It is currently Sat Sep 22, 2018 1:51 am



Forum locked This topic is locked, you cannot edit posts or make further replies.  [ 31 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: 1982 CX500TC
PostPosted: Tue Jun 18, 2013 1:23 am 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
Since Mark's chimed in with his project, I should probably write a little about mine.

It's a 1982 Honda CX500TC motorcycle: 496cc, 80 degree v-twin. It was the first 'factory' turbocharged motorcycle and its EFI system laid the groundwork for Honda's PGM-FI system used in many of its cars throughout the 80's and 90's.

It started out looking like:
Image

At the moment it looks like:
Image
Image

The EFI system is separate from the ignition, though that changed for the 83 650T, when the two merged into one ECU.

The 500T stock EFI system uses 3... yes THREE... MAP sensors for operation: one for ambient pressure, signal sourced near the air filter, one sources its signal between the throttle plates and the cylinders, and one sources its signal between the turbo and throttles. It operates in a speed-density mode at small throttle angles, when the turbo isn't boosting, and transitions to alpha-N operation with MAP compensations when under boost.

Since it lacks any type of feedback such as O2 / lambda, and they didn't want warranty claims, it runs rich... rich enough that the spark plugs needed replacement after a thousand miles or so at best... and the stock plugs were gold plated and cost $5 each back then, and are no longer available. Split-fire plugs were introduced around the same time and I could only coax a few thousand miles off them.

Ignition, as I mentioned, was a separate controller. Static timing is ~25* BTDC. It uses a MAP sensor for its operation. At high manifold vacuum, it should advance timing to ~45* and under boost would retard around 1* per PSI boost.

It uses 2 VR sensors for the cam sensors, which trigger injection, and two VR sensors for the crank position, triggering ignition. Sounds good, right?

The crankshaft has 1 tooth detected by two VR sensors. So, the way the ignition system works is the tooth sets the static timing and the MAP sensor output (Pign) changes the gate threshold on the thyristor which triggers ignition. High vacuum equals low voltage threshold and high pressure equals higher voltage threshold, therefore advancing the ignition under vacuum and retarding ignition under boost.

Might be good for what it did, but not so good for the future.

Almost impossible to implement any ignition timing control with 1 tooth flying around, and made even more difficult with two cylinders firing 280 and 440 degrees apart, thus no 'regular' timing... odd fire at its oddest?

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Tue Jun 18, 2013 1:50 am 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
I mulled around about how to get acceptable crank signals for an ECU to use.

Rotary position sensors were considered, and though some sensors were capable of great accuracy with regards to the position of the crankshaft, that method isn't very compatible with EFI systems such as O5E... and most everything available at the moment... since at power-on, the system won't know anything about the crankshaft position until it detects a 'missing tooth', a signal it expects to get but won't get with a rotary sensor. A rotary position sensor will send signals all day long, for as long as it turns, but like mentioned what I needed to have happen is that 'missing tooth' once per crankshaft revolution. That would mean a controller in there to count pulses and blank one out as needed. Too much time and money would need invested for that to be a solution for the backyard mechanic.

A bit of discussion about electronic counter-measures and stuff ended up with the realization that the bike has cousins- CX/GL models in 500 & 650 cc variants using carburetors and mechanical ignition advance mechanisms mounted off the end of the crankshaft where the 500T had nothing... its advance / retard was vacuum controlled, remember?

Anyhow, it was a short leap to take this (parts that my bike doesn't have but can be made to use):
Image
Image

... and turn it into this:
Image
*- credit to mk e, a.k.a. Mark for doing the machining on this.

Yeah, there's no 'missing tooth' there yet; I'll take care of that when I have it mounted to the bike since this is also involved in a Microsquirt project for a friend and that seems to be particular about missing tooth placement.

I'm hopeful that greater accuracy over fuel control, and greater accuracy over ignition, coupled with the ability to operate in closed-loop from idle to full boost should be able to get my 500cc's a noticeable increase in power... hopefully over 100hp to the ground.



More to come....

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Wed Jun 19, 2013 1:45 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
Okay, the theory is to use a stock VR sensor from a carb'd bike, along with the toothed wheel Mark put together for me (shown above).

Not installed, and definitely not quite aligned correctly but:
Image

Judging by the metal tang exposed on the sensors and the teeth on the gear, I think we have a winner!

Interestingly enough, it looks like the tang on the VR's is about the same width as both the teeth on the gear and the valleys so there shouldn't be much dead time between impulses, as in will probably look like a fairly clean sine wave on a scope, save for the missing tooth region, and if the firmware (and hardware) supports it, zero-volt crossings on both positive-going-negative and negative-going-positive might work.


On another note... I have an idea for a hand-held tuner which I might build into my dashboard. Comm's need to be settled on though.

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Sun Jun 23, 2013 7:16 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
Word has come down that the trigger is on its way to me.

Time to clear some garage room to drop the engine, and find an oscilloscope to test the wheel and pickups' response.

Part of me thinks Mark did some testing with it but isn't telling anyone. Part of me thinks Mark didn't test VR sensors with it when he should have since he had them in his hands.

What could be gained from using some of the most rare VR sensors on the planet for developing your system? ;)

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Sun Jun 23, 2013 9:37 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:45 am
Posts: 729
Location: PA, USA
I didn't do any secret testing .....I"m so over loaded with stuff there's just no way to side more in. Plus I've got a pretty good collection of trigger wheels and sensors to play with any time the need arises.

For you the key is going to be sensor to trigger wheel distance. The teeth are like 1mm wide so the gap wants to be small....I'd shoot for .005"-010" as a good starting point I think.


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Sun Jun 23, 2013 9:47 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:45 am
Posts: 729
Location: PA, USA
Talk about over loaded....I just noticed your VR sensors still sitting on my bench :o

They will go out to you tomorrow :oops:


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Mon Jun 24, 2013 1:32 am 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
:o
At least you noticed before I did. ;)


Adjusting the gap between the sensor and the trigger might be fun to figure out.
At least I'll get to do something. :D

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Tue Jun 25, 2013 2:07 am 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
... stuck finding injectors that will bolt in ....

Everyone around here is 'special order' so there's nothing on the shelves to test fitment against.

:evil:

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Wed Jun 26, 2013 2:28 am 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
Received the trigger wheel Monday, actually. Looks like it's going to work.
Mark added 3 tack welds to guarantee the gear won't come off. ;)
Thanks, Mark.


Will need to do some tests: low speed, where the amplitude of the signal will be least, is likely where I'd have the most problem since I am using a VR sensor.

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Wed Jun 26, 2013 6:08 am 
Offline
User avatar

Joined: Sat May 11, 2013 9:45 am
Posts: 729
Location: PA, USA
abecedarian wrote:

Will need to do some tests: low speed, where the amplitude of the signal will be least, is likely where I'd have the most problem since I am using a VR sensor.


what are you going to use to convert the VR signal to something you can input to the 5634 board? ....or are you planning to test with some other ecu HW?


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Wed Jun 26, 2013 12:30 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
mk e wrote:
what are you going to use to convert the VR signal to something you can input to the 5634 board? ....or are you planning to test with some other ecu HW?

I have a few people interested in using this type trigger on their bikes using MS so I may go that route initially. Otherwise, I'll probably do something around the LM1815 for testing purposes, or maybe order a board from JBPerf.

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Wed Jun 26, 2013 2:52 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
Continuing with my post above, I think the first thing I'm going to do is make a test stand. No reason to tear the bike apart if this isn't going to work.

I've decided to work up a MSP430 dev kit, into a pulse counter and a tachometer, taking signals from an LM1815 and the showing the results on an LCD.

Now, to find an LM1815 on the cheap....

*edit- Ordered a couple LM1815 samples from TI... and a brushless DC motor control kit. :D
*edit 2- ordered 9926 and 5216 samples from Maxim too, the latter for another project.

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Thu Jun 27, 2013 3:31 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
To expand, I have requested:
- 2 LM1815 Adaptive VRS chips
- 2 LM5101B High Voltage High-side / low side gate drivers
- 2 AMC1200 4kV peak Isolated Amplifier for Current Shunt Measurement
- 2 DRV8312 6.5A 3 phase brushless DC motor driver with inrush protection (PWM control)

... the DRV8312 is back ordered, but the others have shipped FedEx at $0 cost. :D

Maxim has yet to post anything regarding my sample request.


*edit: Nevermind... Maxim has replied.... 4 9926 (VR sensor interface) and 4 5216 (16bit rail-rail DAC) should be en route shortly.

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Wed Jul 03, 2013 1:44 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
abecedarian wrote:
To expand, I have requested:
- 2 LM1815 Adaptive VRS chips
- 2 LM5101B High Voltage High-side / low side gate drivers
- 2 AMC1200 4kV peak Isolated Amplifier for Current Shunt Measurement
- 2 DRV8312 6.5A 3 phase brushless DC motor driver with inrush protection (PWM control)

... the DRV8312 is back ordered, but the others have shipped FedEx at $0 cost. :D

Maxim has yet to post anything regarding my sample request.


*edit: Nevermind... Maxim has replied.... 4 9926 (VR sensor interface) and 4 5216 (16bit rail-rail DAC) should be en route shortly.

Parts came in.

Maxim created a new order and changed the requested quantity from 4 to 2. But... they didn't cancel the original request so now I have 6 9926 and 5216 chips. :D That uMax IC footprint is pretty tiny.

And TI said the 8312 was back-ordered but I received them with the others.

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Thu Jul 25, 2013 11:04 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
Been a little busy with things... so, sorry- no updates for the bike.

Unofficially, someone running a bike like mine was clocked at Bonneville doing over 150 MP/h using the stock EFI system, I think. Gearing limits the bike to around 135, which means he was running well over the 9500 RPM red line. Not bad for a bike with pushrods, no?

I bet O5E could've got him a bit faster. ;)

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Fri Jul 26, 2013 8:38 am 
Offline
User avatar

Joined: Sat May 11, 2013 9:45 am
Posts: 729
Location: PA, USA
abecedarian wrote:
Gearing limits the bike to around 135, which means he was running well over the 9500 RPM red line.


Or that he used a tire with a larger than stock OD ;)


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Fri Jul 26, 2013 12:50 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
mk e wrote:
abecedarian wrote:
Gearing limits the bike to around 135, which means he was running well over the 9500 RPM red line.


Or that he used a tire with a larger than stock OD ;)

I have to correct myself- the top speed is 140.

He used the biggest that would fit- a 130/90-17; stock is 120/90-17.
Circumference difference is around 2 1/4" so maybe that's possible.

I'd still like to think O5E would let him get some more power and speed though.
:D

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Sun Jul 28, 2013 6:43 pm 
Offline
User avatar

Joined: Wed May 15, 2013 4:18 pm
Posts: 14
What a fantastic project, I look forward to reading many future posts :D

The CX500 is a real favourite of mine. The non turbo versions offered a startling 60mpg (UK) with some owners reporting more.

Cafe racers looks really ace too. I have often day dreamed about restoring a Norton Commando, CB750 of CX500.... but really have too many projects as it is.

Rob,


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Sun Jul 28, 2013 6:43 pm 
Offline
User avatar

Joined: Wed May 15, 2013 4:18 pm
Posts: 14
PS, nice machining MK :)


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Mon Jul 29, 2013 2:48 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
RobT wrote:
What a fantastic project, I look forward to reading many future posts :D

The CX500 is a real favourite of mine. The non turbo versions offered a startling 60mpg (UK) with some owners reporting more.

Cafe racers looks really ace too. I have often day dreamed about restoring a Norton Commando, CB750 of CX500.... but really have too many projects as it is.

Rob,

I'm not sure I'm going to go pure café. I've been thinking about a small classical styled upper fairing, something like the photo below. My problem will be getting the radiator looking good with it since it's wider and a few inches closer to the forks (to allow room for the turbo) than the stock CX radiator.

Or maybe I'll do something like my avatar. :D

RobT wrote:
PS, nice machining MK :)

I agree.


Attachments:
SteveF.jpg
SteveF.jpg [ 83.99 KiB | Viewed 30389 times ]

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.
Top
 Profile  
 
 Post subject: Re: NanoSkwert hardware
PostPosted: Sun Nov 16, 2014 9:26 am 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
Vapor-ware?

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: NanoSkwert hardware
PostPosted: Fri Nov 28, 2014 10:29 am 
Offline
User avatar

Joined: Sat May 11, 2013 9:45 am
Posts: 729
Location: PA, USA
abecedarian wrote:

Can you recommend injectors with sufficient flow and appropriate open/closing times to support a 500/650cc engine at 900-1000 RPM idle (two cylinders, by the way), and flow enough fuel at 10000 RPM to generate > 100HP?



This site is good too
http://www.rceng.com/technical.aspx

You really don't need to worry too much about the low end with resolution new ECUs have, the 5xxx are down to the micro second.....figure anything with a duty cycle of 60-85% is fine and if you are willing to buy matched injectors or tune each cylinder then a duty cycle down to 25% or slightly less is fine. I'm pretty sure Paul has a 25% DC on his BBC airplane engine and it idles just fine for him.

I would say focus on finding something that fits with no less than 330cc/min and don't worry about it unless they are over 1000cc/min


Top
 Profile  
 
 Post subject: Re: NanoSkwert hardware
PostPosted: Sat Nov 29, 2014 8:11 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
I've actually talked with RC and they recommended either 270 or 310 would be good enough for the bike. :roll: The stock 330 injectors peaked out at ~90% duty @ > 65 pounds fuel pressure so can't see their recommendations being acceptable. Next size up they have are 370- RC has both low and high impedance versions but I worry about them being too large since the bike would foul plugs if idled for too long, with the 330's @ ~38PSI fuel pressure.

Oh well... don't want to rehash an old discussion and derail the thread.

Where's the ECU? :lol:

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: NanoSkwert hardware
PostPosted: Sun Nov 30, 2014 7:55 am 
Offline
User avatar

Joined: Sat May 11, 2013 9:45 am
Posts: 729
Location: PA, USA
abecedarian wrote:
I've actually talked with RC and they recommended either 270 or 310 would be good enough for the bike. :roll: The stock 330 injectors peaked out at ~90% duty @ > 65 pounds fuel pressure so can't see their recommendations being acceptable. Next size up they have are 370- RC has both low and high impedance versions but I worry about them being too large since the bike would foul plugs if idled for too long, with the 330's @ ~38PSI fuel pressure.

Oh well... don't want to rehash an old discussion and derail the thread.

Where's the ECU? :lol:


You want to break this out into a separate topic?

I think you're right that if you have a high end data point that says AT LEAST 330cc/min then any talk of less is silly.

Lets talk about idle though....you shouldn't be having a problem and the fact that you are points to the ECU having very poor resolution, questionable tuning or perhaps to leaking injectors. My money is on questionable tuning thinking that tolerances on injectors didn't used to be what they are today and the OEMs had to be sure all the parts would work.....so they probably used a longish injector deadtime assumption so nothing is lean but most end up rich down low.

Once you switch to a programmable ECU upgrading to the 370s should cause no problem at all and you should be able to fix the idle fouling problem you have now. Notice that I said should and that because I've run into issues with uneven airflow in the past (stock Harleys had a 15% front/back difference!) that simply couldn't be tuned out without an ecu that has individual cylinder tuning ability.....which is what brought me to motec years ago. If the problem you're having is about the same in both cylinders....you should be able to tune it right out with any ECU


Top
 Profile  
 
 Post subject: Re: 1982 CX500TC
PostPosted: Sat Dec 06, 2014 7:35 am 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
Not that I'm asking you to, but if you happen to take a look at some of the things Honda did while developing this bike, you'd see they did do some things that should help balance flow between the cylinders, which I will not enumerate here as it's mostly moot since whatever I do has to work with whatever I have.

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
Display posts from previous:  Sort by  
Forum locked This topic is locked, you cannot edit posts or make further replies.  [ 31 posts ]  Go to page 1, 2  Next

Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Theme designed by stylerbb.net © 2008
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
All times are UTC - 5 hours [ DST ]