC compensation. All the stuff that most amateur tuners just gloss over and accept will be crap.

Having done exactly 1 car, and fuel-only mind, my view is that stuff is the interesting part!  I daily-drove my MS'd 190 from about April-June with no significant issues, and after I got things "drivable" my fiddling frequency went from daily to weekly to every few weeks.  Eventually I got to the point that I was trying to work out things like the car dying at idle when stationary but turning the wheel.  Ah, massage the PID idle tuning, ok, cool that works now.  Having said that, it takes a different mindset to work out those problems.   

I'm dying to get back driving it, I did a bit of disassembly to clean up things once I knew the install was solid, then it got to 95* out and it was parked without any shade.  Sorta tempered my enthusiasm for working on it at the moment.  Finally supposed to cool down into the high 80's, so maybe I'll have it back next week.