[99 Forester clutch issues]

Took that into account and pulled slowly to high rpm and very slowly and deliberately depressed clutch.

 

Makes no difference.

 

The RPM symotom is best explained as if the result of high revs is as if it's revs suck the clutch pedal down. Rev it up, then depress the clutch and the hydraulic resistance has shifted considerably towards the firewall. Deeper bite point iow.

 

It has me baffled and frustrated.

[99 Forester clutch issues]

Clutch-fork broken/cracked?

 

Have said that, I can't think of a scenario when high RPM or WOT would make a difference to clutch feel/performance.

Cluthc fork broken etc. would yield same problems all the time, just like about every other thing you could imagine clutch related. Fork is cast piece, not the stamped one, it's in very good shape. it makes no sense. I'm clutching at straws. But its 100 procent true. It's RPM influenced if not caused and baffling.

 

I wish I could film the "pedal feel". I'll try and get something on video that makes sense if I can. Untill then... Beuller?

[99 Forester clutch issues]

If you keep moderate pressure (not enough to actually press it to the floor) on the clutch pedal and it sinks to the floor then you need a clutch master cylinder. Might explain why the pedal keeps getting lower and lower as you drive.

 

I make the clutch "misbehave" on demand now. High RPM shifts basically.

 

I performed this test today before driving, after letting the car sit for a while and after some harder driving. No sinking so I guess my MC is ok. Once the clutch gets stuck after harder driving it requires pulling the pedal back up 3 to 5 times in order to pump it and it to regain enough pressure to pop back up by itself. 

 

Have you pulled the rubber dust boot off of either cylinder to check for fluid? Fluid inside the boot means the cylinder is bad.

 

It's also somewhat common for the rubber hose(s) to degrade internally and will prevent fluid from exiting the slave cylinder properly.

 

There will always be some play in the release fork/TOB mechanism. That's nothing to worry about.

 

I have pulled the dust boot off, one of the first things I did when reinstalling the slave, did again when master and slave were flushed yesterday. Dry. I can compres the slave by hand and there is no feeling of fluid along the wall of it, seems to seal well.

 

It's good to finally get an answer on the fork/TOB mechanism and it's play. Searched the internet far and wide without much result.

 

The rubber hose is a good call. I'll see if I can find one tomorrow. I had an '80 Chevy Blazer once that had sat for so long the hoses on the front brakes functioned as a one way valve, pump the pedal once and on they went, wouldn't let go though. 

 

The thing that gets under my skin most and baffles me most is that it is RPM related. At the least made worse by high RPM. If I drive like a granny the bite point drops a little but is pretty much ok, it varies more than it should be the pedal pops back up every time without fail and though the pedal effort is a bit heavy it's a functioning clutch. As SOON as I go up in RPM, wether at WOT or part throttle building up very slowly to it, the bite point starts to drop. If I build RPM slowly and shift it's not as bad but after 2 shifts this way the pedal stays dropped. 1 hard pull at WOT is enough to make it drop otherwise.

 

Once I've done this "test" the bite point is at the firewal or sometimes I can't shift anymore. I can regain some clutch actuation by pulling the pedal up, push in, pull up, push in until it pops up on it's own again, usually about 5 times and then pump it a few more and it won't be back to "normal" but the car can be driven. Take it easy and it slowly comes back to normalish as described above. 

 

I live in the city and it's all stop and go traffic most of the time. Clutch needs to work and well. Plus, it's pretty hard not do WOT pulls in 1st and 2nd when it's safe out of bordem. 

 

Anyone have a theory on the relation of RPM. So far I've heard or considerd: crankshaft play fore/aft, which to me seems pretty rediculous. The idea being that at higher RPM the crank would move and cause a deeper clutch which just doesn't compute to me.

 

I've also had it proposed to me that the slave/MC either internally leaking or having air trapped will cause my issue at higher RPM due to vibration of the drivetrain agitating whatever is causing the problem ie. air or internal seal leaks. Again, to me not so plausible.

 

I have my own theory (or hope as the fix would be the simplest).

 

That at higher RPM the fork needs more pressure to release the bearing which normally wouldn't be felt or noticed at a high rpm/WOT shift by the driver because of quicker more spirited shifting and hydraulics doing what they do when they function well, amplifying applied pressure. So that as a result at higher RPM my bite point becomes deeper because there is more pressure to overcome. Any takers? lol

 

Something else occured to me just now, my fork has no spring holding it in place (pulling it towards the radiator). The play in the fork that is normal could be deleted as pressure plate is spun faster at higher RPM thus making the distance between the pin on the slave and the clutch fork farther and moving the point at which the clutch engages/disengages....

 

Any takers on that one?

 

Most importantly: what to do with my RPM related clutch drop and pressure loss....

[99 Forester clutch issues]

If you keep moderate pressure (not enough to actually press it to the floor) on the clutch pedal and it sinks to the floor then you need a clutch master cylinder. Might explain why the pedal keeps getting lower and lower as you drive.

That's a good idea. I don't wanna just throw parts at it. Though if it sinks it could be a bad master or slave... loss of pressure.

 

When it gets caught it seems to get caught on the pivot point that the pedal is mounted to. I don't think it's the mc. It would be very coincidental. but stranger things have happened.

 

who has done a pull type clutch recently and had the bearing in and observed back and forth play? or on the fork once installed, or would like to have a look under the hood of their car for me? I've got a good half inch/1.5cm of play. I'm wondering if that is correct, although there is no obvious link between that being off and my clutch's behaviour. It all point to hydraulics

 

ideas?

[99 Forester clutch issues]

my 99 forester has the pull type clutch. It's a EDM Turbo S.

 

Replaced pressure plate, TOB and friciton disc.

 

Since then strange clutch behaviour.

 

Fork seems to move back and forth very freely and after having "rested" clutch travel in the pedal is normal and engagement etc too. After a few shifts it bites lower and lower. If I push to high RPM it seems to speed this process up. Untill the pedal gets stuck. I can pull it back up and it eventually becomes usable but it's obviously not right.

 

I did notice that the play the TOB has in the pressure plate is about 1cm or more. Seems excessive. No spring to it either which as a result means my fork doesn't push towards the firewall as I'd expect.

 

The old pressure plate and TOB have the same play so I though nothing of it untill now have issues. 

 

Fluids been flushed and bled as precaution. No change. By the looks of it with the IC off the slave simply can't engage the fork enough to pull the clutch to disengage, although the first few times it does. What gives? Is my TOB popping in and out of the pressure plate? Or has my bone dry otherwise never problem having slave gone bad?

 

Going crazy on this. need this thing for daily driving. Insight is apreceated.

 

cheers

[90's EJ20 questions...]

You need to supply 12v to the CEL output on the ECU. Circuits always sink power better than supplying power. With the test meter, put one side on the CEL pin, and the other side to a 12v source.

 

Even if there are no codes stored it will blink at a steady rate.

 

All the codes can be found on my site.

 

http://www.surrealmirage.com/subaru/engine.html

 

 

I'll try and work on getting the values tonight. Been busy....

 

no worries man, thanks for letting me know you haven't forgot, sometimes threads just die, I like to see them through...

 

I'll see e'm when you have time,

 

David

[90's EJ20 questions...]

Legacy777 how about hittin me back with those values!?

 

Here's a few pics, maybe that will inspire some more responses?

 

EJ20 fresh outta the Legacy:

 

 

Beetle transmission mounts getting beefed up for EJ20 fitment:

 

 

Engine in the Beetle, this was a test fit, it's been followed up by some cleaning and propper fitment of the overflow etc.

 

 

Then fitted a later style rear decklid etc.

 

 

It became time for inspection so I did a quick rattle can respray of the rear decklid and fitted older style lights. I don't have pics yet but have fabbed an exhaust that replicates the orgional Beetle exhaust and gives a boxery sound

 

 

Have the radiator mounted up front.

 

The whole car looks like a regular Beetle with the apropriate Suby power under the hood.

 

Here's a vid taken on the fly this is with the Beetle style exhaust, gives a little more low end in person. I intend to get some higher quality vids up on youtube soon. But it's great fun for a Beetle.

 

[90's EJ20 questions...]

I'm not an expert, but I may be able to answer two of your questions...

 

I tried to read for codes but the check engine light wire from the ECU gives a max of 0.6 volts when the IGN is put to ON and does read anything after the engine is started.

--> You might be checking voltage to a 12 volt positive? Check engine light from ECU is 12v, must put through light to ground to work. Try checking volatge to ground in test mode (black connectors hooked up), it should come on, then flash trouble codes. A test light works perfectly for this. Even if no trouble codes, it should come on when in test mode.

 

I also only have the two Green wires and only one black terminal.

--> Do you mean two green plugs and one black plug?? One of the green plugs should be a ground (black and a colored stripe (cannot remember the color)) and if you can fit the black plug in there it will work as it should. Subaru has these two plugs wired to the same ground wire. If you can't figure that out just run a jumper from black plug to ground for test mode.

 

-->That's all I can help with, like I say , I'm no expert myself.

 

Thanks, that's very helpfull actually, as I only had the + side of the two black terminals. Now I know I can just ground that and should get any readings off the Check Engine light. Does this light not usually cycle on when you turn the ignition on though? It doesn't even give me 12v then, just the 0.6v using a multimeter with the + on the check engine wire and the - grounded.

 

I got a knock sensor today, hooked it up. Yet to see what the difference is as by the time all was said and done I had to go home and leave the Bugaru behind.

[90's EJ20 questions...]

*bump*

[90's EJ20 questions...]

I'd thought of this but the wires and plugs are less than 5k miles old, it's always possible. And it's functioned just fine for the last 10k miles without the knock sensor hooked up...

 

 

Any other thoughts?

 

Seems like it was the number 4 plug wire had come loose while I had been checking vacuum hoses that day... we'll see if this holds up.

 

My knock sensor isn't hooked up.

 

With everything else in good mechanical order. What detriment will the lack of knock sensor have?

 

It retards timing as a default without it so I slight lack of performance and gas mileage is to be expected?

 

I have the stamped engine code off the motor and am wondering if anyone can help me identify the year.

 

I tried to read for codes but the check engine light wire from the ECU gives a max of 0.6 volts when the IGN is put to ON and does read anything after the engine is started.

 

I also only have the two Green wires and only one black terminal.

 

I need to know if I have a pre or post 94 engine for future reference.

 

It's an EJ20 as stamped in the front of the block I also found "62" stamped on the passenger front side.

 

The more important enginen number stamped on a pad next to the throttle body on the drivers side is:

 

470633

 

Any help? Google yielded only WRX and Sti stuff...

 

EDIT:

 

From what I can find so far, the 91-93 Subaur Legacy EJ20's DIDN't come from the factory with a knock sensor, they cam with the wiring and provision for it, but no Knock sensor. Can this be right?

 

Would this have been a quick fix to bring emissions down?

[90's EJ20 questions...]

The pistons are a larger diameter. The casting and the liner are larger diameter. It's not really a larger bore of the same block, but a larger cast cylinder. You could not bore an EJ20 to the size of EJ22 pistons.

 

That said, they are physically the same external dimensions. As long as the EJ20 is a SOHC, you should be able to bolt an EJ22 right in with the same intake and wiring.

 

Cool, I'll keep that in mind. Thought as much, good to hear conformation. What about ECU and MAF interchangeability? I have a few spare ECU's/MAF's for the EJ20...

 

I'm assuming exhaust etc. is also the same.

 

 

Spark plugs or Plug wires. Coupled with lack of knock sensor

 

I'd thought of this but the wires and plugs are less than 5k miles old, it's always possible. And it's functioned just fine for the last 10k miles without the knock sensor hooked up...

 

 

Any other thoughts?

[90's EJ20 questions...]

Welcome to the USMB.

 

I added a link for the video and moved your thread over to the transplant forum. It may get some additional comments here.

 

Thanks, sorry about that. Just a bit excited to get all the info down! Let's hope so.

 

 

 

If I did the math right, that's about 18-19 mpg. That's not horrible, but I agree, that's not great for how light the beetle is.

 

That's right on. Exactly as you said, not horrible but not great considering the Beetles weight. I have gotten as high as 35mpg on a straight 20 mile highway cruise but that was 4k rpm constant, an ideal way to get great mileage. I have talked to a guy who had an EJ20 in his Beetle who got exceedingly better mileage around town...

 

 

 

 

tube is probably for the canister purge valve. It's for emissions purposes. You can just plug that tube. Is it this tube?

 

http://main.experiencetherave.com/subaru/images/TB/DCP_1511.JPG

 

Actually that isn't it. It's a bigger diameter, proably 10mm/ or 1/4" to 3/8"

and it's directly off the side, I'll take a picture of it.

 

That pic does raise another question I have. The Suby throttle body has 2 linkages on it...does it matter which one I use and which one is for what?

 

 

 

What are you running for engine management? The stock computer? You need to have knock sensor hooked up. If you are running the stock computer and don't have the knock sensor hooked up, that is probably one of the reasons you're getting poor gas mileage. The ECU will retard timing to avoid any chance of detonation. This will hurt gas mileage and will cause you to lose power.

 

Do you have the check engine light hooked up? If not, I'd suggest wiring a bulb to that pin on the ECU, and then checking the ECU for stored codes to see what's in there that could be giving you issues. Any more info you can provide about your setup will help in diagnosing.

 

Stock Subaru ECU with custom wiring harness (basically stripped down to engine managment)

 

I came across your site by way of a different forum and was just looking over the ECU read out tech article. I'll be hooking up a check engine light asap for diagnostics.

 

The knock sensor would be located behind the intake on the drivers side of the engine in front of the intake right? I'll have to double check if it's hooked up, I know I have an extra wire in my harness that isn't hooked up...

 

 

Is it sticker on the engine or in the casting?

 

I haven't looked yet as it's tight against the firewall in the Beetle. I believe it cast. And for ECU read out purposes I need to know what year this engine is. I know I have the two green plugs and 1 black one, so I'm thinking it's a 94/93

 

Can you refer me to a good source to make sense of the casting number?

 

 

 

It sounds like the water pump isn't able to keep the required flow moving through the system and it's stagnating at that high point. How much additional hose did you add to run the radiator in the front? What size hose?

 

Depending on what you have setup, you may have to put some sort of electrical booster pump in-line to keep the system from stagnating at idle.

 

I have the radiator up front with aprox 14ft of 40mm hoses going back and forth in total.

 

I changed the thermostat yesterday to a 79c and took out the pin that sits in the bleedthru hole. Bled it by putting the nose up and car at a 45* angle. Seems to have worked so far and I have a much better flow at idle.

 

I haven't had a chance to do some realworld testing as the EJ20 has developed a driveability issue.

 

Bog under load, say high gear low RPM or WOT acceleration. So on a flat road at 1200 rpm in gear, no matter what gear, if I try to accelerate hard it shakes/bogs/sputters/missfires, if I let off the throttle and accelerate gently it's either less violent or smooth.

 

It starts a little hard but nothing dramatic for an engine with 300k kilometers, idles fine (every so often once warm shifts 30 to 50rpm sporadically). Off idle if I blip the throttle I can hear it missfire slightly, sounds like a missfire, could be a bit of a bog etc. Gradual acceleration is silky smooth.

 

It's not constant, it does this 3/4 of the time, every so often it will be fine and it just started doing this 3 days ago (symptoms were present before changing thermostat).

 

I was thinking of checking the CTS, TPS and then the ECU for codes first. Perhaps wiping the ECU because it thinks the engine has overheated frequently the past 60 miles (with the trapped airpocket reaching temps of over 120c) and I read a post somewhere that an EJ25 ran like crap after overheat and fine after resetting the ECU.

 

I was also thinking the CTS might be fried, literally from that stupid airpocket. Or that the TPS has gotten dodgy, although it is smooth at higher RPM, or if throttle is applied slowly.

 

Last thoughts are fuel pump or pressure regulator, what should the fuel pressure be and where is the regulator?

 

What is the most likely culprit in your opinion?

 

I don't have them off the top of my head now, but do have some documentation at home. The CTS uses a standard Bosch NTC calibration curve. I'll try and get you the gas gauge and CTS info when I get home.

 

Sweet, thanks I apreciate that! This way I can get a gas gauge working and rule some things out, if you could be so kind as to look up the TPS and cam sensor values that would be great.

 

 

I run 10w30. With miles that are on it, I'd probably stick with 10w30 or 10w40. The lack of backpressure shouldn't be an issue.

 

Allright man thanks, I was running a mineral 10/40 but might go to 10/30 as I do rev the hell out of it if I'm honest.

 

BTW, one final question. I can get my hands on a EJ22 with half the mileage and 20 extra horses are well felt in a Beetle. Is this just a strocked EJ20 or does it have a bigger bore? I wanna get the EJ20 running well either way because I'll still use it, just curious about the EJ22, ECU and wiring all the same?

 

Thanks A LOT for your help so far man!

[90's EJ20 questions...]

Hello all

 

I've recently put an EJ20 in my 72 VW Beetle. I thought I had a few questions but it seems to be quite a few!

 

It's a great runner although the gas mileage around town seems like it could be better.

 

about 8km to to the liter in an 800kg car. Should it be better as the car is lighter? I do run higher RPM's than in a Legacy though, so perhaps that evens out...

 

Also I have some plumbing questions.

 

There's a tube coming off the throttle body that has no hose on it, doesn't seem to be hurting anything just wondering what it is....

 

I was hoping for a picture of a legacy engine from the firewall side...

 

I also don't have a knock senor, but my harness does have the provision for it, any harm here? Pro's cons?

 

Where can I find the exact engine code (to the right of the intake on the rear of the engine standing in front of it?) if I have that number can I identify the exact engine type and what it came out of?

 

Also having an issue with a VERY stubborn air pocket getting stuck just at the CTS, it's pegged my aftermarket gauge a few times but the coolant itself is about 80 celcius but the gauge reads up to 125 celcius. Keep it at 3-4k Rpm for about 40 seconds and it pushes through and drops to a nice 80-85 celcius....very frustrating. If turn the engine off for over 30 seconds the air seems to get trapped there again as it's a higher point and at the very end of the cycle thus getting the least amount of push as it's just in front of the waterpump...

 

Anyone have values for the CTS (ohms)?

 

Values for the gas gauge in ohms? (i'm using the Suby pump and want to match to Beetle gauge)

 

Starts great cold and hot, runs well, no black smoke indicating a rich condition.

 

It has about 300k kilometers on it and is still running strong. Just want to be informed and keep my engine in the best shape it can be.

 

Any ideas on getting that stubborn pocket out? Keep in mind that I'm running coolant lines front to back on the beetle with the radiator up front so that's quite a bit of coolant for the pump to push...

 

What weight oil do these engines like? I'm running a pretty open exhaust, is lack of backpressure an isseu?

 

Thanks in advance and hope to gain some knowledge here! Love these Suby's!

 

A video of my bug, gotta make a better one, this was the day the exhaust was made just wanted to hear how it sounded: