Jump to content

Starlet Ep91 4efe turbo conversion.


Recommended Posts

  • Replies 267
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Posted Images

I've got a c64 crown gear ratio\size made to bolt to a c52 bolt pattern lsd unit needed for the drive shaft spline size. It just all needs bolting together. The crown wheel has been case hardend and ceramic beed polished. 

From what I can see this is all I need I've swapped the bearings in the bell housing for the lay shaft using the c64 bearing in the 4efe bell housing. 

I was reading some were that there might be a inner drive shaft cup that might fit the corrola tsport C63 box and also cup to the starlet drive shaft. If so then a much easier conversation to do for the starlet lads and by far the best box out there. The c64 has a longer 6th gear giving a higher topspeed on the C63 which I've been told tops out around 140mph. 

With the covid situation around the world I've lost members of my work team. as well as deadlines being moved forward to try and beat a second world wide lockdown. 

I can't see the gearbox being put in this side of Christmas. 

 

 

Edited by Sam44
Link to post
Share on other sites

Older turbo kit that came with the car. (Just alittle squabby) against new kit setup 

This was as all the sensors was being positioned and wiring installed. 

2x egt.

Intake temp repositioned.

2x nock monitoring sensors.

Wide band 02

Repositioned original O2 sensor. 

Repositioned map sensor close to throttle to achieve a 1 to throttle against map response. 

16014847704071169933414.jpg

160148485913782547603.jpg

16014849089771966451441.jpg

Pictures to follow 

Fuel pressure sensor. 

Low coolant level warning sensor. 

Battery relocation

Twin fuel pump relay setup and wiring. 

Edited by Sam44
Link to post
Share on other sites
2 hours ago, Sam44 said:

160148485913782547603.jpg

 

What happened to the Audi twin plate throttle body? Did you manage to get it fitted and tested? Any pics?

Also with the MAP sensor being relocated to the manifold have you experienced any pressure fluctuations or interference? I remember reading another build thread were the vac hose was shortened and the "filter" was removed and the owner experienced all sorts of fuelling issues because the lack of filter made the response too sensitive. 

Like the addition of EGT sensors pre and post turbo. 👍

Link to post
Share on other sites

These are old pictures I'm just sorting threw. 

They are piling up on me. 

The bay and car look totally different now. (Tidy & clean).

Yeh I'm rooting out all them errors. Like fuel burning in the exhaust giving false O2 readings and increasing egt and cylinder temps (overfueling)

Also plotting egt against turbo spool up. 

A very handy tip is to also plot it against spark kv duration and dwell gives a great picture of cylinder temps and pressure. Again for overfueling !!! Safety as thy say. 

Even the old factory fuel filter is still in this picture. 

To answer the question on the map sensor.  No .the throttle does become very responsive and can seem jerky. The only impact was the vibrations from the engine that impacted the signal. By placing it closer to the throttle body does affect the standard fueling map by increasing its base reference (more vacume present at cylinder 1 end of the manifold)

 

Edited by Sam44
Link to post
Share on other sites

It's mainly because cylinder 1 runs the hottest. the inlet design (position of cylinder 1 in relation to the throttle), and fuel rail. with cylinder 1 also the last injector on the rail. There's a possible link as to why this cylinder runs the hottest. 

Also with it running the hottest im foucing onit. 

Using cast pistons this cylinder is the 1 to watch (expansion in the bore)

Edited by Sam44
Link to post
Share on other sites
11 hours ago, Sam44 said:

To answer the question on the map sensor.  No .the throttle does become very responsive and can seem jerky. The only impact was the vibrations from the engine that impacted the signal. By placing it closer to the throttle body does affect the standard fueling map by increasing its base reference (more vacume present at cylinder 1 end of the manifold)

Weird, I remember an aussie guy saying that he was reading boost fluctuations of about 1psi (I figure from the wastegate regulating boost pressure) and this in turn was causing fuelling issues as the ecu was reacting to the wavy boost signal. He said that after reinstalling the inline "filter" (damper in reality) the fluctuations were halved and reduced the fuelling issues to an acceptable level. He had some very good data logs that showed it well.  I was going to suggest a different MAP sensor (perhaps Honda k-series) that was designed for direct mounting to manifold but no need. I'm sure that some aftermarket ecu's can electronically filter their input signals to prevent this also.

8 hours ago, Sam44 said:

It's mainly because cylinder 1 runs the hottest. the inlet design (position of cylinder 1 in relation to the throttle), and fuel rail. with cylinder 1 also the last injector on the rail. There's a possible link as to why this cylinder runs the hottest. 

The corolla intake manifold will solve the first problem.

But if you're worried about uneven fuelling along the rail have you considered feeding from both ends and returning from the middle of the rail? Seems like overkill on a short 4 cylinder really but might help.

Or feed fuel from the other end of the rail so the leanest injector is now cylinder 4? Just thinking out loud really!? 😕 Might show a difference in EGT on cyl 1 (or cause a meltdown!)

Interesting stuff though.

Link to post
Share on other sites

Yeh the corrola manifold is a great bit of kit, and definitely a fix for 1 of the main engine/cylinder power balance issues. Another interesting issues is the low kv output of the standard internal ignition coil, using uneven length leads this affects flame propergation and cylinder timing, especially when boost pressure is introduced. 

As you have mentioned if a pipe is used on the map sensor it dampens/delays the signal. This is the same on all manifolds and the longer the pipe the more this effect. if you then introduced t joints and things like piggy back ECU or bov or gauges things get very interesting and now we can see turbulence in the map signal. 

He might of had a underline issue like a leaking valve in the head it sounds to me (back pressure waves). 

It's not being worried about these things, more I'm using sensors and logs to find the errors and pinpoint them. This ultimately brings efficiency and reliability as well as being able to spot potential problems as we up the power levels. There are fuel surges in the rail and line but the dual pump system had eliminated these. I'm currently running a return to the tank system with a under chassis fuel cooler on the return line to help remove and reduce injector temps as well as charge temps, so a fuel rail like what you have suggested would require me to redesign the full system. 

 

I'll put up a picture of the Toyota fix using a antisurge check valve on the delivery side of the rail. That was introduced on all later e series engines across the range. 

 

 

IMG_20191010_234709.thumb.jpg.ad56bf810363ce0719c3ff144444e892.jpg

In the picture notice the additional valve on the delivery pipe/side of the fuel rail.

This is a inline check valve to maintain rail pressure and reduce pressure surges, a Toyota factory item. 

Look at the real length of this map sensor pipe above, it literally creates its own plenium chamber (it create a pressure delay compared to the manifold pressure). This gives a loss in throttle response and delay in the map lookup/reference tables. Lost hp.

Another very good basic mod along side this is to use a aftermarket engine earthing as well as placing an additional earthing line to the vehicles bulk head, this is were all the in vehicle earth's run to (return to battery). It's important to place a earth line to the cylinder head for the sparkplug (return to battery) and one on the inlet manifold this is were you will find the engine sensor earthing bank (return to battery). 

This sharpens up the main fueling sensors inlet map/temperature sensor to an impressive point. 

 

 

Edited by Sam44
Link to post
Share on other sites
13 hours ago, Sam44 said:

It's mainly because cylinder 1 runs the hottest. the inlet design (position of cylinder 1 in relation to the throttle), and fuel rail. with cylinder 1 also the last injector on the rail. There's a possible link as to why this cylinder runs the hottest. 

Also with it running the hottest im foucing onit. 

Using cast pistons this cylinder is the 1 to watch (expansion in the bore)

What delta have you seen across the cylinders?

Link to post
Share on other sites

The difference is anuff for a colder spark plug to be run in cylinder 1, compared to the rest. 

With this engine being a front facing exhaust manifold design, I'm getting the egt sensor as close to the exhaust head outlet as possible to get a very accurate cylinder reading. 

These are a thin mild steel manifold. 

Edited by Sam44
Link to post
Share on other sites
4 hours ago, Sam44 said:

As you have mentioned if a pipe is used on the map sensor it dampens/delays the signal. This is the same on all manifolds and the longer the pipe the more this effect. if you then introduced t joints and things like piggy back ECU or bov or gauges things get very interesting and now we can see turbulence in the map signal. 

There are fuel surges in the rail and line but the dual pump system had eliminated these. I'm currently running a return to the tank system with a under chassis fuel cooler on the return line to help remove and reduce injector temps as well as charge temps, so a fuel rail like what you have suggested would require me to redesign the full system. 

 

I'll put up a picture of the Toyota fix using a antisurge check valve on the delivery side of the rail. That was introduced on all later e series engines across the range. 

 

 

IMG_20191010_234709.thumb.jpg.ad56bf810363ce0719c3ff144444e892.jpg

In the picture notice the additional valve on the delivery pipe/side of the fuel rail.

This is a inline check valve to maintain rail pressure and reduce pressure surges, a Toyota factory item. 

I meant an actual filter in line (as fitted by toyota 23265) which has the effect of smoothing the map sensor fluctuations.

 

filter.jpg

I had a fuel pulse damper on my old weekend toy and I removed it with no ill effects when it failed and was NLA from the dealer. I understand its function and sometimes its is only fitted to reduce NVH into the cabin rather than a weird conflict that can cause lean spots or fuel delivery problems. Quite a few aftermarket FPR's have this feature built in nowadays any way.

I've checked the online Toyota parts catalogue and it is only listed for the very early 96 model 4efe and not listed for the 4efte at all. Maybe Toyota realised it wasn't necessary on a stock car.

Hope you figure out if there is actually a problem, just remember your own advice Sam, try not to get carried away!

Edited by Claymore
Link to post
Share on other sites

Yep this also damps/delays the map signal. (Map sensor filter) and the longer and wider id of the connecting pipe. This becomes a volume of stored pressure, so size (length and diameter) are a big factor here. The aim is to read direct manifold pressure. 

I was referring to the fuel injection 4 cylinder comment with the picture above. 

To answer your thoughts on the antisurge fuel rail valve. Fuel pressure pulses can be seen on a electrical fuel pressure sensor placed in the rail. Made even worse when the low output alternator when placed under high load. 

Brad ho has the fantastic MK1 GT build on this site has told me about a bolt on higher amp output alternator option off a toyota land cruiser. This will load the engine slightly more. 

As soon as I installed to dual pump system the fueling of the vehicle went instantly rich??!!. How is this explained. 

This also reduced my 372cc injector duty cycle at the pre set hp, increasing the injectors potential hp limit. Pinpointing what the factory valve is for. 

We also replaced or should I say added a high amp relay and wiring (earth return to battery) to the twin fuel pump system. This relay is operated off the standard pump wiring. We have placed 1 pump on the right side and 1 pump on the left side of the fuel tank to limit in tank surge affects whilst cornering, using none return valves in each pump to filter line (pump side of the joining Y pipe. This is to provide a fail safe should 1 pump run dry or fail. 

We have also fitted a higher capacity battery in the boot (the factory battery capacity is very low. 

The system has the desired affect and cleared all and any pressure problems. 

A side note to the UK starlet owners is the UK ep91 fuel pump is again different than the JDM turbo unit. A great upgrade for better power and a must when boosting is to replace the pump. 

The kamso fuel pump is a bargain and from my experience and from what information I can find online about it. It's a very good pump possible made by innovative motor sport as a cheap option. 

The above mentioned factory anti surge valve was first introduced on the higher hp 5e engines that ran the same pump as the 4efe starlet. 

This was my first clue as to why the valve was there. 

 

Edited by Sam44
Link to post
Share on other sites
On 10/1/2020 at 4:33 PM, RobSR said:

Yes but what is the actual difference you've seen in Deg C?

Because I'm not running 4x egt sensors on each cylinder I'm using the hottest cylinder and the spark plug to work out temperature coefficient on each cylinder. 

I'm not with the results at the minute.

Edited by Sam44
Link to post
Share on other sites
15 hours ago, Sam44 said:

Yep this also damps/delays the map signal. (Map sensor filter) and the longer and wider id of the connecting pipe. This becomes a volume of stored pressure, so size (length and diameter) are a big factor here. The aim is to read direct manifold pressure. 

Totally agree, Toyota probably chose the firewall position to put the sensor in a safer environment and allow it to be used on different manifold designs etc. This led to the the vac hose and filter which will cause a small delay. When it comes to modifying I understand that it's all up for improvement and its nice to know that moving it hasn't caused you a problem and you could use the original sensor.

I have always been a bit apprehensive about the length of vac hose required to link the manifold and built in piggy back map sensor (about 1m away in the glovebox!). Thankfully I will only be using the det3+ onboard map for boost control not fuelling. I should be able to datalog the delay between Toyota MAP and det3+ MAP signals.

That's what I like about your project, it always gives me something to think about. 👍

Link to post
Share on other sites
5 hours ago, Claymore said:

Totally agree, Toyota probably chose the firewall position to put the sensor in a safer environment and allow it to be used on different manifold designs etc. This led to the the vac hose and filter which will cause a small delay. When it comes to modifying I understand that it's all up for improvement and its nice to know that moving it hasn't caused you a problem and you could use the original sensor.

I have always been a bit apprehensive about the length of vac hose required to link the manifold and built in piggy back map sensor (about 1m away in the glovebox!). Thankfully I will only be using the det3+ onboard map for boost control not fuelling. I should be able to datalog the delay between Toyota MAP and det3+ MAP signals.

That's what I like about your project, it always gives me something to think about. 👍

Haha, yes.

I was waiting for you to bring the piggy back up.

Very very true a 1meter pipe to this sensor makes me shudder inside. As well as all the inline t offs I see on people's builds going to all sort of devices even multiple gagets. 

I had my fingures crossed. I was thinking come on lads. ho will get to the offset. Especially as alot of piggy backs do use there map sensor for engine loading. I knew you would hit it on the head. 

You can see every system no matter how small. can be tunned and sharpend up. Every little thing. 

This is a 30 year old production line car build to be cheap and soft and friendly for your gran to drive to the shop. The design engineers understand every system and whilst testing will tweak everything for the targeted customer group.

Then we get hold of them and turn them on there heads. 

Find them errors (some time purposely introduced for friendlyness) and you will find power and efficiency.

Thanks you for the information. I truly love good digestable info. 

It's silly o clock in the morning were I am, and I find my self checking the starlet owners club haha. Im really enjoying this. 

 

 

 

Edited by Sam44
Link to post
Share on other sites

It would be good to get a data sharing section up.

I'm reluctant to put data logs and data logged/calculated onto the build thread because it's not most people's cup of tea. But for the ones ho are interested it could transform the way these cars are seen as well as making 200hp+ reliable, easy reachable 4efe't, 250hp none forged 4efte

At present if I mention the 4efte to people in the JDM know. They look and laff and say wait for it to blow up. Threw Popular you tube starlet tunning channels I fully understand why this view is taken (poor little cars). 

As well as 100hp n/a. 

Edited by Sam44
Link to post
Share on other sites
2 hours ago, Sam44 said:

Haha, yes.

I was waiting for you to bring the piggy back up.

Very very true a 1meter pipe to this sensor makes me shudder inside. As well as all the inline t offs I see on people's builds going to all sort of devices even multiple gagets. 

I had my fingures crossed. I was thinking come on lads. ho will get to the offset. Especially as alot of piggy backs do use there map sensor for engine loading. I knew you would hit it on the head. 

Cheers, I get it right most times (I think!?). The rest of the time I'm still happy to learn something.

Thankfully the Emanage blue (most popular?) uses an external MAP sensor that can be placed close to the manifold with a long electrical connection so should be better than the onboard piggy's.

Currently figuring out my vac hose needs and positions, would be easier if I could use the fte manifold as it seems to have plenty of options for vac take off's rather than the fe or corolla manifold which are limited. The fe is probably easier as the aluminium looks thick enough to be drilled and tapped for hose barbs unlike the rolla mani that has insufficient wall thickness to take the threads. Weld on threaded boss's might be an option.

Edited by Claymore
Link to post
Share on other sites
  • Sam44 changed the title to Starlet Ep91 4efe turbo conversion.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...