When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
I have the injectors up in the throttle bodies. That tends to be a little problematic on throttle lift. But it was not causing any drivability issues. That screen shot was the first day of running pure Speed Density below 3500 RPM and pure Alpha-N above 3500. I would have most likely got around to dealing with that on the next tune but as it turned out, we are replacing that motor with a B18C ITB righ now.
For learning purposes, Megasquirt is one of the best EMS. Super complete hardware and software, a big community at msextra.com, and it works on almost every engine based thing.
I have access to a dyno that I can get on almost any time. We normally do initial tuning on it. It is also great for doing back to back testing like dialing in the AFR / Wide Band / Engine to what that motor wants for timing and AFR. Once we get that dialed in we head for the track for what is about the second half of the tuning process. You simply can not do the dynamic stuff on the dyno. All sorts of strange things show up on track. Things like temperature control under the hood at speed, fuel starvation at high gee loads, transitions like shifting or throttle control in a slide.
The trick is having at least 20 samples per sec but 50-60 hz is better. My Motec on my 300 HP jetski can go up to 1000 hz but that is a little over the top. And having software designed for looking at this sort of thing.
Looking in this view is normally fairly painful but is normally what most people are forced to deal with.
Now if you look are the same data sorted to a view similar to what the tuning software is dealing with, you get RPM along the bottom, MAP up the left and the AFR in the field shown as a color, you get this. On the right I am showing the same motor with RPM, TPS and AFR. Rich is blue and lean is red. The AFR scale is on the right.
It gets fairly apparent where this motor needs some touching up but for the most part, it is running really well. This motor is an ITB Honda D series running Megasquirt and data viewed with MegaLogViewer HD. Hondata data logs would look just like this.
Hope this helps with getting started with DIY tuning. I tune about 5 race motors with this method a year. Most of them I never see, let alone drive or even hear. It is all done with data logs. I will be happy to help anyone learn to do it. For me, tuning is the fun part.
Just for fun, here is my race motor doing a few throttle stabs. This total trace is only about 3.5 sec long. The motor is accelerating at 13,100 RPM per sec. Not bad for tuned at home.
Does Hondata tuning software allow Alpha-N (TPS and RPM) based tuning? Does it allow Speed Density below some MAP or throttle position and Alpha-N above these points?
The fun part of tuning comes when you can clearly tell what your motor is requiring. And then give it what it really needs.
i believe that anything DIY is great. there is something to be said when you can say "yeah i did that", but that also applies in the instance you blow something up. Just because someone has a shop and a dyno, doesnt mean they are professional tuners and it surely doesnt mean that they provide a more quality final result than you can provide yourself. i myself, have tuned and continue to tune my turbo b16 setup. it has been running and running strong for 3 years now. it has never been dyno'd. i found that by doing it myself i have learned alot, and i am proud of what i have. i do not trust local tuners as far as i could throw them. my friend has a supercharged 5th gen and is the type that unless its name brand and professionally tuned, its junk. needless to say, a poor tune by a local "reputable" tuner, did severe damage to his engine. and here i am, 3 almost 4 years later, still running strong (and my car beat his on several highway pulls lol). it is what you make of it. i am an extremely picky person, and to sit there and watch someone through a window make changes to my car, and knowing that if something goes wrong their excuse is "well you signed a waver, and you cant expect a warranty on highly modified cars". no thanks. i will say that dyno's are a great tool, and if i could rent dyno time from a shop, and i was the one tuning the car, i would for sure. but unfortunately i have not found anyone to do that yet.
for my particular setup i am still OBDO, running BRE with real time updating, datalogging, etc. there are for sure waaaaayyyy more better systems out there if i convert to obd1. however this was cost effective at the time and still continues to be reliable for me. i based everything off a OEM b16 map for starters. tuning AF is pretty self explanatory. as for ignition timing, my tactics were pretty archaic. i mearly took that last column of igintion values in the stock map, copied them, and pasted that column into every boosted ignition column. than i retarded timing .80 degrees per psi. over the years i have reduced that .80* and checked plugs continuously. i know this is not the ideal way, and certainly doesnt provide MAX power, but its reliable thus far.
for me , i have certain goals that i would like to meet on my new engine setup running obd0, after that i would like to upgrade to obd1. however the thing for me about the conversion, i need to buy a dizzy which has been discontinued through honda, and i dont trust aftermarket really. but a new one from napa is $350, than a conversion harness, than a tuning platform maybe 600 bucks, than a professional tune maybe 500 bucks minimum. you have $1500-1600 in obd1 conversion and a tune. now in the grand scheme of building a quality build, thats not awful especially when it could make or break your engine but it is a lot of money.
too many people hear DIY and automatically call it junk or unreliable. how did the big name tuners get their start? DIY tuning, chipping, burning , etc etc. i say do it yourself as far as you can. once you get to a point where you have met your goals or reached the limits of what you or your software is capable of, than maybe look into a good tuner. at least you can be proud of what you achieved. for me that is why i build and work on my car. because i enjoy it and i am proud to say i did it. i would have no pride in anything if someone else built my motor, put it in my car, tuned it for me, etc etc. that makes me just a driver. but to each his own. bottom line is you can tune yourself, safely and reliably, it just takes more time, more learning, reading, etc. good tuners will street tune a car after the dyno anyway just to polish everything up. so if they do it why cant you?
Here is an example of what I am looking for when I look at a data log as a scatter plot. Any area that is clearly red (Lean or around 15 AFR the way I have things scaled) would need fuel added in the VE table. Clearly blue areas would need the fuel removed. See the boxes that I highlighted in this shot.
FWIW, Megalog Viewer can also be used with Neptune or anything else that will allow you to spit out a .csv. There's so much value in scatters and other types of graphing that you just can't do natively in the stock-ecu based systems.
I was looking at some data coming off a Hondata logger. This is a fairly typical log. The first screen shot is the basic fuel tuning plot where RPM is on the bottom and MAP or Manifold Absolute Pressure is the vertical axis. AFR is in the field as a color. 12 AFR is Blue and red is 15 AFR.
The big red strip across the bottom is where the motor is in a downshift and the injectors are turned off. The actual AFR is way leaner than 15 but I have the MAX scale set to 15 so that area shows up as bright red. This gives me better resolution in colors where I really care.
Notice how his MAP falls as the RPM climbs. This is an indication that his throttle body may be a little small or that he has an air filter sizing issue. I will get into more detail as to what I can see later.
Notice how his MAP falls as the RPM climbs. This is an indication that his throttle body may be a little small or that he has an air filter sizing issue. I will get into more detail as to what I can see later.
Have fun learning to tune.
Andy
Interesting. So ideally, should the MAP line stay straight or slope up as RPMs increase? Also, could the decreasing MAP line indicate the manifold is not optimized for high RPM use? Thanks for your insight and I really appreciate you chiming in this thread.
I guess is depends a little where the MAP sensor is located. Normally they are in the the big "Log" section of the intake. Restriction in the TB, CAI or filter can all show up.
I happen to run ITBs on my race car. Here is a pic of my old motor. It was running 42 mm throttle bodies off a motorcycle.
Regarding the slope, I like to see about a 3 KPA drop or less at full power. It takes fairly large throttles to get there.
Here is data coming off my Yamaha FZ1 liter bike running 52 mm Throttle bodiess and running a Dynojet data logger. You can see the MAP drop on that motor also, but the revs are way up there.
Here is data off my CRX racecar. It does have MAP issues but it also had packaging issues if I made an air box, throttle bodies and filter that was really big enough.
As a added note, I often see extremely high intake air temps in data logs, huge voltage swings from the alternator, fuel starvation and all sorts of other things. You simply have to be willing to look at the data. Lots of this stuff is almost impossible to find on a dyno.
Thanks for the info. I like the ITB setup. Are you running a MAP sensor on your setup or is it strictly TPS vs. RPM (alpha-N)? Also, why do your graphs differ from each other? The axes are switched and the colors are supposed to represent AFR but your last graph is using some other numbers.
The software I use allows me to flip and scale things any way I want.
Lately, I have been running speed density below 4500 rpm and Alpha N above 4500. I could go alpha N above 20 pct throttle or above 80 kpa. The SW is real adajustable. Later, we can get into all of this stuff if you want.
Here is a plot coming off a fairly hyper centrifugal blower motor operating over the entire possible operating range of the motor.
The Black circle is where the motor idles at about 50 KPA and about 1600 RPM
The Blue box is the typical operating range. This is the area that the majority of time as seen by the number of dots.
The Red line is where the motor operates as the motor pulls up against a dyno
The Green circle is the motor at full power at max RPM of about 8300 RPM and about 180 KPA or about 12 lbs of boost.
The Purple area is normally only seen in a throttle stab from low RPM. This is where acceleration enrichment has a huge effect and wide band sensors a questionable at best.
The Yellow area and everywhere near the yellow area is only seen in a down shift.
NA motors will not see any of these dots above 100 KPA
Hope this helps (a little) understanding what this stuff is showing.
Awesome. Appreciate the feedback. I just looked into Megalog Viewer and I see that it can be used for multiple dataloggers provided it's in .csv format. This is good news since the datalogger I'm currently using for Lancer Evolutions saves data in a csv format. Would you happen to know if there is software out there that is similar to Megalog Viewer? I just want to know my options.
MegaLogViewer is about the best thing going at our level. I wish it would take raw NepTune logs and map exports so I wasn't limited to the parameters in the .csv's, but you choose your battles here I guess.
08-12-2016, 06:41 AM
Re: DIY tuning, a discussion
