shayne
Omnipresent
1998 Blueprint
Posts: 8,639
|
Post by shayne on Apr 21, 2009 15:46:56 GMT 10
It looks to me that the new (yet to be released) up-spec model Adaptronic will be what I need. Seems to be about the best thing on the market until you get to Motecs at double the price. Fully programmable in regards to RPM points, load points, separate fuel and ignition mapping for each cylinder, will run closed loop with a wide band O2 sensor, you name it she does it. I totally agree - for the dollar they are by a huge margin the best thing around. The Motec/Autronic's are better, but cost a fair bit more and are likely to have a bunch of features that you just won't need. Bit of a shame their basic model won't do a twin. The up-spec one will when it comes out, but it will be double the price they tell me.
|
|
shayne
Omnipresent
1998 Blueprint
Posts: 8,639
|
Post by shayne on Jun 13, 2009 16:25:53 GMT 10
After some more thought, the VFR800 pump still seems to be the best option. A lot of the parts between it ad the SP throttle bodies just match up too well. The VTR & VFR tank shape are also similar, so the pump will fit in the rear where the VFR pump mounts. I did consider using a VFR tank as well, but they just don't fit well enough for my liking. The gaps just aren't right at the rear of the tank when you fit them to VTR. So I will need to modify the VTR tank to fit the submerged pump. The bit that will be tricky will be relocating the VTR breather and overflow outlets, which are going to be in the way. They will be a pain I think. I still think this is a better option than using an external pump, as it will be so neat, and there will be no need for fuel lines running back to underneath the seat like I have now for my low pressure fuel pump. I will need this space for the ECU. I have also forgotten to mention that I will definately be using the aftermarket ECU to control the ignition as well. Having a programmable ignition is a big advantage, and it won't be that hard to do. Not compared to the rest of it anyway. The only problem is I don't have the time to do it yet. Need to be patient.
|
|
shayne
Omnipresent
1998 Blueprint
Posts: 8,639
|
Post by shayne on Jul 25, 2009 20:01:04 GMT 10
I have done a little bit more on this project in the last week. Not much, but working out a few components. Have a look at your rear cylinder head, and you will see a casting that protrudes out on the left hand side at the rear. It looks like it might fit a CCT or something similar, but it is blanked off. It is a casting to fit a Camshaft Pulse Sensor, and it works by reading the rotation of a Cam Pulse Rotor that bolts to your cam gear on the exhaust camshaft. Both these components are Varadero parts, and provide a signal to the ECU. I will use them to do the same job on the VTR with an aftermarket ECU. As the camshaft assembly is the same as the VTR they should fit without modification. The VTR has a Pulse Generator Coil, which is activated by a rotor on the end of the crankshaft. It has about 7 teeth on it, with the remaining section smooth. Looks like a sprocket with some teeth missing. It operates the ignition. Honda's EFI bikes have the same set up but they use a rotor with a full section of teeth, usually about 12. It would seem that an ECU that controls ignition and fuel will work better with the full rotor. Once again, the Varadero has a 12 tooth rotor which should bolt straight on in theory, as both bikes run the same crankshaft. 12 hole fuel injectors are the best option, and as mentioned previously, I have worked out that VFR800 injectors fit into the SP1 throttle bodies if you use a combination of seals from both bikes. Honda changed the injectors in the VFR in 2006, as the part number are different from earlier models. They are all 12 hole injectors, but Honda updated them in conjunction with the VTEC changes in 2006. These injcetors are the same parts as used in CBR1000's from 2004-2007. Reckon they should do the job.
|
|
|
Post by paulbassett on Jul 25, 2009 20:37:12 GMT 10
Good luck with your project.
How expensive are the injectors?
|
|
|
Post by colinlinz on Jul 25, 2009 22:11:19 GMT 10
It sounds like it is all coming together. It will be interesting to see how it all goes when you finish. It's interesting to note that they obviously designed the motor with the future use of EFI in mind.
|
|
shayne
Omnipresent
1998 Blueprint
Posts: 8,639
|
Post by shayne on Jul 27, 2009 15:58:06 GMT 10
It's interesting to note that they obviously designed the motor with the future use of EFI in mind. You could be right. The Varadero has been injected for a number of years, so perhaps Honda planned it from day one. As for the injector cost, not sure about a new price from Honda, but they can be had off ebay for under $100USD for a set of four. I need 4 as each throttle body has two injectors, so that works out well.
|
|
Mish
Senior Member
2007- Red
Posts: 299
|
Post by Mish on Aug 9, 2009 23:52:10 GMT 10
Hey Shayne, I read that Honda have done or are going to do a fuel injected VTR250. Would any of the sensors off these be of any use?
|
|
shayne
Omnipresent
1998 Blueprint
Posts: 8,639
|
Post by shayne on Aug 11, 2009 15:38:00 GMT 10
Hey Shayne, I read that Honda have done or are going to do a fuel injected VTR250. Would any of the sensors off these be of any use? It is possible, as Honda use the same parts on a number of bikes. The Varadero is the best option though, as those parts will definately fit.
|
|
|
Post by gboezio on Aug 23, 2009 0:08:19 GMT 10
Hi Shane, I thought that I could post some of the infomation here as well, for the VTR you will need about 36 to 42 lb/h injectors per cylinders or 350-450 CC, mine are 750 and are a pain to set at idle, they run at 49 % duty cycle at 9500 RPM @ 7.5 psi @ 11.0 AFR. The crank wheel is a 12-3, last tooth trigger one coil, first tooth trigger the other, the last tooth is aligned with front TDC, the stock ICU run wasted spark, the 12-3 is not ECU decodable, the gap is way too big, the timing is all over the place at low RPM or cranking, as much as 45 degree off, as the engine compress, it slows down and the ECU assume a constant speed, beside this the long flat spot is generating a false teeth during the 90 deg hole messing up the readings, yellow wire is sensor positive and yel/white is negative. I used a Varadero wheel and cut one tooth off + cam synch, a full wheel could be used as well with cam synch. The VTR is difficult to tune using MAP, the signal is pulsing a hell of a lot, for a NA setup using ITB's, I think I'll use Alpha-N algorythm (TPS+RPM) like most factory bikes.
I'm hesitating to post my VE and timing tables since they are not perfect yet.
I can't wait to see how yours will turn out, Cheers GB
|
|
|
Post by colinlinz on Aug 23, 2009 10:39:18 GMT 10
While the hardware is fairly easy to adapt and fit, the difficulty is always going to be in getting the programming to work. Manufacturers spend heaps of time developing maps and still manage to stuff them up. The TL was one example where Suzuki had to develop a new map and download it to the existing customers bike. This is in fact a common occurrence these days. Factories continue developing the software and licensed service centres download the up grade when you go in for a service. This can't be done at non dealer shops as they don't have access to the server based diagnostic platforms.
In saying that, and as much as I'd hate to see his balding head get bigger, if anyone has the patience and pedantic character to get it right outside of a specialist tuning workshop, it would be Shayne.
GB, you may just get it all to work too. Best of luck with it and keep us updated.
|
|
|
Post by colinlinz on Aug 23, 2009 10:45:48 GMT 10
Shayne, I've made the thread a sticky, so now your going to have to pull your finger out and start doing something
|
|
|
Post by gboezio on Aug 24, 2009 2:22:02 GMT 10
I'll help, mine is running strong. I just use a different induction system than most folks will use, I'm running a single 60mm throttle body and a 1.9 l plenum so the VE maps may differ, but I can help doing the wiring, dwell, sensors config, getting the tach to work and a whole lot of things. This is fun
|
|
shayne
Omnipresent
1998 Blueprint
Posts: 8,639
|
Post by shayne on Aug 24, 2009 16:21:21 GMT 10
Plenty of work to do, no doubt about it. I have the fuel pump and bank angle sensor in my parts collection now. Need to sort out some injectors now, so I will have to look for a supplier at some stage.
I think that your turbo may be a factor in the MAP sensor problem Giovanni. Honda use this sensor in the SP (RC51), VFR800 and Varadero which woud would all have similar pulses, so it must work well enough.
|
|
|
Post by colinlinz on Aug 24, 2009 17:29:10 GMT 10
I can help out with some injectors. What about a couple of Mercruiser ones, or maybe even some Optimax ones. I wouldn't worry about the etec ones, they are just too big.
Intake manifold pressure will fluctuate a lot on a twin at idle, but if you have the other sensors in the system, some sort of mapping strategy should be able to be employed. For example ignore the MAP reading when the TPS registers idle, and just use a base value. Although you would still want the MAP to sense ambient air pressure when the ignition is switched on, so the mapping can adjust itself to altitude.
|
|
shayne
Omnipresent
1998 Blueprint
Posts: 8,639
|
Post by shayne on Aug 25, 2009 10:09:47 GMT 10
Well I know that Keihin injectors fit, and that Honda use the same injectors on a number of their bikes ie the VFR800 and CBR1000 on certain years use the same part, so I assume that they will handle the flow requirements of a number of engines. I could buy them and be fairly safe I think.
But. The manifold I have uses four injectors, which is for a dual stage fuel delivery. So I do not know whether I would be better off with two smaller injectors and two larger one's, or four the same. And I have no way of finding out the specs for the Honda parts so I need an aftermarket supplier that have info on the injectors they supply. Then I would need to work out what delivery requirements the engine would have to get the right injectors.
A bit to work out yet.
|
|
|
Post by gboezio on Aug 25, 2009 10:29:58 GMT 10
Intake manifold pressure will fluctuate a lot on a twin at idle, but if you have the other sensors in the system, some sort of mapping strategy should be able to be employed. For example ignore the MAP reading when the TPS registers idle, and just use a base value. Exactly, what happened in the last 24 hours, I could not get enough headroom to tune since it idles at 75 Kpa, so it gives me 25 Kpa to tune idle, light load and high load, I tried many settings to no avail, it' s impossible to tune, the light load and idle squares are the same, so it idle very rich and bogs down a lot, if you have bigger cams and run individual throttle bodies, it will pull even less vacuum and become impossible to fine tune, I swear, the MAP is all over the place. But there's hope Since the turbo won't spool much under 3100 RPM, I can use the throttle angle (Alpha-N) from 0 to 2300 rpm then it will blend up to 3100 and become full speed density so it will be able to deal with the hairdryer. The setup is a rough guesstimate and as it beats the heck out of my fine tuned former low rpm setup that took days to make. It drives like a charm now and pull more violently than it ever did. I suggest to either do an hybrid alpha-N like this, or to go to full alpha-N since you run no turbo, but MAP is very doable, at cruise it will see 60 Kpa and on decel down to 30, you may get a bit less with ITB's. As the crank sensor, the RC-51 uses a different setup on the flywheel, the VTR 12-3 wheel is useless but the Varadero works wonder, I cut off one tooth and added a Hall effect sensor on the cam, Honeywell GT-101, won't skip a pulse since then. The CBR wheel will not work, it's smaller and not embossed. The turbo will not affect my crank and cam sensors. Here's a scope of the pulse gens, Both uses the falling edges as trigger and this is what you need to do too : Now how the ECU see it, pay attention to how the pulses are uneven during cranking, the compression makes wider pulses, this is why the 12-3 wheel is bad, imagine 3 teeth missing, the timing lands way off :
|
|
shayne
Omnipresent
1998 Blueprint
Posts: 8,639
|
Post by shayne on Aug 25, 2009 17:52:54 GMT 10
I don't understand why you had to cut one tooth off the crank trigger. It is a problem with the Microsquirt ECU you are using?
Honda's ECU uses the crank trigger why can't you?
|
|
|
Post by gboezio on Aug 25, 2009 18:50:02 GMT 10
I wanted to have some crank reference, I was afraid that my cam synch could be borderline and sometimes synch on the wrong crank pulse offsetting the timing by 30 degrees. With the notch on the crank signal used as the timing reference it would be either power stroke or exhaust stroke at the right crank angle. I'm very afraid of a sudden 30 degree advance, it would destroy the engine in seconds. If you use the stock Varadero components, you may get a stable signal and a useable 12 tooth setup. The stock Varadero Cam sensor will look cleaner on your bike, mine is a hack and will probably be dropped on the track a few times (one so far). Some of it is Microsquirt related, like the fact that I could set the first tach to 30 deg ATDC by cutting the tooth I wanted, this will leave time to the processor to fire the injectors, charge the coils, advance, etc The processor can't handle too much things at once, if the tach is too late, it will run out of time and fire late, sometime loose spark. The stock wheel will set the tach 120 degree ATDC and running wasted spark will leave about a half turn to complete many operations, no good, I could not rev more than 4500 RPM Honda's ECU uses the crank trigger why can't you?
If you mean the 12-3 stock wheel, the Honda ECU is working in a way that Microsquirt can't be configured. If you mean the 12 tooth and cam synch, I think you need the extra code to handle the 3 teeth from the cam synch, or cut off 2. They made a special code for Honda's V twin. You will need to retain the stock ignition box (the six wires), it works wonderfully, just ground the inputs 3 ms and spark (open, no ground) for 1 ms. Basically set Dwell at 3 ms and spark duration at 1 ms. It took me time to find this, the sparks are so powerful that they will ignite any fuel ratio and avoid flooding the engine, even if I had to torch the plugs 4-5 times. Home made cam synch: Sensor housing, still the Macgiver way : I hope this helps GB
|
|
|
Post by vtrfighter on Sept 2, 2009 22:29:21 GMT 10
|
|
shayne
Omnipresent
1998 Blueprint
Posts: 8,639
|
Post by shayne on Sept 3, 2009 19:35:53 GMT 10
Cheers.
|
|
|
Post by dgb on Sept 24, 2009 20:38:09 GMT 10
You'll pick up a tiny bit more top-end power, but the main difference will be the driveability and economy. For example, there's a racing class in the US that runs a 1600cc Toyota engine (4 x 400cc pots) and when the rules called for carbies only (twin 45mm DCOE Webers) they got 238hp maximum. When the rules changed to allow EFI, the power went up to 242hp, and that's almost dyno error. But the big difference was that they improved the mid-range power at 6,000rpm by a full 20hp. The reason for this is the very fine control over the mixtures that you get at nearly every rev point and throttle position. You just can't get that with any carby. Carbies have the slight advantage of introducing the fuel further out than a typical EFI installation and that cools the incoming air a bit more, so they can still make up near the same top-end power as even a good EFI engine. You can of course mount the injectors out in the open end of the inlet trumpet to get the same advantage but then at small throttle openings the fuel tends to puddle a bit and then the engine isn't good stooging around town and at idle. Also because you can run the mixtures at pretty much whatever you like, as well as good power you can also get good fuel economy. An engine I built a while back for one of my road cars made 185hp from 1.6 litres and drove just fine from about 1800rpm on, but still got 42mpg or 6.7 L/100km. FWIW I'm currently building another 1600cc engine for one of my road cars that will be about 240hp and it'll still be totally driveable and (hopefully!) also get good economy. For the 500cc pot of the VTR, I would recommend something about 50mm ID, 42mm is too small for best power at high revs but it'll be perhaps a little nicer around town. Cars and bikes are two very different things, as many of the F1 engineers have found in MotoGP. While they can make great engines when it comes to power, they have failed when it comes to rider connection. This is the area that is so difficult for EFI to compete with carbies. Power doesn't make a quick bike, even if it's mid range. It's the connection between the bike and rider, the rider has to be able to feel what's happening at the tyre, and for this to happen with EFI the system needs to be very refined. With throttle body size there is one thing to remember. Carbies need to have a modest size as too large will slow the air velocity down. This will in turn reduce the venturi effect and reduce the ability to draw fuel. EFI doesn't suffer from this. All we really need to consider is the best velocity to help use the mass of the air to charge the cylinder at the rpm we want. The VTRSP1 uses a 54mm body, but the SP2 uses a 62mm body. This gives an indication of the differences, and just how far we could go with our engines while still be perfectly fine for street use. Here's a bit of interesting history. John Britten turned the superbike world upside down in the early 90's with his literally home made V twin superbike. He also stated that the feature which would get the most attention when he took this bike to America would be the electronic fuel injection. If memory serves me correctly it was a Links system. This bike toyed with factory ducati's and honda's. Britten gave most of the credit to the fuel injection.
|
|
|
Post by colinlinz on Sept 26, 2009 22:19:32 GMT 10
Well it did after cracking a cylinder lining and welding it up, and up to the point it blew a reg/rec and the battery went flat causing them to DNF in the dying laps of the race.
He was a brilliant and driven engineer, he also had a great team of dedicated people by his side.
|
|
|
Post by dgb on Sept 26, 2009 22:54:40 GMT 10
And his quote after the race in reference to the electrical problem was "serves me right for using a ducati part".
For those that don't know John Britten died in 1995, aged 45 of a melanoma.
|
|
|
Post by colinlinz on Sept 27, 2009 9:36:20 GMT 10
He could have chosen worse. He could have went with a Honda unit
|
|
Mattjin
True Stormer
I don’t think you are stupid. You just have a bad luck when thinking.
Posts: 1,168
|
Post by Mattjin on Dec 20, 2009 13:30:26 GMT 10
gboezio, which ecu are you using? I must have missed it going through the thread.
Shayne, if you have not purchased an ECU yet do alot of thinking and be wary of alot of advice you find online. You have to be more careful choosing an ecu for a bike compared to a car. A weird flat spot at low speed can send you off the thing, whereas a car will just keep going.
What should be done on these particular motors is to run a blending of TPS and MAP. But for the MAP reading it needs to be both mechanically and electrically filtered due to the unstable signal. TPS alone will not provide for changes in engine load for the same throttle input, and MAP alone is not stable enough at low rpm. To mechanically filter the signal is easy, just use a restrictor in the line around 0.8mm as close to the inlet as possible. You would pickup from both throttle into a "T" and then immediately run a restrictor. Too small and the signal will lag, too big and it wont dampen enough. Ideally an Air Flow Meter of some sort would run better, but packaging it would be near impossible.
My job is engine management programming specialising in vehicle emissions compliance, and have been doing this for just on 20 years, so I have quite a bit of experience with this.
|
|
shayne
Omnipresent
1998 Blueprint
Posts: 8,639
|
Post by shayne on Dec 21, 2009 17:07:32 GMT 10
Giovanni is using a Microtech I think. If you have a look on the Superhawk forum you can read about his bike.
As for me, I would not use an ECU such as this. I want as much control as I can get, and the ability to run separate fuel and ignition maps for each cylinder, and have unlimited rpm and load points. I will be using factory Honda sensors for the inputs.
As for the ECU, at this stage I think the best option is the new Adaptronic SuperECU. Not sure if it is on the market yet, but it must be close. As my project has stalled while I build a new house and shed, I have not looked at them for a while.
|
|
billzilla
Senior Member
2005 - Matte Black
Posts: 338
|
Post by billzilla on Dec 22, 2009 14:28:37 GMT 10
As for the ECU, at this stage I think the best option is the new Adaptronic SuperECU. Not sure if it is on the market yet, but it must be close. As my project has stalled while I build a new house and shed, I have not looked at them for a while. I've mentioned the Adaptronic here before, the problem with the earlier model was that it can't do the odd-firing that the VTR engine has. I'm hoping that the new version can, as they're a great little unit, by far the best bang-for-buck available. I'm not sure if you can add load points on the map like you can with a Motec, but on the Motec with some race engines if you find the standard 250rpm gap is too large you can just add more points until it all smooths out.
|
|
shayne
Omnipresent
1998 Blueprint
Posts: 8,639
|
Post by shayne on Dec 22, 2009 17:11:27 GMT 10
Have a look on the Adaptronic web site. There is a lot of discussion on the Super ECU. It will do just about anything from what I can remember. I swapped a few emails with them, and it will run a twin and have no limit on load or RPM points, so you can put them where you want. They tested a prototype version on a Aprilia RSV.
|
|
Mattjin
True Stormer
I don’t think you are stupid. You just have a bad luck when thinking.
Posts: 1,168
|
Post by Mattjin on Dec 22, 2009 18:26:23 GMT 10
I would be leaning more towards either a Haltech or Motec. Both are a fair bit more established with less bugs and both will allow full control over mapping and odd-fire engines. We actually use alot of Haltech ECUs when emissions compliance is paramount, and this is usually the most demanding situation you can put an ECU through. Mostly this is due to their closed loop control, which isnt really a top priority in this application though. Adaptronic are a good midrange ecu and are getting there slowly. They will soon be a real force, but as they progress so too do the more advanced systems.
The general rules you should apply is this... firstly if a hand controller can be attached it is usually not a good ecu, and secondly, race applications need less ecu power than street applications (most people think it is the other way). The cheap ecu might get you down the drag strip, but it wont give the the finer control needed for a street application.
If anyone has any ecu specific questions, I am more than happy to give some open advise.
|
|
|
Post by colinlinz on Dec 22, 2009 18:56:28 GMT 10
|
|