Were Should I Spend my EG33 Dollars to Kick Ass.

[Dessertrunner]

Okay new thread,
Based on what everyone is saying we need to create a new thread.

What we need to do is set up some sort of if you make this change you will get this much power from your engine. I may not explain myself correctly but here goes.

What we can do is Bazza can create new ECU maps for the different ECU and motor upgrades using my dyno then for a small charge if someone wants to upgrade and add Know spec cams they can then contact Bazza and he can supply new ECU maps to suit. I think as regards the USA we would need to be able to borrow a Hydra or get a secondhand unit. Because Tom has done a heap of work on the the adapter and they plug direct in people will be able to just buy the unit from Tom and load Bazza maps and away they go. They don't have to cut the harness or anything.

My Engine dyno currently has a complete ECU plug and harness fro a SVX so we can just plug them straight in.

What we plan is a Upgrade path for these engines, steps that are from simple to complex.

------ What we would do is take a stock engine run with existing ECU and measure power which creates base line --------

Offer the following upgrades:- (Some of these will come with recomendations that the cooling system for example needs upgrading)

Upgrade 1.
Install the new ECU Hydra or what ever and develop maps that will do things like allow 1,000k higher red line. Improved power.

Upgrade 2.
Change Cams out to new ones, and provide new spec cams. (this is a simple bolt on change, not much work but a big improvement)

Upgrade 3.
Headers improved exhaust flow.

Upgrade 4.
New injectors & airflow meter (new map)

Upgrade 5.
Shaved head & ported.


Anyway you get the idea what we want is that every part that is added at each stage stays through the futher upgrades and that each stage gives a know power increase.

So what do you all think.
Tony

[SVXRide]

Tony,

I like it!! Note: the Hydra requires that you dump the stock MAF, as it uses a MAP.


Suggested additions:

1. IRIS always open vs IRIS always closed vs IRIS controlled by ECU/Hydra
2. Changes in intake runner path length
3. Head work - bone stock vs port matching (variant of your #5)
4. Equal length vs unequal length headers (variant of your #3)

Cheers
Bill

[TomsSVX]

Ooooo i really wanted to do this years ago.... subscribed :D

Tom

[SVXRide]

Ooooo i really wanted to do this years ago.... subscribed :D

Tom

I remember talking with you about this! :cool:

Hey, call me when you've got a few free minutes. I want to talk to you about the Hydra and control of the IRIS.

Bill

[TomsSVX]

I remember talking with you about this! :cool:

Hey, call me when you've got a few free minutes. I want to talk to you about the Hydra and control of the IRIS.

Bill

Yea if i had the cash id have been all over it. Will give u a call tonite

Tom

[Dessertrunner]

Tom how many Hydra have you sold, also what tha cost?
Tony

[TomsSVX]

Havent sold one in years. I think we sold less than 10. Hate to say it but budget builders wont spend the coin for an ems even though no matter how much u spend on a build you will get more bang for your buck with a better ems. I would have to make a call for pricing...

Tom

[Dessertrunner]

Okay so maybe we need to forget Hydra and we will just do our bit with the cams etc. We are not doing it to make money more for people to find out what works and by how much.
Will keep you posted.
Tony

[SilverSpear]

Subscribed :D

[TomsSVX]

I feel the megasquirt ems is going to be a more popular option for many if you want to try that

Tom

[icingdeath88]

I feel the megasquirt ems is going to be a more popular option for many if you want to try that

Tom

That's what I'm betting on. Appreciate the vote of confidence.

[subarvair]

I am using an eg33 in a LM Corvair project and would really enjoy being a part of this discussion, although I haven't been on this forum much in the past. You guys have so much more experience than me with this engine.
Mods are: forged pist(11-1 cr), mild cams, equal length long tube headers, ported heads, and megasquirt Ms3/3x.
Finally got it running but still have a small synch problem with the crank signal.
Mike

[SVXRide]

I am using an eg33 in a LM Corvair project and would really enjoy being a part of this discussion, although I haven't been on this forum much in the past. You guys have so much more experience than me with this engine.
Mods are: forged pist(11-1 cr), mild cams, equal length long tube headers, ported heads, and megasquirt Ms3/3x.
Finally got it running but still have a small synch problem with the crank signal.
Mike

Mike,

Welcome to the conversation!! Can you provide additional information on your build? Source for pistons and cams? Size/length of headers? How you've got the megasquirt set up? Extent to which you ported the heads?

Thanks.

Bill

[Dessertrunner]

Hi Mike welcome,
You may know this, the crank angle sensor is not at top dead centre, Bazza may jump in and tell us how much its out.
Also on my Wolf ECU we had to put a resister in the circut to clean up the signal coming back from the sensor as it was to dirty for the ECU.
Tony

[bazza]

Hi Mike welcome,
You may know this, the crank angle sensor is not at top dead centre, Bazza may jump in and tell us how much its out.
Also on my Wolf ECU we had to put a resister in the circut to clean up the signal coming back from the sensor as it was to dirty for the ECU.
Tony

It's around 60 degrees out. You can calibrate it with a timing light.

My ecu (LinkPlus G3) has in built functions to clean up the signal. Does the Wolf not have this? Also the link supplies it's own ground reference to cam and crank sensors - does the Wolf do this? Could be why you need to run a resistor.

[Dessertrunner]

I was under the impression that Steve J had trouble with his Motec reading the signal as well. Its all good at lower revs it breaks up at high revs.
Not sure what the Wolf has.

[bazza]

I was under the impression that Steve J had trouble with his Motec reading the signal as well. Its all good at lower revs it breaks up at high revs.
Not sure what the Wolf has.

Do you think it's an issue with the wolf?

[oab_au]

Most of the early units were set up for Ign points, so the input to the controller is a low impedance input, that needs a high current flow, to trigger the controller. The later ones have both low and high impedance inputs, that can use all types of trigger units.
Posted this on another thread.

The out put of the inductive sensors is really a voltage signal, not like a Hall effect out-put that can switch a transistor. They don't have the current to drive a transistor base circuit, they have to drive a high impedance input. like a 47K resistor, so that only the voltage signal gets through. This decouples the current from the input circuit, so it does not short the signal voltage to ground.

Harvey.

[icingdeath88]

I am using an eg33 in a LM Corvair project and would really enjoy being a part of this discussion, although I haven't been on this forum much in the past. You guys have so much more experience than me with this engine.
Mods are: forged pist(11-1 cr), mild cams, equal length long tube headers, ported heads, and megasquirt Ms3/3x.
Finally got it running but still have a small synch problem with the crank signal.
Mike

I have seen your posts on the megasquirt forums, and I would love your input in this thread: http://www.subaru-svx.net/forum/showthread.php?t=60860

As to the sync issues, did you try adding a resistor to the crank input to the MS?

[icingdeath88]

It's around 60 degrees out. You can calibrate it with a timing light.

My ecu (LinkPlus G3) has in built functions to clean up the signal. Does the Wolf not have this? Also the link supplies it's own ground reference to cam and crank sensors - does the Wolf do this? Could be why you need to run a resistor.

The megasquirt has both grounded at the ECU, and I think may likely require a resistor, for the reasons Harvey stated. I will have to play with it to verify. Mine does sync up at cranking speeds.

Out of curiousity, how much do the wolf and link g3 ecus cost, roughly.

[bazza]

It's interesting as most of my mates here in Aus with the EG33 tend to run decent ECU's, one runs Motec M4, I run LinkPlus G3, another runs a LinkPlus G4.

This is picture of my crank sensor setup, never had an issue, just works:
http://i265.photobucket.com/albums/ii209/bazzasti/ECU.jpg

[bazza]

Out of curiousity, how much do the wolf and link g3 ecus cost, roughly.

Mines an old one, prolly could get it for $1000-1250 thesedays. Brand new G4 or the Vipec is prolly around 2K.

Another option is the Microtech LT16S, especially if you're turbo / track/ don't need street manners (closed loop idle or closed loop fuel) or the IRIS valve. Easiest ECU on the planet to tune and can tune while on the track racing.

[Dessertrunner]

As I understand it the signal breaks up at high revs only, given that Motec had the problem I am surprised you haven't found some issues Bazza.

[bazza]

As I understand it the signal breaks up at high revs only, given that Motec had the problem I am surprised you haven't found some issues Bazza.

Why do you say that?

[Dessertrunner]

My guess is I have it wrong but I was on the understanding that the Motec was suppose to be one of the best ECU.

[bazza]

My guess is I have it wrong but I was on the understanding that the Motec was suppose to be one of the best ECU.

They are quite a good units, overpriced IMO but it's like anything else - if you don't set it up properly or don't know exactly what you're doing then it will have issues.

[Dessertrunner]

Well peoples I order the part for the 5 speed box so we can rear drive it to run the dyno. I have a hell of a time getting the power throught the engine to the dyno. I tried a drive shaft like most of the dyno guys say you can do but I had the distink impressiion it was giving the crank a bad time. We tried a fly wheel with shaft mounted to it and it was the same. Talking to Bazza he mentioned the rear drive lock for the 5 speeds so thats what we got. After that its a bit of rewelding and Programing a PLC to run the loading. Long story but I inherited a $6,000 plc that I have no use for at work so we may plug that sucker in. When can then build on the dyno about 60 different logging test point.

Also inherited a software program the we could use that will talk to the PLC. Its called Citec, worth about (10k) so if we get real keen we can set that up as well.

So lets see how we go. The only problem I have is hours in the day.

Tony

[subarvair]

I have seen your posts on the megasquirt forums, and I would love your input in this thread: http://www.subaru-svx.net/forum/showthread.php?t=60860

As to the sync issues, did you try adding a resistor to the crank input to the MS?

Having seen some of my MS posts, you know how many stupid newbie questions I have asked and what a steep learning curve the MS has been for me. A signal filtering function within the tuning software may solve the problem. I'll know in a month when I get back to the project.

[bazza]

Well peoples I order the part for the 5 speed box so we can rear drive it to run the dyno. I have a hell of a time getting the power throught the engine to the dyno. I tried a drive shaft like most of the dyno guys say you can do but I had the distink impressiion it was giving the crank a bad time. We tried a fly wheel with shaft mounted to it and it was the same. Talking to Bazza he mentioned the rear drive lock for the 5 speeds so thats what we got. After that its a bit of rewelding and Programing a PLC to run the loading. Long story but I inherited a $6,000 plc that I have no use for at work so we may plug that sucker in. When can then build on the dyno about 60 different logging test point.

Also inherited a software program the we could use that will talk to the PLC. Its called Citec, worth about (10k) so if we get real keen we can set that up as well.

So lets see how we go. The only problem I have is hours in the day.

Tony

Awesome work mate. Can't wait to do some real life testing and tuning on these things.

[subarvair]

Mike,

Welcome to the conversation!! Can you provide additional information on your build? Source for pistons and cams? Size/length of headers? How you've got the megasquirt set up? Extent to which you ported the heads?

Thanks.

Bill

Bill,
I have been away from the project for 6 months so don't have all the specs in my head. The pistons are CP and slightly over but not much. Cams were done by Delta(tacoma,wa) and were basically putting the exhaust profile(244x8.1) on the intake and grinding the exhaust to 255 x 8.5. The headers are either 2.25 or 2.5 into 3" and about 25" tubes. I hope that's not too big. The final flow report on the heads said that the straight exh was 190 cfm, the dogleg was 185 and the intake was 260 cfm at the new lifts. In my Corvair application, I only used the engine harnesses and the 4 plugs at the engine. When I left the project in May, it was running on fully sequential spark and fuel but with the crank signal problem at 1200 to 2000 rpm. I have since read where there is a signal filtering function within the tuning software that may solve the problem. If not, I'll be using some type of resister combination.

Mike

[RallyBob]

The headers are either 2.25 or 2.5 into 3" and about 25" tubes. I hope that's not too big.


Seems a bit too large. We went with 1.75" primaries and I was concerned they were too large. For most applications 1.625" is more than enough to support 250-300 whp. But given our final power and rpms the 1.75" worked out okay.

I can't imagine such short and large primary tubes unless you were making 550-600 whp and spinning up 10,000 rpms.

JMTCW...

[dynomatt]

Hi Mike,

I run MS and had ignition issues initially...there's a thread on the MS forums about it and references the fix we found.

I'm happy to give you my files if that helps?

Matt

[icingdeath88]

Hi Mike,

I run MS and had ignition issues initially...there's a thread on the MS forums about it and references the fix we found.

I'm happy to give you my files if that helps?

Matt

The VR conditioner that's used in yours I think is different than the one that's in the MS3. Do you know if there are a couple of adjustable "pots" on the board? So it may not be directly applicable, but I, and probably Mike if he searched the MS forums, was able to find the old posts by you and the guy who helped set it up initially, that cover what was done: http://www.msextra.com/forums/viewtopic.php?f=94&t=24695

Matt, do you know where along the wire the resistor was added?

Also, a question. On the MS forums one of the fixes for sync loss is adding resistors, but sometimes they are added inline with the signal wire, and sometimes, they are added across the signal and ground wires. Or sometimes both. What does this accomplish and why would that be a remedy? Harvey? Or anyone?

[bazza]

I don't know how you guys can put 2-3K+ worth of parts in your engines, take months and even years to build it all up patiently, nicely with a lot of love and then run it with something so unreliable, troublesome and nasty. Don't you think it would be better to spend that coin on the ecu and run the stock internals or at least reduce internal work etc?

The ECU is the most important part in your car, my advice is to invest in it (doesn't have to be expensive, just run something known to work without drama) or you will pay the price via by wasted time, poor dyno results or higher costs for a tuner trying to figure it out / fix the signal issues.

[icingdeath88]

I don't know how you guys can put 2-3K+ worth of parts in your engines, take months and even years to build it all up patiently, nicely with a lot of love and then run it with something so unreliable, troublesome and nasty. Don't you think it would be better to spend that coin on the ecu and run the stock internals or at least reduce internal work etc?

The ECU is the most important part in your car, my advice is to invest in it (doesn't have to be expensive, just run something known to work without drama) or you will pay the price via by wasted time, poor dyno results or higher costs for a tuner trying to figure it out / fix the signal issues.

Once setup properly, which does take a little experimenting for each new application, they run fine. Just takes a little figuring out. Right now, I can't do much work on the hardware of the car, since basically all my spare cash is tied up buying a house, so I have the time to iron the bugs out before I'll be ready to get back to the real work.

Basically, it's only such a pain in the ass the first time around, then it will be easy for the next guy to get his setup. I am just unlucky not to be that second guy. Mike too.

And then once figured out, we can be saving, what? almost a grand compared to a hydra? That's a solid amount of money that could go to parts or whatever.

[bazza]

Once setup properly, which does take a little experimenting for each new application, they run fine. Just takes a little figuring out. Right now, I can't do much work on the hardware of the car, since basically all my spare cash is tied up buying a house, so I have the time to iron the bugs out before I'll be ready to get back to the real work.

Basically, it's only such a pain in the ass the first time around, then it will be easy for the next guy to get his setup. I am just unlucky not to be that second guy. Mike too.

And then once figured out, we can be saving, what? almost a grand compared to a hydra? That's a solid amount of money that could go to parts or whatever.

I've actually tuned a mate's MS and it was the nastiest thing I've ever tuned lol, so I'd say the fun doesn't quite end after setting it up. Looking at their website they're anywhere from $550-650 then a lot of time an effort to get it working. Plus then who wants to risk a very expensive engine whilst setting up an ecu - I certainly wouldn't.

At that price there are other options - you can buy a 2nd hand Microtech on ebay for a 6 cylinder for $400 odd bucks. You can then send it to Microtech and have them set it up for the EG33 which costs about $150 and then a handset to tune it for $200. So you could be setup for $750 odd and the ecu is FAR superior to the MS. The microtech ecu actually holds the world record for the quickest EG33 / Subaru / WRX on the planet still to this day. Or you trawl the forums until you find a 2nd hand Linkplus G3 or Vipec come up for sale etc.

There are a lot of other better options, we just gotta be smart about it.

[oab_au]

The VR conditioner that's used in yours I think is different than the one that's in the MS3. Do you know if there are a couple of adjustable "pots" on the board? So it may not be directly applicable, but I, and probably Mike if he searched the MS forums, was able to find the old posts by you and the guy who helped set it up initially, that cover what was done: http://www.msextra.com/forums/viewtopic.php?f=94&t=24695

Matt, do you know where along the wire the resistor was added?

Also, a question. On the MS forums one of the fixes for sync loss is adding resistors, but sometimes they are added inline with the signal wire, and sometimes, they are added across the signal and ground wires. Or sometimes both. What does this accomplish and why would that be a remedy? Harvey? Or anyone?

The signal from these type of sensors is a low level AC sine wave. The voltage rises to peak then falls to zero, to cross to the opposite voltage. It is the zero crossing that they use to trigger the controller.
If the signal is reduced by a mismatch of the impedance, the voltage will be lower, and the zero crossing point will not be as sharp, and harder to detect.

The problems stem back to the Inductive pick-up that the engine has. These are an electrical generator, that produces a voltage wave form that depends on the circuit load, and the rate of change, rpm.
This allows the signal to be too much voltage, from the speed or a higher than normal impedance, or too low a voltage from too low a impedance. Also because they generate a such a low voltage, it is prone to interference from all the other electrical equipment. To be reliable they need the input circuitry to be matched to the inductor.

If the impedance is too low, fitting a resistor in series with the signal will prevent the voltage from being reduced. If the impedance or the voltage is too high, fitting resistor across the signal or the pot as a voltage divider, will reduce the level,

It would be better to replace the inductive unit with a Hall Effect unit. These produce a 5v square wave that is immune to interference, and does not change with speed or load. A shutter could be fitted behind the crank pulley.

Harvey.

[oab_au]

Seems a bit too large. We went with 1.75" primaries and I was concerned they were too large. For most applications 1.625" is more than enough to support 250-300 whp. But given our final power and rpms the 1.75" worked out okay.

I can't imagine such short and large primary tubes unless you were making 550-600 whp and spinning up 10,000 rpms.

JMTCW...

I agree with Bob there. The larger dia will reduce the velocity, and the 25" chimes in at about 11500 rpm.

Harvey.

[subarvair]

Seems a bit too large. We went with 1.75" primaries and I was concerned they were too large. For most applications 1.625" is more than enough to support 250-300 whp. But given our final power and rpms the 1.75" worked out okay.

I can't imagine such short and large primary tubes unless you were making 550-600 whp and spinning up 10,000 rpms.

JMTCW...

Bob,
I agree. Went back and checked my notes. What was recommended to me was 1.75" into 2.5" with 25-26" long primaries. I think the builder talked me into 2" into 2.5 to better match the opening of the elbow exh after porting. Hope I don't regret that.
Thanks for the imput.
Mike

[intellibomb]

Very interested in the cam upgrade... if that means ecu and map and injectors I guess those too... I want a little more kick in the pants very nice thread I will be watching diligently thanks for all the hard work guys!