3FE MAF to AFM conversion

[Moby]

3FE AFM to MAF conversion

Some time back I posted a thread about my in car PC, LC1 wideband O2 and upcoming AFM to MAF conversion (https://forum.ih8mud.com/60-series-wagons/421776-lc-1-wideband-o2-pc-install.html). I'm using a Split Second PSC1-005 with the inverted 5-0v signal to match the stock AFM signal (http://www.splitsec.com/technotes/AFM_to_MAF_Conversion.pdf). I finally got some more time to work on this the past couple of weekends so I'm starting a thread.

Progress may be slow but the truck starts, idles and free revs light throttle up to 2000rpm. I tuned without the stock O2s connected, aiming for 12.5-13.5 (better to run a bit rich and have the ECU use the O2 feedback to trim the fuel back than to run lean and have the ECU have to add fuel). I then plugged the stock O2s back in and let the ECU learn a little. Wideband indicates that with very little learning the ECU has trimmed nicely and is running well. I haven't driven it yet (see next) but so far this is going well. I've already managed to smooth out a rich spot with the stock AFM where for some reason I'd run 12-13:1 (rich) between 1100-1400rpm. Immediately above and below this range was always in the 14:1 range.

I'm having a couple of problems that I still need to work out. First, I've got a 31 code. The ECU doesn't like something with the VS or VC signal (AFM signal and power respectively), but it is probably a signal out of range issue, maybe at initial start up. It doesn't seem to impact anything, but I'll be chasing it down nonetheless. The second problem is that the PSC1-005 has an aux driver to energize the fuel pump. The stock AFM will only allow the fuel pump to run if air is actually flow (this is a safety measure). My PSC1-005's driver doesn't appear to be working (I've confirmed that the stock +12 that I'm feeding it from the stock AFM wiring is under it's 200mA limit - it's ~100mA). I'll need to call Split Second on Tuesday to see what is going on. I've simply jumped this for the time being.

Otherwise this is going very well so far. The goal is a little better mid range and higher power (2000rpm+) and to keep the ECU from going into open loop (and overly rich) as easily at 2500+rpms.

Not very exciting, but here are some pics:

 

[Moby]

. . .

 

[62Cruiser]

I am excited to see how well this works. The stock AFM can be a pain and expensive if you have to change it. I hope this is a better option! Subscribed...

 

[Spook50]

This will be interesting to follow. I'd like to install a PC interface so I can use my laptop to monitor my truck, and maybe at some point install one of those double DIN PCs that've been mentioned in the Electronic Toys section.

Is that a plastic bag you have wrapped around your air filter? I'm curious if you can use a setup like that in conjunction with a snorkel. I do NOT want to use a gauze/oil based filter (like a K&N) instead of an OEM Toyota filter, but it would be nice to have a way to flow air more freely into the engine than the stock 62 airbox allows for.

I plan on copying what you did with your ignition once I have the parts and ditch my Mallory setup. Be great to see how my rig does with your ignition, a MAF and the H55 on long road trips and even in day to day driving.

 

[Moby]

My PC monitor is in a double din mount in the stock stereo location. If you do this pay the extra for the anti-glare coatings that are usually offered. The stock stereo location has a lot of glare. The serial cable in the pic above coming out of the Split Second box is running to the PC.

The bag is a pre-filter. Just makes it easier to knock the dust off. The filter is an Amsoil. No oiling required. On the list of things to do is eventually bend up a dam to isolate the filter from the hot air coming off the engine and radiator. Just haven't gotten to it yet.

I drove it around the neighborhood this morning and did some more tuning. Very light throttle and low rpms. Lots more tuning to do but it is coming along pretty nicely....

 

[Moby]

Worked up to half throttle or so and up to highway speeds. It seems smoother and pulls stronger but the tuning isn't good enough to really lay on it yet. I'll probably spend another week or so tuning myself and then get it on the dyno with a tuner that I've worked with before. Then we'll see how far off I got it

 

[Spook50]

Worked up to half throttle or so and up to highway speeds. It seems smoother and pulls stronger but the tuning isn't good enough to really lay on it yet. I'll probably spend another week or so tuning myself and then get it on the dyno with a tuner that I've worked with before. Then we'll see how far off I got it

Nice. Sounds like it's really coming together. I wonder if it'll be best to dial it in for each specific vehicle it goes on, or if once you find your ideal settings those will work for all/most 3FEs.

 

[Moby]

I don't have enough experience with this to say how my eventual calibration might work on a stock or close to stock engine. It might be a starting point. You would still need a wideband O2, I can't imagine doing this without one (obviously folks did, but a fast wideband gives a lot of confidence and makes the tuning go pretty quick).

 

[FJ40Jim]

Good work Moby.
Questions: Why even keep the stock ECU at this point?
Wouldn't it be easier now to just change to an aftermarket control system that accepts the WBO2 and MAF inputs?
Or is there an issue that won't allow stock toyota bits to play nice with the DIY EFI kinda stuff?

 

[Moby]

Good work Moby.
Questions: Why even keep the stock ECU at this point?
Wouldn't it be easier now to just change to an aftermarket control system that accepts the WBO2 and MAF inputs?
Or is there an issue that won't allow stock toyota bits to play nice with the DIY EFI kinda stuff?

A standalone would give complete control but you'd either need custom hardware (see below for some options) to interface with the stock wiring or you'd have to build a wiring harness from a standalone box to the stock EFI inputs/outputs (I'm picky about wiring, the most important aspect of reliable EFI, and a scratch built harness using Toyota plugs would need a crimp tool). Once you have the hardware side you'd then have to build all the calibration maps from scratch (timing, fuel...) which is a lot more work.

This is all possible but more work than I need (and more than I want to spend, both in time and dollars). I just want to reduce the intake restriction and keep it from going into open loop at part throttle highway speeds. With this approach all I'm having to do is calibrate the MAF signal to match the stock AFM signal and trim it back some at the rich points.

For those interested in a fully programmable system for the 3FE wayneosan in the 80's section built a megasquirt unit with stock wiring plugs. That would be the cheapest way to get full standalone. Also the guy in Oz that built the turbo 2FE used a standalone system (but I think he changed out some of the stock sensors and the stock injectors, which might avoid having to build a standalone wiring harness).

 

[Spook50]

This is all possible but more work than I need (and more than I want to spend, both in time and dollars). I just want to reduce the intake restriction and keep it from going into open loop at part throttle highway speeds. With this approach all I'm having to do is calibrate the MAF signal to match the stock AFM signal and trim it back some at the rich points.

Is all the tuning and calibration done on a software level?

Is the WBO2 needed for the MAF to operate, or is it just an invaluable tuning aide?

 

[Moby]

Is all the tuning and calibration done on a software level?

Is the WBO2 needed for the MAF to operate, or is it just an invaluable tuning aide?

All the tuning and calibration is in software. The wideband is just a tuning aid. The way that I have it is that I ordered the PSC1-005 with the datalogging addition. This gives you 4 0-5 volt inputs that will show up on the calibration and data logging software. I'm using 3 of the inputs currently:

1. Innovate wideband O2
2. J&S Safeguard knock sensor output
3. AFM signal (this is the output from the PSC1; because the output is inverted from 0-5 to 5-0 after the calibration values are applied seeing the actual output just makes it a little easier to think through the process).

So my wideband is actually connected to the calibration software. This isn't necessary though, you could use a standalone wideband or even tap your stock narrow O2 and run that to the calibration software. Not nearly as effective but reasonable.

Split Second has pretty good docs on their site that cover the calibration software (named R4) and the AFM to MAF conversion (including wiring diagrams): Tech Notes Their tech help has also been very good.

 

[billmc]

Very interesting thread! I'm surprised that it goes into open loop at 2500+. Is that normal? I would've expected it only if the TPS was sending WOT, but then I'm no expert.

Bill

 

[Moby]

Very interesting thread! I'm surprised that it goes into open loop at 2500+. Is that normal? I would've expected it only if the TPS was sending WOT, but then I'm no expert.

Bill

From what i've observed it is not based on WOT only, the AFM signal is a factor. But without more definitive reverse engineering, this is obviously subject to error.

 

[Moby]

Fixed the fuel pump driver issue. It needed to be enabled in the Output settings. I have it set to turn on over 1 rpm or over .3 AFM input. So this preserves the safety feature of the original system.

 

[torfab]

Joel, have you ever thought of building a megasquirt system for your 2fe?

 

[Moby]

Yeah, in fact wayneosan (mostly in the 80s section where he is just finishing a 5.3 swap for his 3FE) actually built a megasquirt board and found a 3FE wiring plug - so a plug and play megasquirt system for 3FE. He offered to sell it to me and do a base tune. The advantages of this would be:

1. Ability to use wideband O2s
2. Ability to do sequential injection (perhaps a bit more mpg and lower emissions over the stock batch fire).
3. Full control over timing
4. I think knock sensor feedback

These just didn't make it worth the extra work and expense for me (I already have a J&S Safeguard for #4). And I already had the Split Second MAF system. My next major powertrain iteration will either be a small turbo or a 5.3 swap.

 

[torfab]

a small turbo - or - a 5.3 swap Those are so similar, how will you ever choose?

 

[Moby]

I still have to get that driveshaft shortened first Frankly, if I had to chose right now I'd do an LSx (some version). Why? Mostly the state of my stock wiring harness. My engine has been in and out multiple times. Connectors are wearing out, insulation is getting brittle... But since you did the H55 for me and I got the body rust fixed if the MAF swap is successful (if I end up achieving better control over the transition to open loop) I'm going to try to leave it alone and just drive it for awhile.

 

[Moby]

Got rid of the error code (well, hopefully). I think that the FSM is really serious when it says that it expects 4-5 volts on VS with ignition on/engine off. I was at 3.98. We'll see if it comes back or if that was it.

On the tuning front things are coming along. Higher load tuning has been very interesting. High load is very sensitive, so .73 vs. .8 makes a big difference. I have managed to keep it in closed loop longer than it would have been with the AFM (and longer than I want), and I have managed to get it really rich (9s) just via the air flow signal.