| Author |
Message |
fusername
Samba Member
Joined: March 15, 2006
Posts: 2897
Location: Boston MA
|
 Posted: Fri Jun 01, 2018 12:25 pm Post subject: FI injector timing question |
 |
|
Had anyone ever looked at the injector timing with a scope? I am wondering what the variation in time open is when driving. Like are they on for 5% of the time at idle and 100% at full throttle and 5k RPM? This is driven partly by curiosity I've had for years, and partly because when I went from points to petronix ignition my bus started running super rich and prolly behaved. No idea if the two are related or if something bad happened over the winter when I want driving the car, so I have a long list to go thru. Only thing I know for sure is it is not the tsII sensor.
_________________
[email protected]
Need something custom bent up? shoot me an email, maybe we can make it work!
FORSALE: Thrust cut T4 and 1.9 main bearings
| obnoxiousblue wrote: |
| Maybe Ben Pon's ghost comes and vomits NOS stampings for your bus, but not mine! |
|
|
| Back to top |
|
 |
SlowLane
Samba Member
Joined: July 11, 2005
Posts: 1044
Location: Livermore, CA
|
| Posted: Fri Jun 01, 2018 1:08 pm Post subject: Re: FI injector timing question |
|
|
If you're really curious, figure out how to turn an ignition dwell meter into an injector duty-cycle indicator. Same principle.
FWIW, I briefly installed a Pertronix on my van's engine years ago. Didn't like how it ran with the Pertronix, so I replaced it with a Crane Cams XR700 optical-trigger system. The car ran beautifully with that system for years.
_________________
Present:
'81 Westfalia: 2L, manual. Originally Canadian, now Californiated
Back in the day:
'72 Super Beetle
'69 Camper Van - Corvair powered
'71 Window Van - Transferred Corvair from '69
"With sufficient thrust, pigs fly just fine." - Internet RFC 1925
"They say a little knowledge is a dangerous thing, but it is not one half so bad as a lot of ignorance." - Sir Terry Pratchett
Last edited by SlowLane on Sat Jun 02, 2018 12:15 pm; edited 1 time in total |
|
| Back to top |
|
|
telford dorr
Samba Member
Joined: March 11, 2009
Posts: 3555
Location: San Diego (Encinitas)
|
| Posted: Fri Jun 01, 2018 1:14 pm Post subject: Re: FI injector timing question |
|
|
If you don't have a dwell meter, put the points back in and see if the mileage changes (I suspect it won't).
_________________
'71 panel, now with FI
'Experience' is the ability to recognize a mistake when you're making it again - Franklin P. Jones
In theory, theory works in practice; in practice, it doesn't - William T. Harbaugh
When you're dead, you don't know you're dead. The pain is only felt by others.
Same thing happens when you're stupid. - Philippe Geluck
More VW electrical at http://telforddorr.com/ (available 9am to 9pm PST) |
|
| Back to top |
|
|
Tcash
Samba Member
Joined: July 20, 2011
Posts: 12844
Location: San Jose, California, USA
|
|
| Back to top |
|
|
SGKent 
Samba Member
Joined: October 30, 2007
Posts: 41031
Location: Citrus Heights CA (Near Sacramento)
|
| Posted: Fri Jun 01, 2018 2:43 pm Post subject: Re: FI injector timing question |
|
|
the interval is not static.
Coil secondary creates a small pulse in the coil primary as it collapses after the points open and the primary magnetic field quits. The ECU reads this pulse and opens the injectors. The coil RPM, TS1, TSII, door position and direction, all send information the ECU which then adds interval to the time the injectors are open. The ECU never subtracts, it only adds to the base interval. If the coil pulse is affected by changing from points to an electronic ignition module then the ECU base pulse length changes so the base injector pulse is altered. That is why some electronic ignitions don't work well with L-jet. Also the strength of that pulse at the coil matters so when someone puts in a super coil and electronic ignition that can affect FI too.
_________________
“Most people don’t know what they’re doing, and a lot of them are really good at it.” - George Carlin |
|
| Back to top |
|
|
Wildthings
Samba Member
Joined: March 13, 2005
Posts: 50352
|
| Posted: Fri Jun 01, 2018 3:18 pm Post subject: Re: FI injector timing question |
|
|
| telford dorr wrote: |
| If you don't have a dwell meter, put the points back in and see if the mileage changes (I suspect it won't). |
--^^^^^--
I would go this route and look carefully from something that got knocked/left loose while I was doing the original work. |
|
| Back to top |
|
|
fusername
Samba Member
Joined: March 15, 2006
Posts: 2897
Location: Boston MA
|
| Posted: Fri Jun 01, 2018 8:25 pm Post subject: Re: FI injector timing question |
|
|
Oh I have plenty of instrumentation to check it out, scope etc, was just curious if anyone had a base line on time at idle whole warm. Something is very wrong with my FI, and it's probably a loose wire, it ran fine last fall until my starter went.
I'll post a longer follow up when I get further along. Afaik, the ecu sets the on time based on afm, ts2 and ts1, and battery voltage. Fuel delivery is furtheret affected by vacuum affecting fuel pressure and rpm. Injectors fire every second trigger from the coil.
_________________
[email protected]
Need something custom bent up? shoot me an email, maybe we can make it work!
FORSALE: Thrust cut T4 and 1.9 main bearings
| obnoxiousblue wrote: |
| Maybe Ben Pon's ghost comes and vomits NOS stampings for your bus, but not mine! |
|
|
| Back to top |
|
|
Wildthings
Samba Member
Joined: March 13, 2005
Posts: 50352
|
| Posted: Fri Jun 01, 2018 8:40 pm Post subject: Re: FI injector timing question |
|
|
| If you were for some reason getting a double blip each time a magnet passed the pickup you might well get twice the fuel your need. |
|
| Back to top |
|
|
aeromech
Samba Member
Joined: January 24, 2006
Posts: 16971
Location: San Diego, California
|
| Posted: Fri Jun 01, 2018 11:27 pm Post subject: Re: FI injector timing question |
|
|
Why don’t you stick a noid into one of the injector connectors and watch the flashes as you run up the engine. Might be too fast to tell though
_________________
Lead Mechanic: San Diego Air and Space Museum
Licensed Airframe and Powerplant Mechanic
Licensed Pilot (Single engine Land)
Boeing 727,737-200-300-400,757,767
Airbus A319,320,321
DC9/MD80
BAe146
Fokker F28/F100
VW type 1 1962,63,65,69,72
VW Type 2 1971 (3 ea.) 1978, 1969
VW Jetta
VW Passat
Capable of leaping tall buildings in a single bound |
|
| Back to top |
|
|
furgo
Samba Member
Joined: September 06, 2016
Posts: 944
Location: Southern Germany
|
| Posted: Sat Jun 02, 2018 1:48 am Post subject: Re: FI injector timing question |
|
|
| fusername wrote: |
| Had anyone ever looked at the injector timing with a scope? I am wondering what the variation in time open is when driving. Like are they on for 5% of the time at idle and 100% at full throttle and 5k RPM? |
To answer your original question, I did this a while ago on the bench, as I couldn't bring the instruments where the bus is parked. I wasn't diagnosing an issue, it was more to visualize and better understand how injector pulse generation worked.
I've got two spare ECUs: one for a '79 bus (Federal, manual) and one for a '78 Opel Commodore (same L-Jetronic, but with 6 cylinders). They both drive the same type of injectors, with the same flow rating (the ones with the blue connectors on the bus, the ones with the grey connectors on the Opel). I used a spare harness, spare AFM and a potentiometer to simulate the TS2, a frequency generator as the terminal #1 input and observed the injector waveform on the oscilloscope.
The reason I never posted the results was because I could not get the bus ECU to produce an output. I never found out whether it was because it was expecting a different clock input signal or whether something else was broken.
However, I did get the Opel ECU to produce an output. From the inside, they are nearly the same. Of the three big ICs that do most of the analog computing work, two were the same for both ECUs. The only visual and functional difference I could tell was in the third IC, the one that does the frequency division on the first stage: while the bus ECU divides the terminal #1 frequency by two (4 injectors), the Opel ECU divided it by 3 (6 injectors).
| fusername wrote: |
| Oh I have plenty of instrumentation to check it out, scope etc, was just curious if anyone had a base line on time at idle whole warm. |
If you are interested, I can post here what I measured back then on that ECU, with the caveats above.
I simply didn't want to put it in its own thread due to it not being a bus ECU. The results are probably similar, but without having confirmed it, I didn't want to make assumptions and potentially generate confusion.
| fusername wrote: |
| I'll post a longer follow up when I get further along. |
Looking forward to that!
| fusername wrote: |
| Afaik, the ecu sets the on time based on afm, ts2 and ts1, and battery voltage. Fuel delivery is furtheret affected by vacuum affecting fuel pressure and rpm. Injectors fire every second trigger from the coil. |
That sounds correct. The only thing I would add is that the ECU will only start doing this if it sees its clock signal from terminal #1.
Roughly, that signal is a periodic pulse (0-180Hz) of hundreds of volts for a couple of milliseconds, but unfortunately, there are no specs for it. Thus it's difficult to diagnose what's going on if that signal does not seem to meet the ECU input requirements. The input stage of the ECU simply attenuates it, and the frequency division IC then converts it to a square signal of half the frequency for the subsequent stages.
As a final note, I used a 12 V square signal clock input to simulate terminal #1. Some observations:
• The minimum amplitude that the ECU seemed to recognize as a clock input was 10 V. Below that there was no injector pulse generation
• The ECU did not care about the duty cycle of the input pulses –as long as it was below 100%, obviously.
I interpreted that as the ECU clock input being quite tolerant to pulse amplitude and unaffected by pulse length from terminal #1.
_________________
'79 Westy, P22 interior, FI 2.0 l Federal, GE engine (hydraulic lifters)
Decode your M-Plate |
|
| Back to top |
|
|
Abscate
Samba Member
Joined: October 05, 2014
Posts: 22670
Location: NYC/Upstate/ROW
|
| Posted: Sat Jun 02, 2018 2:15 am Post subject: Re: FI injector timing question |
|
|
It’s probably triggering on the rising edge, so that’s pretty common.
_________________
.ssS! |
|
| Back to top |
|
|
furgo
Samba Member
Joined: September 06, 2016
Posts: 944
Location: Southern Germany
|
| Posted: Sat Jun 02, 2018 3:21 am Post subject: Re: FI injector timing question |
|
|
Indeed. When incompatibility issues with electronic points replacement modules are mentioned, they could be due to the magnitude of that signal from the primary. That is, how much the attenuation applied on the ECU's input stage affects it. That said, the voltage pulse on the primary must be high enough to generate a spark in the secondary. So high that I'd think it would make it a moot point, though.
That (and sensor inputs + supply voltage, obviously) is all the ECU cares about to generate the injector pulses.
As far as I know, electronic points modules simply replace a mechanical switch by a transistor. Some of the more expensive ones have got microcontrollers that do fancy things with the primary waveform. But that directly affects the spark itself (time, strength), not the FI system, so I've never understood the claims about those modules affecting what the ECU does, other than making the signal more stable.
_________________
'79 Westy, P22 interior, FI 2.0 l Federal, GE engine (hydraulic lifters)
Decode your M-Plate |
|
| Back to top |
|
|
telford dorr
Samba Member
Joined: March 11, 2009
Posts: 3555
Location: San Diego (Encinitas)
|
| Posted: Sat Jun 02, 2018 9:32 am Post subject: Re: FI injector timing question |
|
|
| furgo wrote: |
Indeed. When incompatibility issues with electronic points replacement modules are mentioned, they could be due to the magnitude of that signal from the primary. That is, how much the attenuation applied on the ECU's input stage affects it. That said, the voltage pulse on the primary must be high enough to generate a spark in the secondary. So high that I'd think it would make it a moot point, though.
That (and sensor inputs + supply voltage, obviously) is all the ECU cares about to generate the injector pulses. |
Indeed! The pulse the ECU sees is substantially greater than the 12 volt battery voltage. If you have one, put a 100x probe on the scope and look at the voltage at the points. You'll see that 350 volt positive peak spikes are common, as well as 100 volt negative peaks. These are from the magnetic field of the coil collapsing and inducing a voltage in the coil primary winding (mirroring that across the secondary). After the spark at the plug ceases, the coil and condenser resonate with each other until the remaining energy is dissipated, or the points close again. The ECU sees all of this (attenuated and filtered). So there's no lack of ECU trigger voltage.
| Quote: |
| As far as I know, electronic points modules simply replace a mechanical switch by a transistor. |
It's a bit more complex than that, I suspect. The inductive pickup in the unit is good at generating transitions as the rotating magnet swings by, thus triggering the coil to fire. But something has to generate the "dwell angle" which conventional points system use to feed energy into the coil, and it's not the magnetic pickup. Circuitry inside the unit has to do this, and it's not that easy, as it has to keep the coil "on" time inversely proportional to engine speed. Some units don't bother with this, but just kill power to the coil for a fixed duration. This yields a variable "dwell" angle that is much greater at low engine speeds. A side effect of this is, unless the design has a current-limiting feature, the coil can run much hotter at low engine speeds.
A while back, I built a transistor driver unit that drives a VW coil to simulate a Model T Ford coil. It used the fixed "off" time method, but had an adjustable frequency. The coil ran too hot at low frequencies, so I had to rework the design to generate a constant "dwell angle" using counter IC's, which fixed the problem nicely.
| Quote: |
| Some of the more expensive ones have got microcontrollers that do fancy things with the primary waveform. But that directly affects the spark itself (time, strength), not the FI system, so I've never understood the claims about those modules affecting what the ECU does, other than making the signal more stable. |
One issue is that when a transistor is used to control the coil, there is no condenser to resonate with the coil primary - just the stray capacitance of the coil itself, and whatever the driver circuit contributes. Thus, the post-firing waveform to the plug may not have the normal points style resonant waveform (it is likely to be much higher in voltage and frequency and shorter in duration). Thus, the waveform seen by the ECU will reflect this. While great for the spark plug, the filtering inside the ECU may not pass this signal the same as a points generated waveform is. The only way to know is to probe the internal ECU signal after the input filtering with a scope and see.
And if the unit does anything really tricky, like multiple firing the plugs at cranking speeds, then the ECU may do things you really don't want it to...
_________________
'71 panel, now with FI
'Experience' is the ability to recognize a mistake when you're making it again - Franklin P. Jones
In theory, theory works in practice; in practice, it doesn't - William T. Harbaugh
When you're dead, you don't know you're dead. The pain is only felt by others.
Same thing happens when you're stupid. - Philippe Geluck
More VW electrical at http://telforddorr.com/ (available 9am to 9pm PST)
Last edited by telford dorr on Sat Jun 02, 2018 12:07 pm; edited 1 time in total |
|
| Back to top |
|
|
furgo
Samba Member
Joined: September 06, 2016
Posts: 944
Location: Southern Germany
|
| Posted: Sat Jun 02, 2018 11:46 am Post subject: Re: FI injector timing question |
|
|
Excellent, thanks for the more detailed insights, particularly on the electronic points modules.
_________________
'79 Westy, P22 interior, FI 2.0 l Federal, GE engine (hydraulic lifters)
Decode your M-Plate |
|
| Back to top |
|
|
fusername
Samba Member
Joined: March 15, 2006
Posts: 2897
Location: Boston MA
|
| Posted: Mon Jun 04, 2018 11:14 am Post subject: Re: FI injector timing question |
|
|
Well I only got as far as taking a few scope shots whole looking at the low side of the coil (terminal 1) and current in the injector to measure on time. I was idling at 1500 room for most of it due to a sticky throttle cable, but it shouldn't affect much. Clearly I am running rich due to excessive injector duration. At 1500 rpm and no load, if I pushed the flapper arm closed slightly and made the ecu think I had less air coming in, the idle revved up plenty. At 1500 rpm the on time was like 8ms I think, wish is clearly way to Long. I am suspecting a bad afm right now, but I don't have a spare, so I am triple checking wiring. If I find nothing I'll put points back in and see if the out put changes. Maybe the pillar stretcher gets confused by the ringing in the terminal and ends out adding extra by accident...
_________________
[email protected]
Need something custom bent up? shoot me an email, maybe we can make it work!
FORSALE: Thrust cut T4 and 1.9 main bearings
| obnoxiousblue wrote: |
| Maybe Ben Pon's ghost comes and vomits NOS stampings for your bus, but not mine! |
|
|
| Back to top |
|
|
furgo
Samba Member
Joined: September 06, 2016
Posts: 944
Location: Southern Germany
|
| Posted: Mon Jun 04, 2018 12:21 pm Post subject: Re: FI injector timing question |
|
|
| fusername wrote: |
Clearly I am running rich due to excessive injector duration. [...].
At 1500 rpm the on time was like 8ms I think, wish is clearly way to Long. I am suspecting a bad afm right now. |
Thanks for posting the follow-up.
If my math is right:
1500 rpm => 25 Hz frequency => 40 ms period
In that case, your 8 ms ON time would give a 20% duty cycle. It's difficult to tell if that's too much or too little without a baseline. The measurement will also be affected by the engine temperature. In any case I personally would not infer a defective AFM with only that data.
If you want to functionally test the AFM though, you can easily check:
1. It's voltage swing:
No need to remove the lid, it's only to illustrate in which direction and at which angle the vane opens in the video.
2. Its spring adjustment. This is something that I'd do as a last resort only. For a basic pass/fail test, checking if the vane moves freely and that it takes a not too light amount of push to move it should be enough.
Other than doing a full adjustment, the AFC manual has some additional checks that can be quickly done with a multimeter. For instance the air temperature signal might be worth testing too.
| fusername wrote: |
| Maybe the pillar stretcher gets confused by the ringing in the terminal and ends out adding extra by accident... |
I don't really understand what you mean there. Googling pillar stretcher returns... rather disturbing results.
In any case, it might also be useful to post some pictures of your scope screen to double-check the duty cycle time.
_________________
'79 Westy, P22 interior, FI 2.0 l Federal, GE engine (hydraulic lifters)
Decode your M-Plate |
|
| Back to top |
|
|
SGKent
Samba Member
Joined: October 30, 2007
Posts: 41031
Location: Citrus Heights CA (Near Sacramento)
|
| Posted: Mon Jun 04, 2018 12:46 pm Post subject: Re: FI injector timing question |
|
|
Ralph - the output of the AFM is voltage. That said, there are multiple points the voltage is taken from so there is the voltage at every one of those points and the accumulative voltage. That combination and the rates at which it changes tells the ECU what is happening and how fast.
Where are you taking your voltage readings? Not to hijack this thread but maybe you've found something missed by others here. I am wondering if it is possible to compare a voltage to RPM map at no load to mixture. I use an air to fuel gauge to set my afm, but maybe if we had a chart of voltage to mixture it might help folks quickly determine how close they are if they lack an air to fuel meter to set their AFM. The only caveat is that most voltmeters are not that accurate unless calibrated. One has to calibrate them against a voltage standard once in awhile.
http://www.voltagestandard.com/DMMCheck.html
_________________
“Most people don’t know what they’re doing, and a lot of them are really good at it.” - George Carlin |
|
| Back to top |
|
|
ac78
Samba Member
Joined: March 18, 2008
Posts: 330
Location: Nashville
|
| Posted: Tue Jun 05, 2018 9:00 am Post subject: Re: FI injector timing question |
|
|
| SGKent wrote: |
Ralph - the output of the AFM is voltage. That said, there are multiple points the voltage is taken from so there is the voltage at every one of those points and the accumulative voltage. That combination and the rates at which it changes tells the ECU what is happening and how fast.
Where are you taking your voltage readings? Not to hijack this thread but maybe you've found something missed by others here. I am wondering if it is possible to compare a voltage to RPM map at no load to mixture. I use an air to fuel gauge to set my afm, but maybe if we had a chart of voltage to mixture it might help folks quickly determine how close they are if they lack an air to fuel meter to set their AFM. The only caveat is that most voltmeters are not that accurate unless calibrated. One has to calibrate them against a voltage standard once in awhile.
http://www.voltagestandard.com/DMMCheck.html
|
I scrolled down too fast and thought I saw this 
http://www.serioustoyz.com/ItemImages/000005/ARTrockemSockemA_lg.jpeg
_________________
78 Westy 2.0l hydr. FI
VW Campmobile
For people who believe a family's
car should be it's castle ! |
|
| Back to top |
|
|
SGKent
Samba Member
Joined: October 30, 2007
Posts: 41031
Location: Citrus Heights CA (Near Sacramento)
|
| Posted: Tue Jun 05, 2018 10:12 am Post subject: Re: FI injector timing question |
|
|
my question of where he is pulling his voltage from is sincere and not a flame. He may have figured out something the rest of us missed.
_________________
“Most people don’t know what they’re doing, and a lot of them are really good at it.” - George Carlin |
|
| Back to top |
|
|
furgo
Samba Member
Joined: September 06, 2016
Posts: 944
Location: Southern Germany
|
| Posted: Tue Jun 05, 2018 10:25 am Post subject: Re: FI injector timing question |
|
|
I don't think he mentioned he was measuring AFM voltage. He said he looked at the injector pulses with a scope.
In any case I agree, without knowing exactly how the measurement was made, it's hard to answer questions.
_________________
'79 Westy, P22 interior, FI 2.0 l Federal, GE engine (hydraulic lifters)
Decode your M-Plate |
|
| Back to top |
|
|
|
