Flare, Resistor and other wierdness - Page 4

#46 09-11-2003, 04:59 PM

1992 Subaru SVX 6-3300cc 3.3L DOHC

Service Manual Corrections - Manual Updates

NUMBER: 18-03-92

DATE: 09-16-92

APPLICABILITY:
1992 SVX

SUBJECT:
SERVICE MANUAL CORRECTIONS

The throttle sensor voltage specifications listed on pages 197 and 208 of 3-2 are incorrect.

Please make corrections to read: .5 volts closed, 4.5 volts open.

MANY Thanks to quick_SVX

In your FACE Butch

Collins

AFBeefcake · #47 09-11-2003, 06:05 PM

I just checked my TPS and it was at .456 volts and the car was Idling at around 300 RPM's
Now its at .496 volts
idling at around 600 RPM's

I when for a short drive with no flare, but the flare was not all the time to start out with. tomaro I'll see if it flares.

mbtoloczko · #48 09-11-2003, 06:10 PM

Its not just where you set the voltage. A bad TPS induces a shift flare by not smoothly increase in resistance. There can be throttle positions where the voltage reads something close to the voltage at closed throttle position.

Chicane · #49 09-11-2003, 06:27 PM

Quote:

Originally posted by mbtoloczko
Its not just where you set the voltage. A bad TPS induces a shift flare by not smoothly increase in resistance. There can be throttle positions where the voltage reads something close to the voltage at closed throttle position.

A brand spankin new TPS cured all my shift flaring.

- Rob

oab_au · #50 09-11-2003, 07:06 PM

Quote:

Originally posted by mbtoloczko
Its not just where you set the voltage. A bad TPS induces a shift flare by not smoothly increase in resistance. There can be throttle positions where the voltage reads something close to the voltage at closed throttle position.

Hi Mychailo,
Yes this is the main cause of failure, It is the vibration that gets it. Cruising at 60 MPH/100kph is where the wiper sits at the same spot on the track, for the most time. The vibration causes the wiper arm to wear through the deposited carbon track.

The early L- Jetronic fuel system were plagued by this fault in the air flow meter. Used in VW.Volvos, etc.

Harvey.

Trevor · #51 09-11-2003, 08:21 PM

Quote:

Originally posted by oab_au

Hi Mychailo,
Yes this is the main cause of failure, It is the vibration that gets it. Cruising at 60 MPH/100kph is where the wiper sits at the same spot on the track, for the most time. The vibration causes the wiper arm to wear through the deposited carbon track.

The early L- Jetronic fuel system were plagued by this fault in the air flow meter. Used in VW.Volvos, etc.

Harvey.

Greetings Mychailo,

If the wiper arm were to wear through the resistance element the TPS would go open circuit. The TPS is connected as a voltage devider and an open circuit would cause major problems. I am certain the TPS is a wire wound linear potentiometer and contact pressure and dirt or deposits is the usual problem.
We are not dealing with a radio type volume control here.

The tests I have suggested cover the best that can be done with the gear expected to be on hand. An absolute test requires a meter with a bar graph or better still an oscilliscope. I have a proposal using headphones but I have yet to remove my to TPS to experiment. I will do so ASAP.

Regards, Trevor.

#52 09-11-2003, 08:36 PM

Well, it looks like I'm nominated...
I've got a scope, (tectronix 465), a TPS of questionable integrity and, the tools to take it to bits. A report with photo's by Monday at the latest... (Like I got time to do this).

Ya can't spell geek without the double "E"
Collins "EE"

BTW What is this

supposed to mean?

mbtoloczko · #53 09-12-2003, 01:54 PM

Quote:

Originally posted by ww111
Well, it looks like I'm nominated...
I've got a scope, (tectronix 465), a TPS of questionable integrity and, the tools to take it to bits. A report with photo's by Monday at the latest... (Like I got time to do this).

Ya can't spell geek without the double "E"
Collins "EE"

BTW What is this supposed to mean?

My suggestions for analysis.

1) Measure resistance over range of throttle positions in "as-is" condition. A few repeats would be good for statistics.

2) If there is a way to squirt electronics parts cleaner into the assembly, then do that, and make more measurements, again with some repeats.

3) Making a plot of the resistance vs throttle position for each measurement run would be very cool if you have the time.

4) And then when you're all done, if you feel like it, hacksaw the thing into two pieces so we can all see how its constructed. :-))

#54 09-14-2003, 06:46 PM

Sorry guys, You'll have to wait for the TPS test and post mortem. However, the Flair is fixed, (knock on wood) yes the TPS was set to 0.500 VDC, but if you turned the engine on, it dropped to (about) 0.3 the ground was bad. That got fixed, the ABS Sensor is in and works, I swapped out the window switch and now have auto up and down again. Painted the hub covers (THANKS way much Larry III) This thing is just about FMC! (Dino sauce who guesses that one first) Next I'm putting the Blizzaks back on and working on the rust.

Thanks to everyone for their help and suggestions. LONG LIVE THE LIST!

Collins 92 pearlie

oab_au · #55 09-14-2003, 07:40 PM

Quote:

Originally posted by mbtoloczko

My suggestions for analysis.

1) Measure resistance over range of throttle positions in "as-is" condition. A few repeats would be good for statistics.

2) If there is a way to squirt electronics parts cleaner into the assembly, then do that, and make more measurements, again with some repeats.

3) Making a plot of the resistance vs throttle position for each measurement run would be very cool if you have the time.

4) And then when you're all done, if you feel like it, hacksaw the thing into two pieces so we can all see how its constructed. :-))

It is not hard to check the TPS.
You just need a volt meter, or ohm meter connected across the resistance. Remove the throttle body cover and open the throttle by the cable drum, slowly by hand, you won't have to move it too far, the fault won't be too far off idle.
Watch the needle of the meter for any sudden movement, if it does show this, just check that position slowly. The bad position will show as a sudden change in voltage or resistance, where the track has been etched out.
A good TPS should show a smooth pattern from one end to the other.

Harvey.

Bobb · #56 09-14-2003, 09:25 PM

Hi Collins, When I lost engine brakeing in my SVX, the resister was bad. I had it replaced and once again had engine braking. Take care, BOBB

Trevor · #57 09-14-2003, 09:35 PM

Quote:

Originally posted by oab_au

It is not hard to check the TPS.
You just need a volt meter, or ohm meter connected across the resistance. Remove the throttle body cover and open the throttle by the cable drum, slowly by hand, you won't have to move it too far, the fault won't be too far off idle.
Watch the needle of the meter for any sudden movement, if it does show this, just check that position slowly. The bad position will show as a sudden change in voltage or resistance, where the track has been etched out.
A good TPS should show a smooth pattern from one end to the other.

Harvey.

\

N. B. If a volt meter is used the ignition must be turned on. It could be assumed from the above that an ohm meter should be used in the same manner. This could result in an unfortunate very suddem movement of the meter. In order to measure resistance the plug should be removed so that the TPS is open circuit.

It IS very very hard to make the connections described above and this does present a real problem.

However the answers to this problem, as well full details on the above subject, have already been published here in the " How To " section.

Note : - If Subaru have gone for the very best, the resistive element is likely to be conductive plastic. This could be confused, on the basis of appearance, with a carbon element. If one is used a multifingure wiper will probably be incorporated. Expensive transducers of this type are used in industrial applications.

#58 09-15-2003, 12:07 PM

I'm talking about testing a unit out of the car. So I won't have throttle or pedal position to report. For voltage I can use either the Variable Power Supply on the bench or a signal generator... Since it's a dual trace scope I can actually compare input to output on the screen. I'm also finding out that the ECU will self adjust to the output but the TCU knows only absolute voltage. Can anyone verify this?

Collins

Trevor · #59 09-15-2003, 05:09 PM

Hi Collins,

My rough test with an ohm meter indicated that the TPS is linear and I do not think you will achieve much plotting throttle position. What you will be looking for is intermittent change in resistance as registered from the wiping contact.

I had suggested a scope or a bar graph meter so as to get over the time delay which is present in an analogue or digital instrument and which could possibly mask a slight ripple resulting from a very small fault area. Very slow movement while testing should show up expected faults. What you are doing is probably more than is in practice necessary.

The ECU will be using the analogue voltage output from the TPS in absolute terms from zero but may adjust for maximum voltage received.
The closed throttle position voltage and setting is therefore important, I would expect that the TPS as a component will have a tolerance of the order of 10% for resistance and 0.5 - 2 % for linearity.

Hope these comments may help. Trevor

P.S.

I would simply connect the TPS as a divider across a few volts ( 5 V. would be indicative of operating conditions ) and measure the output at the wiper
with low vert. gain and some sweep to provide a trace line. Moving the TPS will then show up any irregular output. Not rocket science, as you will agree !

Any possible problem with linearity would show up as out of specification resistance end to end, this being simple to confirm.

Trevor · #60 09-15-2003, 11:50 PM

Quote:

Originally posted by ww111
Sorry guys, You'll have to wait for the TPS test and post mortem. However, the Flair is fixed, (knock on wood) yes the TPS was set to 0.500 VDC, but if you turned the engine on, it dropped to (about) 0.3 the ground was bad. That got fixed, the ABS Sensor is in and works, I swapped out the window switch and now have auto up and down again. Painted the hub covers (THANKS way much Larry III) This thing is just about FMC! (Dino sauce who guesses that one first) Next I'm putting the Blizzaks back on and working on the rust.

Thanks to everyone for their help and suggestions. LONG LIVE THE LIST!

Collins 92 pearlie

Hi again Collins,

Out of interest I have just read through the thread again and now notice that in the above thread you say, Ò the ground was bad. Ò.
Is this what the shop told you they fixed ? If so this does not add up.

If you made the measurements yourself please advise what you fixed in the way of a ground connection. The more info. we can assemble the better how to fix we can record.

The TPS has no Òground " in the accepted sense of the word, i.e. at its point of mounting . In order to prevent earth loops the negative connection is returned via a separate wire in the loom to terminate within the ECU. What is more the cable is provided with a shield connected to this wire.

I will await your reply with interest. Regards, Trevor.