Page 59 of 89 FirstFirst ... 949575859606169 ... LastLast
Results 871 to 885 of 1326

Thread: #6

  1. #871
    Join Date
    12th February 2004 - 10:29
    Bike
    bucket FZR/MB100
    Location
    Henderson, Waitakere
    Posts
    4,196
    I've set timing at anything over 8000rpm(ish) to 44BTDC. On John's dyno, just using the inertia of the roller to work against and with the wideband Lambda installed it was pretty easy to see it start to lean off at certain revs and load and simply add a bit of fuel and redo run. After half an hour, and just before I killed me and my son with carbon monoxide poisoning, it was running and sounding quite good. It's pretty crisp up to 15,000rpm which was as far as I took it. Blips cleanly, idles sweet, takes any throttle at any revs up to about 50% or a bit more. I picked up my rolling road tonight from John's garage and have the bike sitting on it downstairs.
    The neighbours are going to be real happy.
    The Link ECU is very stable so far and adjustments seem very precise. So far the switch from Ecotrons seems to have been a very good idea. I'll probably play a bit more making manual adjustments and then feed the Lambda output to the ECU and get it autotuning.
    I'm going to have to sort out an alternator for it as the current draw means a single LiPo blade only lasts about 5 minutes. i use a big 12V car battery beside the dyno to run it at the moment.

  2. #872
    Join Date
    13th June 2010 - 17:47
    Bike
    Exercycle
    Location
    Out in the cold
    Posts
    5,626
    Is it showing any boost ?

    I had a look this morning and I don't have any 250 stators here sorry.

  3. #873
    Join Date
    12th February 2004 - 10:29
    Bike
    bucket FZR/MB100
    Location
    Henderson, Waitakere
    Posts
    4,196
    I've got a rotor & stator but aren't using it. I used a spare rotor boss and mounted the crank trigger disc on it. Now I need a small rotor I can mount on the outside of the trigger disc that won't fall apart at 19,000 rpm, and associated stator.

  4. #874
    Join Date
    12th February 2004 - 10:29
    Bike
    bucket FZR/MB100
    Location
    Henderson, Waitakere
    Posts
    4,196
    Got the Ecotrons Accurate Lambda Meter(ALM) hooked up to the Link now. Anaolgue input 3(AN3) was setup as Lambda input and that all worked but I had a little trouble figuring the best options to get the ALM and the Link to see the same Lambda value. In the end it turned out to be real easy.

    Setup the ALM to output .5V at Lambda = .7, and 5V at Lambda = 1.3. By default the higher value is larger but I'm not interested in the leaner mixtures.

    Then I chose Calibration table 2 to define what voltage input on AN3 equalled what Lambda. I set the Lambda values to start at .7 and step up in .04 steps. This made the last value Lambda = 1.3 so a perfect match with the ALM. It seems by default the first voltage is .5V and the last is 5V with equally spaced .3V increments.

    The ALM has a very useful test mode where you can enter a value of Lambda and it outputs the voltage determined by the setup table. I tested it at every value in the Link ECU table and it exactly matched so now the Link software sees exactly the same Lambda as the ALM.

    Next is the Quicktune feature where the Lambda measured is matched to a table of desired Lambda values. The software can be setup to make adjustments automatically. There's a manual option as well. It's all a guess at the moment what value to go for as I don't know the optimal value for each cell in the load table. Lambda = .85 should be a good start point

    Actually I'm not even sure the load table is going to remain as MAP versus RPM. It seems a better option might be to make the load table TPS versus RPM and have the desired Lambda table setup MAP versus RPM and making corrections on the fly. I haven't got close to getting my head around that one yet.

  5. #875
    Join Date
    12th February 2004 - 10:29
    Bike
    bucket FZR/MB100
    Location
    Henderson, Waitakere
    Posts
    4,196
    I've tried a few different techniques to try and get some tuning working.
    It definitely auto tunes OK but the time required in a load cell and needing to be real close to centre makes it pretty difficult. It would be a lot easier with a real dyno where you could lock in rpm and use the throttle to vary the load.
    What seems to be working with my setup is I have put "15" in every cell in the fuel map, still using MAP over RPM. Then manually tuned each cell or small range of cells around where it hovers, based on the AFR ratio. It works well enough and you can easily see where you have been because it isn't "15" any more. I haven't done too much because it is in my garage with the roller door up and the exhaust pointing out the door. The other good thing is that you can see how the map is developing and easily enter approximate values in the next load cells to be tuned. This is all being done on my rolling road but would work on an inertia dyno as well. As the revs rise, loaded against the roller inertia, you can see the AFR vary and make an adjustment and then do it again. Might have to have a chat with Rob & Cully

  6. #876
    Join Date
    18th May 2007 - 20:23
    Bike
    RG50 and 76 Suzuki GP125 Buckets
    Location
    Auckland
    Posts
    10,473
    Sorry about the odd angle of the photos but some how the phone and KiwiBiker clash over the orientation of them.

    Click image for larger version. 

Name:	#6 (1).jpg 
Views:	62 
Size:	715.1 KB 
ID:	335930Click image for larger version. 

Name:	#6 (2).jpg 
Views:	65 
Size:	801.0 KB 
ID:	335931Click image for larger version. 

Name:	#6 (3).jpg 
Views:	61 
Size:	637.4 KB 
ID:	335932Click image for larger version. 

Name:	#6 (4).jpg 
Views:	63 
Size:	767.2 KB 
ID:	335933

    Finally got the dyno up and running again and spent an enjoyable afternoon helping Mike with #6. A very interesting efi turbo Bucket project. Incredibly quite bike, the turbo seems to make a great muffler.

  7. #877
    Join Date
    12th February 2004 - 10:29
    Bike
    bucket FZR/MB100
    Location
    Henderson, Waitakere
    Posts
    4,196
    I haven't been making much noise with the bike. We just ran it in the garage for 15 minutes to warm it up and pinpoint the leak. Turns out it is from somewhere around the end of the intake cam where it protrudes through an oil seal. It has an alloy boss pressed into the end and it's either from around there or, hoping, from the 6mm cap screw in the end. The boss and cam are both ground concentric and 24mm so if needed I will press a speedi sleeve over the join with a little sealant on it. Might have to relocate the seal 2mm further out to make it work. I'll run it again tomorrow to see if I've fixed the leak or whether I need the sleeve.
    On the fuel front I've changed the fuel Equation mode from MAP to BAP/MAP crossover. So now the main fuel table has Throttle Position(TP) versus RPM. The ECU uses TP to determine fuel requirement at an RPM up to where boost starts, or where MAP exceeds Barometric Air Pressure(BAP). Once MAP exceeds BAP it uses what they call a 4D fuel table to adjust the fuel as determined by the main fuel table.
    The main fuel table should be able to have quite a few load cells plotted without boost occurring. For those cells where it always pulls boost such as high revs and large throttle it will be a matter of coming up with a number that looks right in the main fuel table. Lets say 80% throttle and 14,000rpm always ends up with boost and we've determined that 50 is a reasonable number in that cell in the main fuel map. In the 4D table at 14,000rpm, if we have say 1.2BAR boost, the relevant cell may have 10 in it. This is a percentage of whats in the main fuel map cell which in this case is 50. 10% of 50 = 5 so the fuel injected is 55.
    If the throttle is opened to 100%, still at 14,000rpm, the main fuel map might have 60 in that cell. If the boost remained at 1.2BAR, in the 4D table the same cell would still be in use - 14,000RPM/1.2BAR, so 60 would have 10% added to it making the fuel injected 66.
    It's actually pretty cool what it does. It does altitude correction with the BAP reading, uses the throttle position and revs and BAP to calculate fuel while MAP is below BAP, and then when MAP exceeds BAP uses the 4D table to add percentages to whatever has been calculated.
    Just got to figure out a technique to tune it.
    One very cool feature of the system is what they call the Mixture Map. Where it is difficult to use Quick tune you can use Mixture Map. You turn on logging either to the ECU or PC if connected and record at least RPM, Load, and Lambda. If you hold the engine at a particular load and rpm for a short time it records in each cell the Desired Lambda, Actual Lambda, number of samples, and the error. If you click in a cell in the Mixture Map the same cell is also highlighted in the main fuel map. You can right click and update and it calculates the correct value for the main fuel map and writes it in. It's all colour coded dependent on how far out it is or whether it is spot on. More noise tomorrow.

  8. #878
    Join Date
    20th January 2010 - 14:41
    Bike
    husaberg
    Location
    The Wild Wild West
    Posts
    11,823
    Click image for larger version. 

Name:	8332185bc56295bfb9208a8db7705a83 (1).jpg 
Views:	48 
Size:	46.5 KB 
ID:	337021Click image for larger version. 

Name:	250 Turbo.JPG 
Views:	51 
Size:	306.9 KB 
ID:	337022
    Drawing of Freddies 250 twin turbo
    note the surge tank size
    Quote Originally Posted by Katman View Post
    I reminder distinctly .




    Kinky is using a feather. Perverted is using the whole chicken

  9. #879
    Join Date
    12th February 2004 - 10:29
    Bike
    bucket FZR/MB100
    Location
    Henderson, Waitakere
    Posts
    4,196
    Not even slightly buckety, but, I'm after a GN250 carb if anyone wants to get rid of one. Seems it will be ideal for my scooter. Actually "ideal" would be a 34mm flatslide downdraught but the wife is already grumpy about what this engine is costing.

  10. #880
    Join Date
    7th September 2009 - 09:47
    Bike
    Yo momma
    Location
    Podunk USA
    Posts
    4,562
    Quote Originally Posted by speedpro View Post
    Actually "ideal" would be a 34mm flatslide downdraught but the wife is already grumpy about what this engine is costing.
    Buy the 34 and a bunch of roses for her indoors. Should do the trick.

  11. #881
    Join Date
    12th February 2004 - 10:29
    Bike
    bucket FZR/MB100
    Location
    Henderson, Waitakere
    Posts
    4,196
    Went for a brand new Mikuni TM33 pumper carb. Very nice. Then found a near perfect manifold tapering from 32mm to 28mm on Webike so ordered that and a new throttle cable and new drive belt. Only an air filter to go now and still married, dammit! I fired it up when it was in the scooter just to see. Sounds the biz. Will be interesting to see how it goes with the big carb.
    Attached Thumbnails Attached Thumbnails Click image for larger version. 

Name:	scooter1 (640x480).jpg 
Views:	53 
Size:	197.7 KB 
ID:	338188  

  12. #882
    Join Date
    12th February 2004 - 10:29
    Bike
    bucket FZR/MB100
    Location
    Henderson, Waitakere
    Posts
    4,196
    Scooter is now going pretty good so it's back to the FZR.

    Connected a battery, hooked up the laptop and Lambda meter, power on, 3rd gear, gave the wheel a pull and it fired up 1st time and idled away sweetly. Lots of tuning to do but will simplify it and not get too worried about precision.

  13. #883
    Join Date
    12th February 2004 - 10:29
    Bike
    bucket FZR/MB100
    Location
    Henderson, Waitakere
    Posts
    4,196
    Had a spare moment so fired it up again today. Took 3 tries this time to start it.
    Disconnected the tube between turbo and inlet plenum as I do not want boost at this time. I want to setup a simple TPS vs RPM fuel table.
    Warmed it up and then created a very cut down fuel table. X-axis RPM values are only 0, 1000, 4000, 11000, & 18000. Y-axis TPS % values are 0, 5, 10, 50, & 100. The same values are at each junction. It fired straight back up which I expected. I retained the lower 3 values for both X & Y axis. Blipping the throttle on the stand and the lambda was staying more or less where it had previously. A bit of dyno time is now required to see if I need to add in any rows or columns though I suspect I could probably just alter the values of RPM or TPS instead. A 5X5 table will be a lot simpler to tune even if using auto-tune.

  14. #884
    Join Date
    12th February 2004 - 10:29
    Bike
    bucket FZR/MB100
    Location
    Henderson, Waitakere
    Posts
    4,196
    Thanks to Rob & Cully of ESE I got the FZR on the dyno this arvo. Initial runs with the cutdown tables were encouraging. First runs were with the turbo disconnected from the intake plenum. Started with 50% throttle runs, monitored AFR, and recorded horsepower. A few adjustments saw an improvement and the bike running sweet with smooth response to throttle changes. Pretty sure it was only making 7hp and not revving much over 13-14000. Went on with 100% throttle runs, tipped in more fuel and made a few ignition changes as well, giving it more advance. It was making 12hp now which is less than it made with the old carbs and a couple of straight pipes but with the turbo clogging things up no real surprise.
    Connected the turbo to the intake plenum and went through making 50% throttle runs. It pulled boost straight away from about 8-10,000rpm and made a few more horsepower. Initially it was lean on boost but with the 4D fuel table I ended up adding about 20% more fuel depending on boost. Full throttle runs saw a bit more boost and more fuel needed, and more horsepower. Final reading was 18hp with a near flat torque curve from 8-15,000rpm.
    I'm hoping to go back tomorrow and chase the last few 1000 rpm. I've also caved in and added a couple more rpm points in the fuel table so I need to do a few runs with the turbo disconnected to be sure the values are correct. I've added matching rpm points in the 4D fuel table so hopefully will be able to recover a couple of horsepower we lost down low while retaining and extending the power at higher rpm.
    I also had an rpm limit programmed in based on cylinder head temp. We weren't controlling the head temp very well and I suspect when we ran the temp down at 65-85 that there was a problem which was limiting the revs. Supposedly the control should have been by a fuel cut but I saw the AFR go REALLY rich around 15-16,000rpm.

    There is a problem which I suspected with the fuel configuration. The main fuel map is TPS vs RPM. This map has values that were determined with the turbo disconnected from the intake plenum. The values are pretty good for a normally aspirated motor. The 4D map is Manifold Gauge Pressure(MGP) vs RPM. The MGP range starts at atmospheric pressure or 0 MGP with all 0s so no extra fuel added off boost. Any increase in MGP above atmospheric will cause the fuel calculated by the main fuel map(TPS vs RPM) to be increased by a percentage. The problem is that the engine pulls boost at 1/2 throttle and flows an amount of air determined by the boost, not by the throttle. It might pull the same boost at full throttle and of course the fuel is increased by the same percentage. In both cases the motor is flowing the same amount of air courtesy of the boost, but the fuel is initially calculated in the main fuel table and there are different values for 50% & 100% throttle and therefore different initial fuel amounts which are then increased by the same %. They can't both be correct. The Link help files suggest this setup but I'm thinking I may need to set the main fuel map to MAP vs RPM but the MAP signal is pretty weak generally. Possibly may have to do it but concentrate the MAP values around 70 - 100kPa to try and get some definition. Of course then there will be no need for the 4D table??? I could do the whole fuel calculation in the main fuel table with MAP values varying from 70 to 140kPa, 100kPa being atmospheric, a lower value being manifold vacuum and a value greater than 100kPa being boost. Or just use full throttle everywhere.

  15. #885
    Join Date
    18th May 2007 - 20:23
    Bike
    RG50 and 76 Suzuki GP125 Buckets
    Location
    Auckland
    Posts
    10,473
    Click image for larger version. 

Name:	Turbo and EFI.jpg 
Views:	36 
Size:	793.6 KB 
ID:	340109Click image for larger version. 

Name:	FZR EFI Turbo.jpg 
Views:	37 
Size:	798.5 KB 
ID:	340110Click image for larger version. 

Name:	2 cylinder FZR 125.jpg 
Views:	36 
Size:	767.7 KB 
ID:	340111

    Speedpro's Turbocharged and Fuel Injected twin cylinder cut down 4 cylinder FZR. The turbo fits in where the outside two cylinders were removed from.
    .

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •