Close



Page 3 of 4 FirstFirst 1234 LastLast
Results 51 to 75 of 92
  1. #51

    Default

    After wasting a day farting around with the PIC micro, I decided it wasn't worth the effort. It would require an external clock to match the accuracy of the SX chip which would negate it's advantage of a smaller footprint. Plus, I don't have an in-circuit programmer for the PIC so I'd have to pull the chip out any time I wanted to tweak the calibration.

    Here is the schematic for the calibrator using the SX18AC;



    And the finished circuit board;



    I neglected to consider the current draw of the in-circuit programmer when picking values for the shunt regulator. I plugged it in and the voltage drops to 3 Volts. Oh well, it works as it should when the programmer isn't plugged in. Don't really want to up-size the shunt regulator to accommodate the programmer since that will just add heat and waste power. Instead, I'll make a dongle for the programmer to power it from a separate source.

  2. #52

    Default

    After finishing the calibrator, I had an idea to incorporate an auto-calibration for the bleed flow. Simply clamp off the rubber fuel line going to the carb then click a button. The calibrator will measure the bleed flow then then save the value. Unfortunately, the SX18AC microcontroller doesn't have EEPROM that can be accessed at runtime, so it was scrapped and the PIC12F675 was used after all.

    The main loop and division routines were pretty much just a cut-and-paste from the SX code but the interrupt had to be completely revamped to work with the internal oscillator. It was tricky to get the 8-bit counter to reliably count a 16-bit value. It took me two days to track down a bug that was causing it to hang when the counter would overflow in the middle of the interrupt routine.

    Here is the tested and debugged code; http://www.gofastforless.com/junk/Fu...Calibrator.asm

    Modifying the circuit board for the new processor was pretty simple. Remove the 18-pin socket, solder in the 8-pin socket then add an LED and push-button switch.



    The LED is the pull-up for GP0. Clicking the switch starts the calibration routine. While the bleed calibration is running, GP0 is made a low output to keep the LED lit. The LED goes out when the calibration is finished.

    Here is the schematic;


  3. #53

    Default

    I got the calibrator hooked up last weekend and it seemed to work well for the first couple days. This morning, on my way to work, the reading was WAY off. The numbers looked like it wasn't subtracting the bleed flow. I suspect that the calibration routine was false triggered when the voltage dipped during startup and it saved a bleed value of zero.

    I'm going to add a timer to the program that waits for a flow signal then delays the main program from starting for a short period so that everything has time to stabilize after starting the engine. Hopefully, that will be enough to preserve the stored calibration value.

  4. #54
    FEP Super Member xctasy's Avatar
    Join Date
    Jun 2012
    Location
    Dunedin 9011, New Zealand, South Pacific
    Posts
    3,961

    Default

    I know you'll get there. Beats the pants off a door Knob!

  5. #55

    Default

    So far, this calibrator has created more problems than it's solved. It wasn't losing the calibration value, it just reads low MPG when it's cold and it reads high MPG when it's hot. My guess is that this is due to the internal oscillator drifting with temperature change. So after spending weeks rewriting the code to work with the internal oscillator, I may need to use an external crystal oscillator anyway.

    Progress is stalled for the moment because I had to tear the front of the motor off to fix some leaks and convert to a standard rotation water pump. If I can get the ECU debugged then I may do a harness swap while it's down. Too many projects, too little time.

  6. #56

    Default

    "Too many projects, too little time."

    This describes my life...
    Jim DeAngelis
    Cornucopia of Useless Knowledge
    Connoisseur of Dearborn Ferrous Oxide
    '83 GT hatch, currently under the knife
    '79 Capri 2.3L n/a, Medium Copper metallic, survivor
    (bought from MRausch82)

  7. #57

    Default

    I was looking at the specs in the PIC12F675 datasheet and there is no way that temperature change would account for the fluctuations I was seeing. It must be a fuel delivery issue.

    Turns out, there was an [inoperable] inline fuel pump under the car. Surely that poses more of a restriction than a 20 micron filter. Instead of replacing it with a straight hose, I decided to put a working 2 psi pump down there to push fuel through the flow sensor up to the mechanical pump. That should eliminate any concern of suction leaks or vapor lock.

    I drove the car to work today and it did not read super low MPG in the morning, and it did not read super high MPG in the evening. In fact, it was pretty darn consistent. Not sure yet how accurate it is but consistent is GOOD. As long as it remains consistent, the accuracy can be improved with the calibrator.

    I added some signal averaging to the calibrator code a few weeks ago and the auto bleed calibration function stopped working. It works perfectly in the simulator but will always return a zero value in the actual circuit. I got frustrated with it and just hard coded a value that was "close enough". Now that the fuel delivery issues are [hopefully] resolved, I'll have to dig into the code again and figure it out.

  8. #58

    Default

    I struggled with the auto calibration code and can't, for the life of me, figure out why it isn't working in circuit. I commented out the part of the code that sets it to zero if no flow is detected. That allowed the calibration routine to run.

    I've been super excited the last few days because the Tripminder has been functioning perfectly! The readings were rock solid, consistent, and appeared to be quite accurate. When I left work this afternoon it was functioning beautifully. Picked the kids up from daycare and went to the store, still good. However, on the way home from the store it went bonkers and was reading insanely high MPG. I ran the calibration again when I got home then pulled the chip to read it. The bleed flow was 4.115 gph, when i calibrated it three days ago it was 5.195 gph. This whole calibrator project has been based on the assumption that the flow through a fixed orifice would be constant. That is clearly not the case.

    I'm going to try it again with the solenoid that blocks the return line. It may work better now that there is a functioning pusher pump in-line.

  9. #59
    FEP Super Member xctasy's Avatar
    Join Date
    Jun 2012
    Location
    Dunedin 9011, New Zealand, South Pacific
    Posts
    3,961

    Default

    Yep, that's electronics, and why we use teams of people to help trouble shoot code. BTDT....


    The constant is influenced by operation conditions. Monitor it, use the information at hand, and you'll sort it.

  10. #60

    Default

    With the solenoid on the vapor return line and the bleed flow set to zero in the calibrator, the Tripminder is working like it should. I'm trying not to get my hopes up but I really think it's all sorted out now.

  11. #61

    Default

    Having the "pusher" pump back by the tank and the solenoid that blocks off the bleed line was the ticket to getting this thing working consistently.

    I unhooked the calibrator and plugged the FlowScan 20B fuel flow sensor strait into the Tripminder to figure out how far off it is so I'd know what calibration value to use.

    The first fill-up was a surprise. Gas pump read 8.814 gallons, Tripminder said I used 8.9 gallons. That's a 1% error! Obviously, a single fill-up isn't definitive but that is very encouraging.

  12. #62

    Default

    After three fill-ups it's at -4% error. If that's true then the FlowScan 20B is putting out 46,000 pulses per gallon. I'll keep logging it to get a more accurate number.

    It's pretty cool that an off-the-shelf flow sensor will plug into the Tripminder and read that accurately. Heck, my odometer is probably that far off. Come to think of it... in all this time I haven't checked the accuracy of odometer. The miles on the Tripminder do match the mechanical odometer but how that relates to actual miles traveled, I can't say.

  13. #63

    Default

    The good news is that after five fill-ups, the Tripminder was reading 99.4% of the pump readings. The bad news is that my Tripminder DIED shortly after that!

    The screen would occasionally blank out. Only for a few seconds and it never lost it's memory. Then one day it started blinking off and on, then went off completely. I've dissected it to try and find the problem. All the capacitors I put in it are still good. There are no visually burned or broken components. The power inverter apppears to be doing something but I'm suspicious of the voltages coming out. Without a working unit to compare it to, I'm pretty much stuck.

  14. #64
    FEP Super Member erratic50's Avatar
    Join Date
    Dec 2016
    Location
    Nebraska
    Posts
    4,575

    Default

    Bummer..... was a cool project

  15. #65
    FEP Super Member xctasy's Avatar
    Join Date
    Jun 2012
    Location
    Dunedin 9011, New Zealand, South Pacific
    Posts
    3,961

    Default

    Quote Originally Posted by mrriggs View Post
    .....
    The screen would occasionally blank out. Only for a few seconds and it never lost it's memory. Then one day it started blinking off and on, then went off completely. I've dissected it to try and find the problem. All the capacitors I put in it are still good. There are no visually burned or broken components. The power inverter apppears to be doing something but I'm suspicious of the voltages coming out. Without a working unit to compare it to, I'm pretty much stuck.

    First, Sorry man.


    On the supply voltages to the screen and the trigger, I worked on the evidence that the peak to voltage wouldn't be IVR controlled.


    May bee that wasn't right. It may have not have been designed to fed the non RMS 12 volt peak that all digital and Quartz Ford and GM and Chrysler Corp clocks seamed too.

    The time when the 1978 2.8 Injection European Granada first used the MPG Tripminder, the flow sensor input was the PWM, and the base voltage for powering up the unit was probably 12 volts. But again, that was only based on the info I found back in the mid teens.

    Thinking back through it, the VF display might be operated on 5 volts....you'll have to check under the lower volt system. The VF tubes were probably supplied by Silicon Valley, but I do know my 1978 Casio Computer Co AA battery VF tube calculator did exhibit differing low noise emitting from the VF tube under operating under differing voltage, so you could do a current throttle on its supply inwards.


    All the evidence I have seen is that the TripMinder and GM and Mopar MPG gauges were just run off the same 12 Volt supply.


    If you see Murilee Martins "The page that Launched 52 Clocks (well, 15 so far anyway)"




    http://www.autoweek.com/car-clock-of-the-week/

    The 78 Chrysler New Yorker clock uses the same display, so 12 volts should be it for any Trip Minder




    However, I've been known to Clock up. You could use the IVR 5 volt peak supply Fords other gauges used for everything 1932 till bout 1983. Of course, you'll have to ask nice for another one on the Wanted page of FEP, and please state area and where you are.


    Accuracy, you nailed it earlier.

    Quote Originally Posted by mrriggs View Post
    The mail man dropped off my Tripminder on Friday. I broke out the scope and function generator so that we can definitively answer the original question of what signal is required.

    The first thing I checked was the trigger thresholds. The fuel input takes a square wave, the High must be over 9 volts and the Low below 4 volts, so this is definitely a 0-12 volt input.

    The distance input is AC, the voltage must swing at least 0.3 volts above and below ground. This is meant to be hooked directly to the magnetic VSS sensor. Triggering it from a digital signal would require a +/- 5 volt supply.

    The standard Ford VSS puts out 8,000 pulses per mile. I just did a quick check of the distance input using the function generator to confirm that the Tripminder is looking for 8,000 pulses per mile.

    The fuel input is in fact calibrated at 48,000 pulses per gallon. To check this, I made a simple square wave generator triggered by a filament transformer. The utility 60Hz is far more stable than my old function generator. With a 60Hz signal to the fuel input it took 13 minutes 18.72 seconds to register 1 gallon. That works out to 47,923 pulses per gallon, less than 0.2% error from 48,000. I figured some of that error may have been my reflexes with the stop watch so I repeated the test with a higher value. It took 1 hour 6 minutes 33.62 seconds to reach 5 gallons, which again works out to 47,923 pulses per gallon.

    I don't have the necessary equipment to properly test the FlowScan 20B fuel flow sensor so I'm just going to hook everything up in the car and see how it goes.

    Sorry if my Yes to your will a 12 volt supply question kill your TripMinder isn't the best.




    Quote Originally Posted by mrriggs View Post
    After three fill-ups it's at -4% error. If that's true then the FlowScan 20B is putting out 46,000 pulses per gallon. I'll keep logging it to get a more accurate number.

    It's pretty cool that an off-the-shelf flow sensor will plug into the Tripminder and read that accurately. Heck, my odometer is probably that far off. Come to think of it... in all this time I haven't checked the accuracy of odometer. The miles on the Tripminder do match the mechanical odometer but how that relates to actual miles traveled, I can't say.
    First. the speedo calibration on all VR sender Fords was 100% accurate. For 1979 Ford Australia used the VDO sensor pickup, but well before that, in fact, Ford USA used the early 1969 DistroVac VR. Each allowed calibration to be accurate within 1 part per 1000 on an 8000 pulse per mile sensor. So that's basically 100% accurate.


    This compares with the following independently verified inaccuracies.

    https://www.edmunds.com/fuel-economy...s-fibbing.html

    Gas Pump MPG = OBC MPG x (1+C)

    I calculate C for each tank of fuel and also C Since New, and C Since Adjustment. Here's the C Since New for my four cars:

    2002 BMW M3 -5.0% ( -0.05 )
    2007 Chevy Cobalt: +0.27% ( +0.027 )
    2014 BMW 535i: +1.8% ( 0.018 )
    2018 BMW X3 30i: -3.6% (-0.036 )

    The 2014 BMW 535i and the 2016 BMW X3 provide for adjusting the OBC MPG. I recently did the 2014 BMW's myself. The 2018 BMW is still under warranty, so I had the dealership do it. With both of these cars, C is drifting in the positive direction (negative and drifting toward zero, or positive and drifting higher). So, I "aimed low" when adjusting the OBC MPG.

    After two tanks of fuel, the adjusted C for the 2018 BMW X3 is now -0.19% ((-0.0019). After one tank of fuel, the adjusted C for the 2014 BMW 535i is now -0.23% (-0.0023).

    So, in the case of newer BMW's the answer to "How accurate is the OBC MPG?" is "as accurate as you want to make it be."

    I wont be a judgmental dip sh!+, but there is a thing called tire wear (Tyre wear in New Zealand, England, Australia, and perhaps Canada).

    A 25.5 inch tire looses half and inch of its tread depth from birth to Kojac Bald. Total diameter loss is therefore 1 inch bald. 24.5/25.5 is a 4% variation in roll-out diameter. Gorsh some jurnos are retarded....

    Anyway, its not about spreading hate, its about the bald headed Eagle truth.

    Who loves ya babe...

  16. #66

    Default

    The fuel flow sensor signal definitely isn't responsible for my Tripminder dying. I should have elaborated further after my initial assessment of the trigger thresholds. The signal wire from the Flowscan 20B fuel flow sensor is an open-collector. That means that it is just a switch to ground, it doesn't output any voltage. The Tripminder was designed to work with an open-collector sensor, there is a 10k pull-up resistor built in.

    Below is a sketch made while probing the power inverter transformer.



    Pin-3 is filtered B+, Pin-4 goes to the switching transistor. It is switching at ~37kHz although the waveform looks odd.

    The outputs of the transformer are all rectified back into DC.

    The output from Pins 7 & 8 look normal (not knowing exactly what the voltages should be).

    Pin-5 is suspicious. Seems like a lot of effort just to make +1 Volt.

    Pin-1 is definitely a problem. Look at the polarity of that diode, the output should be greater than the -9 Volt bias.

    The output of Pin-7 goes to the cathode of the vacuum fluorescent display. Pin-1 goes to the filament of the VFD, which is why it's biased to Pin-7. No voltage to the filament means no heating of the cathode which, of course, means that the display will not light up. There is an explanation for why the display was intermittently cutting out (before dying completely) but the memory was not being reset.

    I pulled the Pin-1 diode from the board and it tested good. Probing the transformer with an oscilloscope shows that there is no positive swing to be rectified. Back to that funky waveform thing.

    There is something clamping the positive voltage swing. Possibly a shorted MOV or protection diode. This circuit is hard to trace because it used multi-layer circuit boards and there are a LOT of things packed in tightly. The switching transistor is getting really warm so it may have some kind of internal protection that is shorted out.

    Of course, Ford being Ford, they couldn't use off-the-shelf components for their electronics, NOOO! Every component, including the diodes, have Ford part numbers on them. Finding a replacement transistor will be a shot in the dark. Assuming it is that and not the transformer, or something else.

  17. #67
    FEP Super Member xctasy's Avatar
    Join Date
    Jun 2012
    Location
    Dunedin 9011, New Zealand, South Pacific
    Posts
    3,961

    Default

    Lol.




    Well, good luck with that. Your Max Planck, not Captain Plankton!

  18. #68

    Default

    Making progress. Replaced the mystery transistor with an ISL9V3040P IGBT and it worked better but causes the feedback circuit to freak out and shut everything down. It runs for about 1mS then cuts off for 2mS.



    Pin-5 is a rock solid +5 Volts. Pin-1 is about 1 Volt higher than Pin-7.

    Tried an IRL2703 MOSFET next. It's up to nearly 50% duty cycle now.



    Pin-5 is still a rock solid +5 Volts and filament voltage is nearly 2 Volts. The negative voltages all look like sawteeth on the scope because of the starting and stopping of the switching transistor.

    Got a couple more transistors on-hand that I can try.

  19. #69

    Default

    Put a 16 Volt TVS diode across the transistor and the duty cycle improved further.


  20. #70

    Default

    I had to ground a wire on my trip computer.

    I pulled it out of a 3.8 cfi and installed it in my 86 5.0 sefi car.

    It read 138mpg average before I got to freeway speeds. After grounding out one if the wires it read about 25mpg at 75mph and was pretty consistent
    2 1986 cougars (both 4 eyed and 5.0)
    1 1987 cougar

  21. #71
    FEP Super Member xctasy's Avatar
    Join Date
    Jun 2012
    Location
    Dunedin 9011, New Zealand, South Pacific
    Posts
    3,961

    Default

    Quote Originally Posted by Haystack View Post
    I had to ground a wire on my trip computer.

    I pulled it out of a 3.8 cfi and installed it in my 86 5.0 sefi car.

    It read 138mpg average before I got to freeway speeds. After grounding out one if the wires it read about 25mpg at 75mph and was pretty consistent
    Perfectly fine Hs, as just one injector is the trigger point as per post#2 on page one.

    If I'm right, one injector is bank fired asynchronously, and at 39.15 psi is 31.4 lb per hour for the 120 HP 3.8 CFi, while the regular Std Output 129-155HP are 46lb, and the 165-185HP High Output has injectors rated at 52lb. So somewhere a regular 19 lb-hr SEFI gives the right duty cycle for the old TripMinder calibration value.


    25mpg indicated ----> was most likely 15 mpg actual at 75 mph in a Sterling built Fox, although its normal to get right into the 25's at 55 mph. It might be implied that if you bump up the fuel pressure, it will read a higher fuel flow, but in closed loop, pulse widths should then be adjusted to suit. Part of the 75 mph cruise could even bee in open loop on acceleration.

    For VV7200 Variable Venturi and 2150 Motorcraft 2-bbls, Trip Minder is flow meter driven.

    For CFi, its injector driven off one injectors triggered pulse width.
    3.8, 31.4 lb/hr,
    5.0 129 to 155 hp, 42 lb/hr.
    HO 5.0 in LSC or Conti Hash 7, 52 lb/hr.

    Going to a 19 lb/hr injector, with a base on the 31.4 lb/hr, the scala factor is reduced from 31.4/19, or 25 mpg inducted is 15 mpg actual.

    If Fuel flow based on 19lbs. injectors.
    When you install larger injectors the computer will automatically adjust fuel injector pulse width to prevent a rich condition. Larger injectors = less pulse width. There is no fuel flow meter for EECIV fed Tripminders, but Ford claims that it also uses the input from the fuel gauge. My car has 24lbs. injectors and the tripminder reads higher fuel economy than I am actually getting but was reading correct before I changed the injectors
    .


    Tune in when "sheltonfilms at http://eectuning.org/forums/viewtopic.php?p=100241 says:-.

    Correction I found the right pages under chapter 14: Data Output Link (its much prettier looking in the PDF)

    The Data Output Link (DOL) provides a communication line between the EEC and
    the vehicle dashboard computer, Tripminder, for the transfer of fuel
    consumption information. The fuel flow information sent by the EEC is used
    for computation of instantaneous and average fuel economy, which is then
    displayed to the driver.
    The Tripminder requires an appropriate integer number of pulses within 100
    msec period. Therefore, within each background loop or each 100 msec period,
    whichever is shorter, the EEC sums the fuel flow and through the injectors
    output since the summing period started, converts this sum into DOL pulses,
    and outputs these pulses at a maximum frequency of 500 Hz during the
    following summation period.
    The fuel flow is converted into DOL pulses according to the following
    equation:
    DOL_COUNT = INTEGER (FUEL_SUM * 7804.19 * INJOUT) + DOL_COUNT
    The FUEL_SUM is then reduced by the amount converted into DOL_COUNTS (One
    DOL_COUNT = 1.282E-4 lbm).
    where,
    DOL_COUNT = Number of pulses to be output beginning
    in the next summation period. One
    DOL_COUNT = 1.282E-4 lbm.
    FUEL_SUM = Sum of fuel flow per injector, which was
    initiated since last summation period. It is
    updated during the Fuel PW output routine.
    7804.19 = (48000 pulses/gal)/6.15 lbm/gal, pulses/lmb.
    INJOUT = Number of injectors per output port.
    (See Fuel Strategy)
    14-2 DATA OUTPUT LINK - GXK0
    PEDD-PTOPE, FoMoCo, PROPRIETARY & CONFIDENTIAL
    FUEL_SUM = FUEL_SUM + LBMF_INJn
    LBMF_INJn = Fuel Flow per injector, calculated from
    FUELFLOWn (n= 1, 2). (See Fuel Strategy).
    ---- ---- ---- ---- ---- ----
    | | | | | | | | | | | |
    | | | | | | | | | | | |
    FUELPW ----- ----- ----- ----- ----- ------
    |<-- 1 BACKGROUND LOOP -->|----
    || |
    || |
    DOL | |
    OUTPUT --------------------------- ------------------------

  22. #72

    Default

    Alright, IRF510 MOSFET is firing constantly although there is still a bit of up-and-down.



    Adding the 16V TVS diode cleaned it up a bit and has given me the strongest voltages yet.



    Pin-1 is -26.6 Volts
    Pin-5 still a solid 5 Volts
    Pin-7 is -28.3 Volts
    Pin-8 is -36.6 Volts

    Unfortunately, the display is still not lighting up.
    It's got a negative voltage for the cathode.
    It's got a slightly less negative voltage for the filament.
    It's got a more negative voltage to pull down the grids and anodes.
    Probing the anodes with the scope shows that it is multiplexing, pins are being pulled up to +5 volts in short bursts.
    Probing the grids shows no multiplexing pull-ups happening. Hmmm... do the grids use a separate pull-up source than the anodes?

    Interestingly, the screen does light up for a brief instant when I remove B+ from the Tripminder.

  23. #73

    Default

    Quote Originally Posted by xctasy View Post
    Perfectly fine Hs, as just one injector is the trigger point as per post#2 on page one.

    If I'm right, one injector is bank fired asynchronously, and at 39.15 psi is 31.4 lb per hour for the 120 HP 3.8 CFi, while the regular Std Output 129-155HP are 46lb, and the 165-185HP High Output has injectors rated at 52lb. So somewhere a regular 19 lb-hr SEFI gives the right duty cycle for the old TripMinder calibration value.


    25mpg indicated ----> was most likely 15 mpg actual at 75 mph in a Sterling built Fox, although its normal to get right into the 25's at 55 mph. It might be implied that if you bump up the fuel pressure, it will read a higher fuel flow, but in closed loop, pulse widths should then be adjusted to suit. Part of the 75 mph cruise could even bee in open loop on acceleration.

    For VV7200 Variable Venturi and 2150 Motorcraft 2-bbls, Trip Minder is flow meter driven.

    For CFi, its injector driven off one injectors triggered pulse width.
    3.8, 31.4 lb/hr,
    5.0 129 to 155 hp, 42 lb/hr.
    HO 5.0 in LSC or Conti Hash 7, 52 lb/hr.

    Going to a 19 lb/hr injector, with a base on the 31.4 lb/hr, the scala factor is reduced from 31.4/19, or 25 mpg inducted is 15 mpg actual.

    If Fuel flow based on 19lbs. injectors. .


    Tune in when "sheltonfilms at http://eectuning.org/forums/viewtopic.php?p=100241 says:-.
    This was an Sefi 1986 low output hundred 50 horsepower motor. Also had an overdrive transmission with 273 gears. All of my 86 to 88 low output cards I've been able to get in the 28 mile per gallon range freeway if they're running right.
    2 1986 cougars (both 4 eyed and 5.0)
    1 1987 cougar

  24. #74
    FEP Super Member xctasy's Avatar
    Join Date
    Jun 2012
    Location
    Dunedin 9011, New Zealand, South Pacific
    Posts
    3,961

    Default

    Quote Originally Posted by Haystack View Post
    This was an Sefi 1986 low output hundred 50 horsepower motor. Also had an overdrive transmission with 273 gears. All of my 86 to 88 low output cards I've been able to get in the 28 mile per gallon range freeway if they're running right.
    Then it looks like the actual fuel consumption is being measured by one injector pulse. So injector pulse fuel consumption and your CFi Trip-master mileage are in synch. It'd have expected 25 to 28 US Miles per Gallon on an Aero Bird.

    Ford knew what they were doing with the SEFI. Great engine!

    Even better is how a bigger, fatter heavier Sterling Fox makes the same fuel numbers as a smaller, thinner, lighter Fox. Aero dynamics and frontal area equal Mo' Econo-Me baby-eh!

  25. #75

    Default

    Hey look... its working!



    I'm not confident that it's totally fixed. I suspect there is still an intermittent fault, it's just choosing to work for the moment. It wasn't working when switched on. After several minutes it started blinking on and off before staying on. I was able to probe the grids while it was blinking on and off and sure enough it was the loss of multiplexing to the grids that is causing it to go off.

    Hmmm... one thing I haven't looked at yet is the dimmer circuit. It's tied into the dash lights so that the display will dim when the headlights are turned on. Thinking back, I seem to remember the display going out once when I turned the headlights on. It worked after turning the lights off and back on.

Page 3 of 4 FirstFirst 1234 LastLast

Posting Permissions

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