Air cooled motorcycle CB900f WUE and CLT question

[NickC]

Hi there,

Tuning underway on my CB900F. Seems to be going well, but had a questions for anyone who has microsquirted an air cooled motorcycle.

What are the settings for CLT temp for the WUE? Since it's air cooled and it is still a little cold around my place, the range of CLT temperature while riding, and depending if cold out or not, it can go from about 35 degC to 90 degC. Obviously, when I leave the WUE at the default 71 deg C, it wants to kick-in when temps go below that. I WUE go to 100% at 45 deg C, but after riding in a bit colder weather, I have seen it drop a bit below that. (unless my thermistor is pooched???)

Anyways, just want to see what kind of set-up some might have that are running an air cooled beast.

Cheers
Nick

[billr]

How about sensing oil temp for CLT?

[slingy88]

Where have you got the sensor mounted?
I have heard of similar things when mounting the sensor in the sump to measure oil temp.
I mounted mine in an oil gallery plug in the head, so you get a mix of oil temp and head opperating temp (more indicative of combustion conditions).
This seems to work pretty well for me, giving a realistic warm up time and steady riding temps, but it dont get really cool here (OZ) so cant comment too much on the low temp running.
cheers slingy

[R100RT]

I concure.
I had my CLT running off a stock GM sensor (and calibration values stock) in a fitting that saw oil temperature "pre turbo". In fact with the oil cooler set up I have those temps weren't climbing nearly quick enough in cold weather for the engines liking (on my set up anyways). I fashioned a small heat sink fitting that mounts behind the right side cylinder (boxer BMW) about halfway between base and cylinder head in the air stream from the fins. This reacted much better, but I ended up playing with the calibration values (top two of the three in the range) to fool the "max temp" that would show on TS (and affect WUE etc.) from those the sensor was actually able to witness in that location (rather unproductive when an air temp reading of 230-250 developing and then skewing other settings etc.)
I really appreciate the ability to log engine operation during warm up as well rides, as you can and will suss out what is going on and be in position to apply values that work for you. (Hard to provide recomendation for your numbers hence, as most engines being vastly different even in subtle ways).
Regards,
Lorne.

[NickC]

Thanks for the replies Guys!

I had tried the oil temp approach and the issue was that the oil temp was not proportional to what the engine wanted, and so WUE was not comming off when needed and then if bike was parked for a bit, it would not come on when it was needed. That's when decided to do with Head temp.

Sensor is mounted on a fin in head above intake track for cyl #2. The readings I get seem to reflect pretty well the actual cyl head temp, the issue I guess is the settings in the WUE wizzard in TS. Default setting make the WUE taper off all the way to 71 deg C, but for an air cooled application like mine, seems too high. I now have it taper off at about 45, but might bring it a bit lower. Last time I was out with the bike, it was about 8 deg C outside, so at higher speeds with low engine loads, the CLT temps went down below the 45 in the WUE settings.

I will keep tuning and playing with it to see if I can get something that I can live with. That and the temp out here will only be getting warmer as we get closer to summer...

Thanks for th inputs, just wanted to see what other air cooled motorcycle guys had for set-ups.

Cheers

[wmax351]

You could look at a spark plug mounted CHT sensor.

https://www.denniskirk.com/cylinder-hea ... 408514.sku

The nature of air cooled engines is they have a wide operating temperature range. Something like the later Airhead or any oilhead BMW's or certain other bikes will have a thermostatically controlled oil cooler, which tightens the range down a bit.

You could consider running an "electronic choke" that uses table switching to give an artificially rich mixture for warmup, etc, then turn it off after a certain amount of time, so you don't have unwanted enrichment at lower head temperatures.

[NickC]

ya, the manual approach is one that I was thinking of too. I will probably be looking at that as an option, since my handlebar controls are off a GSXR with the cable choke lever on it, would be a good place to set it up.

Thanks for the input!

[NickC]

Good day all,

So after an extremely busy summer and not being home most of the time, I am finally getting back to tuning my bike,

I had questions, I guess more on tuning approach. Hopping some guys that there are guys running MS on a 4 cyl bike that could chime in with their experience.

My bike is running GSXR ITB's and I am currently running it in SD mode. My MAP at idle stays around 70kpa. I have a wideband on it and I am datalogging with Shadow Logger and my Android Smart phone. I have it set-up so I have no AE or DE and no EGO correction so that I can tune the VE table. Seems to work well, I am able to plot in MLV the target AFR and the measured AFR. From this, I am able to tune the VE values in the table to get them as close as possible. I am assuming my approach to this is correct? Once VE is tuned, then I will tune the AE and DE.

Thing is, I seem to run my bike most of the time in the upper load range, MAP around 80kpa, but the default AFR table puts me in an AFR range in the low 12. Now I am correcting my VE to get my measured AFR to match the targer AFR, but then I'm guessing I will need to bring the AFR values in the areas I seem to ride the most to something more in the 13.2 to 13.5? (assuming the bike is happy in that area!)

The other thing, I noticed the spark table does not go even close to the full advance that this bike use to see when it was stock (38.5deg @ 3200 rpm). Now the table I have is generic, and need to be adjusted, bike runs ok, but I think I need to get closer to the 38.5 deg past 3200rpm in the load range I am running. Any comments?

And lastly, based on all this info, would it be wise for me to use the ITB mode which blends SD and Alpha-N ? Or is this doable only in SD?


Here is my MSQ.

Tune before Aug18 Am Run[1].msq

Tune before Aug18 Am Run[1].msq

I am having problems loading my datalog, I will try on another post.

Cheers
Nick

[NickC]

Today's datalog is too big in size to upload, so if needed, I will get another log that is shorter.

Nick

[R100RT]

Nick, I'm 2 cylinders shy of the tuning advice you've requested, but should be able to add a couple of comments.
I have found that road tuning (I take a lap top along on the bike ) requires that you really "mix it up" on the routes you choose. I had frequented an area close to home that had very little traffic & lots of straight sections but in the grand scheme of things then learned how the seemingly proper tune would react differently when all the other road conditions might be met on trips and such. Hence, the map value you seem to be residing in may be a good one to watch, but other lighter cruising and slightly decelerating conditions could require much different settings. I moved away from any stock maps for all the usual suspects like VE/ AFR target, and especially Ignition which I continue to mess with as I get more confident with high boost states and rapid roll ons. Wall wetting seem to be an ever present factor when more than casual transitions are developed and I haven't solved but the ultra rapid afr change as a high load gear change is broached can be mitigated by a very brief "pause" in throttle state at the top rpm. Not scientific but equally transmissions fail more easily when rowed too aggressively on some of the older iron (or aluminum if you will).
Guess the bottom line is that without dyno time the job is doable but requires a lot of diligence and careful road time. Currently, looking towards a 10hZ gps and applying some horse power logging on latest release of TS to finish the final aspects of my msq.
Lorne

[NickC]

Thanks for the advice.

Yesterday, I set-up an ignition table that resembles the original ignition curve. Basically 10 deg at idle and 38.5 deg above 3200rpm. That and I had debris in my fuel filter again, which I think was preventing a steady fuel pressure. Cleand the filter and took it today and drove miles better than yesterday. Even with AE disabled.

So I think for now, I will keep trucking in SD mode, datalogging and using MLV and VE Analyzer to keep dialing in my VE table. Once I get that tuned in, I will start looking into AE.

Are you running your bike in SD mode?

Nick

[ewflyer]

Nick,

Another Microsquirted Motorcycle guy here...

I've always run my first project bike, an EX-250 (Kawasaki 250cc water-cooled parallel twin) in Alpha N mode. The small cylinders of this high RPM engine just didn't offer enough kPa resolution for SD so I had to choose Alpha-N.

I consider my EX-250 a success because after quite a bit of tuning, a lot of datalogging, VE Analyzer finished off with some careful hand-entered final tweaking, the bike runs great from cold start and all up and down the rev range. I've been pretty satisfied with Alpha-N on this bike.

So, my opinion is that Alpha-N can be a good solution on a small high-revving engine.

But I wouldn't still be here on the MSExtra forum if I was completely finished striving for the perfect Do-It-Yourself motorcycle fuel injection solution. So I'm here gathering information and working on a new project, a Suzuki GSF400, 400cc water-cooled inline four. My hope is that I can manage to make this bike run well in SD. To reach this goal I've turned to Don McPhail (dontz125) with his QuadraMap product (multiple MAP sensors with a "lowest-only" discriminated signal delivered to the Mircrosquirt) and in addition to that improvement I'm not ruling out the possibility of having to use some of MSExtra's MAP-sampling ability in order to get enough kPa resolution for the bike to run well in SD.

I haven't gotten the bike up and running yet, I'm still working on rebuilding/reassembling it since it was in pretty rough shape when I bought it.

[grom_e30]

iv got a 1999 gsxr 600 that I started with alpha-n and it was good but i found at low throttle riding through town and traffic it to be a bit inconsistent so I decided to switch over and give itb a go my bike idles at about 80 kpa warmed up even higher when cold, i started out on my alpha-n tune and went for a long ride so I could data log to find the 'itb load tps switch points' eg what % throttle got the bike to the baro switch point of 90kpa easy to find if you use mlv and scatter plots. I left the 'itb load at switchpoint' all at 50% so its equally split, half map and half tps effectively and did a few drives with auto tune running and I found there to be a big difference drives so much smoother now, night and day difference for me. 80 kpa for me at idle works out to be about 40% itb load. iv left my spark table and afr target table on alpha-n but will be trying those on itb when i get some free time to give it a go as the bike is my daily ride atm.

[R100RT]

Sorry Nick, had intended to comment on SD versus Alpha N, and things got busy (at work )
I have always gone with speed density, my original reason being an inducted engine. However, I found some of the proposed reasons "Not" to use it haven't really been that challenging.
For instance, the potential for wild map readings from wildly fluctuating intake readings at lower speeds. For my engine this should have been pretty well worst case being a 2 cylinder boxer - however I have taken a small feed line off each runner and combined through an orifice @ 0.010" roughly, into a C02 cartridge (others describe using a small fuel filter but that just didn't do it for me) that is ported on the end to feed a) map sensor, b) boost gauge, c) water injection hobbs switch, & d) fuel pressure reference. An insane blend of electronic and mechanical gadjetry .
I have no noise or resonance to contend with and enjoy ultra calm map readings etc. Nothing earth shattering here but it does work, and function with no lag or impediment to rapid but calm response.
I'm inclined at some point to experiment with blended Alpha N low state and SD higher up, but that can be for another day when I've run out of MS based challenge and intrigue on other aspects of the all important "perfect tune" that I feel still needs a bit more in areas that become more subtle and mostly un noticable to all but the highly discerning squirter lol.
Its interesting to hear from other bikers and what's working, MS needs more motorbikes
And I also use my bike for the daily commute, and try for a couple of long runs each summer to rallies and such. How can one leave it parked in the garage after pouring so much work into her is my philosophy.
Lorne

[NickC]

I agree with you about running the bike. I have two young kids that go to daycare, I need to fab baby seats for my bike to use it more!!! lol

So I tried the ITB mode, and it made a fairly big difference. The higher RPM's are allot more usable. It improved driveability quite a bit.

R100T, what is your fuel set-up? where did you mount the fuel pump? I still seem to be having heat soak issues on hot days with my pump. It gets noisy and if I lwt it cool off, it's fine. Problem is, when it gets noisy, obvioulsy this affects my fuel pressure, and then .... Well you know.

Nick

[R100RT]

I agree with you about running the bike. I have two young kids that go to daycare, I need to fab baby seats for my bike to use it more!!! lol

So I tried the ITB mode, and it made a fairly big difference. The higher RPM's are allot more usable. It improved driveability quite a bit.

R100T, what is your fuel set-up? where did you mount the fuel pump? I still seem to be having heat soak issues on hot days with my pump. It gets noisy and if I lwt it cool off, it's fine. Problem is, when it gets noisy, obvioulsy this affects my fuel pressure, and then .... Well you know.

Nick

I wanted to keep my "antique" 83 BMW format as close to standard as possible (hold on, that is a farce , adding fuel injection and turbo to a fared R100RT can fool some of the people some of the time, but ---) and kept the twin fuel tank ports by fabricating a combination "feed/ return" fitting that allows my fuel returned via a 4" standpipe that ensures fuel blending and potential heat soak won't be a factor on the fuel until you're dangerously close to running out anyways. The pump I utilized is a Bosch roller style from about 1990 vintage VW rabbit application, its mounted into a small reservoir that ensures there is never any aeration, or "slosh" interruption. Hence, that is mounted adjacent to the rear wheel on the frame kind of tucked into a spot that looks pretty normal, has the twin feed lines combine into a single bottom port inlet. The return from the injector "manifold" plumbing hits a small pressure regulator that promotes removal of any air that might develop and splits the return again so that the fuel tank flanks (kind of fit down on each side of the frame on these old girls) will keep fuel balanced inside the tank.
I have never felt any heat developing on the tank during trips or hot/ slow operation so I assume the system is happy. That pump probably doesn't have too huge a flow rate which also helps.
These boxer engines do generate heat, especially with forced induction and the aftercooler and oilcooler mounted at the front, but the cylinders poking out the sides is a definite advantage.
Sounds like you'll get it figured out, if keeping the fuel, or pump cool proves to be challenging you might want to add a smallish "air to liquid" core to pull heat back out that way?
Lorne

[NickC]

So I re-located my fuel pump, and works allot better. Also with the ITB mode, runs allot better too.

I had a question concerning VE Analyzer in MLV and ITB mode, It analyzes and modifies the VE table just as it does in SD mode, right? I think my datalogs don't seem to tip into Alpha-N mode that often, so I seem to be getting better and better in the SD portion of the VE table, but the upper Alpha-N mode, the analyzer is not changing much.

[grom_e30]

veal tunes the whole table sd and alpha-n parts. not sure about mlv though.

[NickC]

So I have been riding my bike quite a bit lately trying to get it tuned in, and I got to say, the thing is running better and better every day. The ITB mode really woke it up.

I was wondering if anyone who has an ITB set-up running the ITB mode could share what their VE table looks like in 3D mode?

The analyzer is adjusting both the SD and AN side of things, but it seems that the SD mode is demanding more fuel than AN?

Any comments?

Here is my current tune, that is after the analyzer has done it's thing with my datalog and I have not altered anything to smooth it out. And my AFR table probably needs a bit a tweeking. for the purpose of tuning my VE table, should I just put my AFR table at 13.5 everywhere I run?

Sept 3rd tune.msq

Nick

[grom_e30]

here is my current tune for you to have a look at its all been done on the road no dyno, the shape of the table is kinda close bar the middle hump you have, but i would say you need to re scale the load axis as you have 4 rows high as sd and you alpha-n is spaced out some between 99-100% load how much change can there be that row could be used elsewhere