HPRE

Menu

Members Rides

Poppit


in
Members Rides

Welcome, Guest. Please login or register.
Did you miss your activation email?
January 31, 2015, 02:47:55 PM

Login with username, password and session length

 

Author Topic: Any Power Commander V experts amongst us?  (Read 486 times)

singhg5

  • Grand Gearhead
  • *****
  • Posts: 2364
  • Karma: 0
Re: Any Power Commander V experts amongst us?
« Reply #15 on: January 30, 2015, 04:20:12 AM »
I can't remember... does the O2 sensor create voltage?  Or does it just change resistance? 

ZIRCONIA type Oxygen sensor GENERATES voltage.

TITANIA type oxygen sensor does NOT generate voltage, instead their resistance varies with the concentration of oxygen around the sensor. 
« Last Edit: January 30, 2015, 04:25:11 AM by singhg5 »
1970's Jawa /  Yezdi
2006 Honda Nighthawk
2009 Royal Enfield Black G5

Ducati Scotty

  • Grand Gearhead
  • *****
  • Posts: 6459
  • Karma: 0
  • 2010 Teal C5
Re: Any Power Commander V experts amongst us?
« Reply #16 on: January 30, 2015, 04:27:53 AM »
Singh, is one of those types used for wide band and one used for narrow?


singhg5

  • Grand Gearhead
  • *****
  • Posts: 2364
  • Karma: 0
Re: Any Power Commander V experts amongst us?
« Reply #17 on: January 30, 2015, 04:40:27 AM »
Singh, is one of those types used for wide band and one used for narrow?

Binary type or narrow band oxygen sensors are of two types - Zirconia or Titania. They are not interchangeable.

Linear or Broad Band sensors may also use Zirconia element but its construction design is different.
« Last Edit: January 30, 2015, 03:31:12 PM by singhg5 »
1970's Jawa /  Yezdi
2006 Honda Nighthawk
2009 Royal Enfield Black G5

SteveThackery

  • Inveterate tinkerer
  • Grease Monkey
  • ****
  • Posts: 327
  • Karma: 0
  • "If it ain't broke, keep fixing it until it is."
Re: Any Power Commander V experts amongst us?
« Reply #18 on: January 30, 2015, 06:02:26 PM »
I can say for sure that the Bullet uses a narrow band sensor which generates a voltage.

As such, it's not entirely clear how the resistor in the "cheater" does the job, but the bias circuitry on the input of the ECU appears to pull the input to 5V when disconnected from the outside world.  Thus the resistor in the "cheater" is presumably just right to pull the bias voltage down to 0.45V.

To answer Mattsz's question: yes, you can indeed monitor the voltage from the O2 sensor when it is disconnected from the ECU, and that is how I've arranged my "black box" flight recorder to work.  You need a very high input impedance on your monitoring circuitry; I don't know the exact requirement but mine is in the tens of megohms.

Thus I can run open-loop (O2 sensor disconnected from the ECU) whilst still monitoring the mixture. 

I can tell you for sure that in open-loop mode the engine runs weak a lot of the time, except at large throttle openings and high rpm.  This is exactly right for good economy, and the rich mixture protects the engine at high power levels.  Perfect.

In closed-loop mode, everywhere it runs weak in open-loop, it runs at stoichiometric* instead.  The rich zones are unaffected and still run rich.

Thus the engine should be a little less economical when running closed-loop (which is a well understood feature of closed-loop running in gasoline engines, but the reduction in pollutants other than CO2 makes it worth it).

Depending on which text book you read, the best mixture for maximum power is around 11:1 or 12:1, and the best mixture for maximum econonomy is between 16:1 and 18.1.

Both of these are outside the measurement range of a narrowband sensor like we've got, so I can't say anything other than "rich" or "weak" from my flight recorder data logs.

*When I say the engine runs at stoichiometric, it is more accurate to say the mixture hunts around stoichiometric.  I can post some data log graphs if you like.
2014 B5, stock apart from disabled auto-decomp.  Also ride a Kawasaki Versys 650 for everyday use.

Previous:
'06 ElectraX (Good bike, had no trouble at all)
'02 500ES (Fully "Hitchcocked" - 535, cams, piston, etc - and still a piece of junk)

SteveThackery

  • Inveterate tinkerer
  • Grease Monkey
  • ****
  • Posts: 327
  • Karma: 0
  • "If it ain't broke, keep fixing it until it is."
Re: Any Power Commander V experts amongst us?
« Reply #19 on: January 30, 2015, 06:16:49 PM »
[quote author=gashousegorilla link=topic=21067.msg235112#msg235112 date=1422569415
 BTW, using your trusty Lab equipment, if you could determine what the stock ignition timing curve looks like.... and or how it varies based on inputs from the MAP and TPS sensors.   That would be pretty cool and useful info.   That rotor has 24 nodules on it... with ONE longer one .  The one longer nodule, when under the crank position sensor is at or near TDC.  I'm not sure.... and I should have checked this when I had side cover and head off the last time..... If the front , middle or even rear of the nodule is at TDC.
[/quote]

Actually, GHG, according to what I've been taught at university, it's probably none of those. 

As you will appreciate, the gap in the rotor is indicated by a break in the regular waveform, and the timing reference point is usually the first falling edge after the end of the gap.  I've attached a picture to illustrate it.  This was taken from a Honda CB500R, though, not a Bullet.

Regardless of that, the reference point is normally somewhat before the most advanced point that the ignition timing can be.  For instance, if a particular engine has a maximum designed spark advance of 30 degrees BTDC, the reference point might be (say) 40 degrees or more BTDC.  Some engines are arranged to have the reference point exactly 90 degrees BTDC (there's a good reason for this).  After picking up the reference point, the ECU then waits a precisely calculated time (based upon the desired ignition point and the engine rpm), and after the calculated time has expired it then triggers the ignition coil.

Modern systems using this "delay time from the reference point" method can position the spark to within 0.1 degrees of crank rotation (although I doubt very much that applies to ours!)

HOWEVER - I haven't looked at the Bullet arrangement, so it could just be that they've cussedly done it different from everyone else!  :-)

2014 B5, stock apart from disabled auto-decomp.  Also ride a Kawasaki Versys 650 for everyday use.

Previous:
'06 ElectraX (Good bike, had no trouble at all)
'02 500ES (Fully "Hitchcocked" - 535, cams, piston, etc - and still a piece of junk)

SteveThackery

  • Inveterate tinkerer
  • Grease Monkey
  • ****
  • Posts: 327
  • Karma: 0
  • "If it ain't broke, keep fixing it until it is."
Re: Any Power Commander V experts amongst us?
« Reply #20 on: January 30, 2015, 06:26:21 PM »
BTW, using your trusty Lab equipment, if you could determine what the stock ignition timing curve looks like.... and or how it varies based on inputs from the MAP and TPS sensors.   That would be pretty cool and useful info.

Absolutely agree.  It's not trivial to do this, because the data rate I would need in order to record the rotor waveform is much higher than I'm currently recording at, so I might need a punchier microcontroller.   I'm still mulling this over and will definitely look into it.

Part of me thinks it might be easier to emulate the various sensor inputs (including the rotor sensor) using kit on a workbench, such that I could "characterise" the ECU off the bike under controlled conditions - systematically working through the rev range at different throttle openings, etc.  That might turn out to be easier than trying to do it all "live" on a bike that's running up and down the road (I don't have access to a rolling road).

I'm working on this in my head, and when I've done some real measurements I promise to report the results as soon as I have them.
2014 B5, stock apart from disabled auto-decomp.  Also ride a Kawasaki Versys 650 for everyday use.

Previous:
'06 ElectraX (Good bike, had no trouble at all)
'02 500ES (Fully "Hitchcocked" - 535, cams, piston, etc - and still a piece of junk)

Scotty Brown

  • Scooter
  • **
  • Posts: 76
  • Karma: 0
  • Scotty Brown
Re: Any Power Commander V experts amongst us?
« Reply #21 on: January 30, 2015, 06:44:30 PM »
Anyone else have the feeling of Sausage being made ?  Some things I just don't want to know.  However for those that do.........................It's O K .............

Ducati Scotty

  • Grand Gearhead
  • *****
  • Posts: 6459
  • Karma: 0
  • 2010 Teal C5
Re: Any Power Commander V experts amongst us?
« Reply #22 on: January 30, 2015, 06:47:04 PM »
Binary type or narrow band oxygen sensors are of two types - Zirconia or Titania. They are not interchangeable.

Linear or Broad Band sensors may also use Zirconia element but its construction design is different.

Interesting.  Thanks for the info. 

gashousegorilla

  • Grand Gearhead
  • *****
  • Posts: 2851
  • Karma: 0
Re: Any Power Commander V experts amongst us?
« Reply #23 on: January 30, 2015, 11:00:10 PM »
[quote author=gashousegorilla link=topic=21067.msg235112#msg235112 date=1422569415
 BTW, using your trusty Lab equipment, if you could determine what the stock ignition timing curve looks like.... and or how it varies based on inputs from the MAP and TPS sensors.   That would be pretty cool and useful info.   That rotor has 24 nodules on it... with ONE longer one .  The one longer nodule, when under the crank position sensor is at or near TDC.  I'm not sure.... and I should have checked this when I had side cover and head off the last time..... If the front , middle or even rear of the nodule is at TDC.


Actually, GHG, according to what I've been taught at university, it's probably none of those. 

As you will appreciate, the gap in the rotor is indicated by a break in the regular waveform, and the timing reference point is usually the first falling edge after the end of the gap.  I've attached a picture to illustrate it.  This was taken from a Honda CB500R, though, not a Bullet.

Regardless of that, the reference point is normally somewhat before the most advanced point that the ignition timing can be.  For instance, if a particular engine has a maximum designed spark advance of 30 degrees BTDC, the reference point might be (say) 40 degrees or more BTDC.  Some engines are arranged to have the reference point exactly 90 degrees BTDC (there's a good reason for this).  After picking up the reference point, the ECU then waits a precisely calculated time (based upon the desired ignition point and the engine rpm), and after the calculated time has expired it then triggers the ignition coil.

Modern systems using this "delay time from the reference point" method can position the spark to within 0.1 degrees of crank rotation (although I doubt very much that applies to ours!)

HOWEVER - I haven't looked at the Bullet arrangement, so it could just be that they've cussedly done it different from everyone else!  :-)


  Brilliant system.  NICE !    Ummmm ..... Has me wondering about a few things. The Honda and the Enfield , I believe at least share the same components.  In that they are BOTH made by Keihin.  So I wouldn't doubt the Enfield uses a similar system, and maybe just as accurate.


  Also with the ignition control on the PC-V.  Are they perhaps  "learning" that trigger point,  intercepted at the crank sensor.  And retarding or advancing that spark , intercepted at the ignition coil.  And with the rev extend feature, as long as they have control of ignition and fuel..... the stock rev limiter starts cutting spark and THEN fuel if I remember right... They may be able to set a redline whereever they want.


Absolutely agree.  It's not trivial to do this, because the data rate I would need in order to record the rotor waveform is much higher than I'm currently recording at, so I might need a punchier microcontroller.   I'm still mulling this over and will definitely look into it.

Part of me thinks it might be easier to emulate the various sensor inputs (including the rotor sensor) using kit on a workbench, such that I could "characterise" the ECU off the bike under controlled conditions - systematically working through the rev range at different throttle openings, etc.  That might turn out to be easier than trying to do it all "live" on a bike that's running up and down the road (I don't have access to a rolling road).

I'm working on this in my head, and when I've done some real measurements I promise to report the results as soon as I have them.


  Damn .... I didn't think it would be that easy.  But much respect to YOU, for getting out that on the road , Testing and tuning in that garage, HANDS on . ;)   And yeah, if you have the equipment a more static test on the bench would be more practical.
An thaibhsí atá rattling ag an doras agus tá sé an diabhal sa chathaoir.

wildbill

  • Grand Gearhead
  • *****
  • Posts: 1137
  • Karma: 0
Re: Any Power Commander V experts amongst us?
« Reply #24 on: January 30, 2015, 11:11:35 PM »
ghg - no wonder I get headaches...oh! the pain! the pain! the pain! fuse blown. lol

Also with the ignition control on the PC-V.  Are they perhaps  "learning" that trigger point,  intercepted at the crank sensor.  And retarding or advancing that spark , intercepted at the ignition coil.  And with the rev extend feature, as long as they have control of ignition and fuel..... the stock rev limiter starts cutting spark and THEN fuel if I remember right... They may be able to set a redline whereever they want.

2011 C5 black/chrome
2012 C5 maroon/chrome 
2013 B5 black
2014 Continental gt
2014 C5 tan/cheery - upgrades and work in progress

2004 mazda-speed miata
2001 bmw z3 2.2

gashousegorilla

  • Grand Gearhead
  • *****
  • Posts: 2851
  • Karma: 0
Re: Any Power Commander V experts amongst us?
« Reply #25 on: January 30, 2015, 11:29:55 PM »
 Hahaha !   Believe me there Wildbill. I give myself headaches !  :o ;)
An thaibhsí atá rattling ag an doras agus tá sé an diabhal sa chathaoir.

gremlin

  • Grand Gearhead
  • *****
  • Posts: 1293
  • Karma: 0
  • "Do one thing each day that scares you"
Re: Any Power Commander V experts amongst us?
« Reply #26 on: Today at 01:27:12 AM »
Anyone else have the feeling of Sausage being made ?  Some things I just don't want to know.  However for those that do.........................It's O K .............

It's all just BS until it actually happens.  Digital scopes are staggeringly cheap these days.  And just about any personal computer built this century can perform I/O fast enough to emulate the sensors involved
1996 Trophy 1200
2011 RE B5
1979 XS11 w/vetter terraplane
1981 XS11 streetfighter
1983 Venture Royale
1982 CB750K
1971 Triumph Trident
1969 CB450
1966 Sears (puch) 250


SteveThackery

  • Inveterate tinkerer
  • Grease Monkey
  • ****
  • Posts: 327
  • Karma: 0
  • "If it ain't broke, keep fixing it until it is."
Re: Any Power Commander V experts amongst us?
« Reply #27 on: Today at 10:18:30 AM »
It's all just BS until it actually happens. 

Well, aspiration rather than BS, I would say.  The data logging side of things has gone pretty well and given me some significant insights already.  I always expected the next stage to take longer.

Digital scopes are staggeringly cheap these days

Actually the staggeringly cheap ones are staggeringly crap, too.  I've spend weeks researching the market, and basically the "worth buying" category pretty well starts with Rigol from the Chinese side of things, and Pico from here in the West.  GBP400 gets you the base model of each, and in fact I spent GBP500 for mine, which is the cheapest 100MHz Rigol.  That doesn't feel like "staggeringly cheap" to me, although I admit I'm not particularly flush with money - perhaps you would find that cheap.

And just about any personal computer built this century can perform I/O fast enough to emulate the sensors involved

Ah!  I'm glad because it looks like you might be able to give me some advice on this.  The only analogue output I'm aware of on a PC is the loudspeaker output.  Obviously this would need some external amplification to get the output up to the 5V range used by the Bullet's sensors.  However, I've found it very difficult to get bit-by-bit real-time control of that output; the driver and API seem to be structured around playing pre-existing sound files.  My software would need to construct and stream a sound file in real time in order to respond to variations in things like engine speed and throttle opening.  The bigger challenge is that I'll need multiple channels, and of course a PC normally only has two loudspeaker outputs (one output socket with left and right channels).

This all sounds quite hard to me (though not impossible) so if you can offer any guidelines I'd be most grateful.

I could also buy a general purpose analogue I/O board for my PC, but these are rather expensive (low sales volumes, I suppose) and the thought of writing the necessary software to emulate all the sensors in real time, as well as logging the relevant outputs, is quite scary.  I'm not much of a programmer myself and fear it might take a very long time to get it right.

FYI I use Delphi 2010 to program to the Win32 API, and Oxygene in Visual Studio 2013 to program to .Net.

Are you able to offer any advice which might start me off in the right direction? 

Many thanks.
2014 B5, stock apart from disabled auto-decomp.  Also ride a Kawasaki Versys 650 for everyday use.

Previous:
'06 ElectraX (Good bike, had no trouble at all)
'02 500ES (Fully "Hitchcocked" - 535, cams, piston, etc - and still a piece of junk)

SteveThackery

  • Inveterate tinkerer
  • Grease Monkey
  • ****
  • Posts: 327
  • Karma: 0
  • "If it ain't broke, keep fixing it until it is."
Re: Any Power Commander V experts amongst us?
« Reply #28 on: Today at 01:18:44 PM »
<posted in error>
« Last Edit: Today at 02:39:45 PM by SteveThackery »
2014 B5, stock apart from disabled auto-decomp.  Also ride a Kawasaki Versys 650 for everyday use.

Previous:
'06 ElectraX (Good bike, had no trouble at all)
'02 500ES (Fully "Hitchcocked" - 535, cams, piston, etc - and still a piece of junk)