Jump to content
Triumph-V8

Megasquirt 2 On Triumph Tr6 With 6 Throttle Bodies

Recommended Posts

Finishing my above project it is time to report

as promised to Nick an James who supported me with

some basic maps for my MS2 project.

 

Tried to use six individual throttles on the Triumph

as it is employed on the TR6 PI but with the MS2.

Advantage should be better idle and better response

on low revs and low throttle opening with wilder cams.

 

In the engine is a 280 degree cam with 108deg seperation

and a much wilder profile than the 143 PS stock version.

 

Made some videos and thank you one more time from here!

 

First video is the engine starting from cold

first with choke and than idling with idle stabilisation.

 

Second video is the warm engine idling

and throttle response.

 

Enjoy!

 

 

 

Share this post


Link to post
Share on other sites

Sounds sweet :thumbsup:

 

Are you using manifold pressure (MAP) or throttle position (Alpha N) as the load sensing? How hard was it to map the low load areas? Small throttle movement makes a big effect.......

 

What type of idle control are you using (seems to work)

 

Cheers

 

Nick

Share this post


Link to post
Share on other sites

Hi Nick et all,

first thanks for the friendly reply.

 

The cam is a 280degree wilder cam than 143PS.

It works fine with the MS2.

Expected a lot of problems from reading the German pages for this topic.

But nothing strange happened, the mistakes I made are well known

even from me and after some trouble with the water temp sensor

and the Ignition (30kOhms on the longest cable)

I renewed what should already have been renewed and all is in limits.

 

I am using MAP and have no trouble.

I figured some time around with "general/lags"

to keep the signals in limits.

MAP at idle ist between 65 and 50 depending on the ignition advance.

It drops to 40 on higher revs and idling which seems to be very okay

to get a fine adjustment with the MAP.

Jumping MAP was expected but never a problem.

I must say I cut the MAP Accel off after it hit my setup sometimes.

In the back there must be some pulsations but with MAP lag set to 45 from 50 first

it seems to be stable and stil fast enough to respond on throttle.

 

TPS made trouble and the acceleration fuel was a pain.

Accel engages too fast without hitting the throttle

setting the gate higher causes flatspots.

Later I switched the Neon bulbs/energysaver lights in the garage off

and found it much better.....

 

So some trouble but nothing blaming on the MS

and very good results with the six butterflies

Share this post


Link to post
Share on other sites

Sorry forgot the explain of idle control.

Set four rev ranges close together

700 730 760 790 for example

At each I add 5 degrees advance

leading to the huge advance of 20 degrees at 700.

But the engine feels well with that and still runs faster

if you go even further.

Share this post


Link to post
Share on other sites

Thanks for the extra info.

 

Acceleration enrichment can be a real pain to get right (not sure either of my cars are really right), especially just off idle - you just have to keep experimenting. Different firmware versions have different strategies available. I've not had great results with any of them.

 

I've done similar with the ignition idle control but couldn't stop it "hunting". Still have extra advance in the top left corner as "ant-stall" - that works well.

 

Cheers

 

Nick

Share this post


Link to post
Share on other sites

Update1:

 

Engine got into trouble when setting up idle.

Due to the noise when syncronising at 2000rpm

I did this deep, deep in the forest.

Just dressed with some dirty cloth, no phone, no money!

Was very confident in the EFI!!!!!!!!!!!!!!!!!!!!!!!

 

The 37 D-SUB plug failed

and just from the beginning

I had the feeling it was a ugly piece of shit.

It contacted just on the last two millimetres

and at last did not contact one channel for the injectors

(which normally is contacted by two pins of the 37)

 

Anyway after some seconds of shock

and connecting to the Laptop

I could see the engine reving with high AFR after start

and than soon engage the EGO control

bringing the engine to stall.

 

Three cylinders are cut off

and the EGO tried to solve the problems

giving the three remaining much too much fuel.

 

I hit the ECU on the passenger floor

shake the connector and than it was fine again.

All in all 15 minutes work, 10 for the brain,

(maybe this could be done faster)

and five to unfix the MS2 from the passenger compartment

and work on the plug..

 

So recommendation ist to look after the plug and if it connects

later than half of the depth of the metall shield around

prepare to swap to a better product!!!

 

Update2:

 

Changed to 6 squirts simultan

to get a better idle.

Makes little difference but noticeable.

Idle is kept higher to get better throttle response

(900rpm after 820 first) when pedal hit from idle.

If going to low I needed too much acceleration fuel

to keep the things friendly

and it makes later driving worse

lowering AFR noticeable always when accelerating.

Share this post


Link to post
Share on other sites
TPS made trouble and the acceleration fuel was a pain.

 

This may seem a silly question...

How, and what have you used as a TPS? Have you simply stuck a hall effect sensor on the end of a butterfly spindle?

Share this post


Link to post
Share on other sites

Hi Gerard,

I am using a CP type injection

which has no comeing out spindle on the side of the idle screw.

 

For that reason I had to make my own spindle which is longer.

I took a 8mm steel item and at the end there is a half moon cut.

 

At the throttle body there is a aluminum plate bolted on

to accept a conventional TPS sensor from Volvo

which you can find in a dozen of other cars.

It accepts the half moon of the spindle

and is turned by spindle changing resistance.

Calibration can easily be done with the MS.

 

Thats all!

 

But the pulsations cause the spindle a little bit to swing at some revs

stock position of the throttle is one for all cylinders far away from pulsation in the front.

 

As the MS measures the speed of the resistance change and not the total amount

the systems tends at some revs to accelerate due to the flattering spindle.

 

First I will figure with the lags setup in the MS and the starting point to engage.

If all goes wrong I will put the TPS to the long spindle underneath the bodies

from where the three spindles were engaged by linkage.

But that would not look that nice :-)

Share this post


Link to post
Share on other sites

Update 3:

 

After some time of driving the EFI systems makes me feel it is not there.....

Just start the engine and drive....

Fuel consumption is about 10 litres per 100Km

and drops clearly under 8 when smooth cruising summerdays!!!

Funny, can drive beyond 450Km with the fuel in the tank.

Never had that before.......

 

Problems with the idle could be related to some noise on the signals.

With the help of Matt from DIY Autotune

I reduced noise from the (-) cable furthermore.

Idle is now stable at 800 rpm, MAP is at 58 with the 280 cam.

 

Just a little thing to improve:

The TPS signal became better and better but still not perfect.

The gate to engage accel is forced by the noise on the signal, not by my needs.

I have to stay a bit away from the noise to suppress unwanted accel fuel.

 

The reason are mainly the six individual throttles

which are a difference to the situation of the TPS sensor

and how it is normally engaged.

 

Very little degrees of throttle spindle movements from idle

create huge change in MAP enforcing accel enrich.

Unfortunately the TPS only recognizes a very small change

just a tiny bit above the noise........

 

With my MS1 this problem never occured,

I do not know the reason why the MS2 with the 3.2.1

extra code reacts that nasty on the noise.

 

Anybody else recognized the same????

Share this post


Link to post
Share on other sites

I presume you've played with the new sampling smoothing settings?

 

The huge throttles on the PI system is one of the reasons I didn't want to use them— they make the MAP and TPS leap.

Share this post


Link to post
Share on other sites

You might try using a logrithmic TPS rather than linear?

 

Also it should be possible to change the accel enrich settings to either be 100% TPS, 100% MAP or an adjustable proporiton of both. It is very hard to get right in my experience.

 

As regards interferance, are you using resistive HT leads? Resistive plugs? Do you have the 25uF capacitor on the coilpack supply?

 

Cheers

 

Nick

Share this post


Link to post
Share on other sites

Hi Nick, Hi James,

could write a book about the consequences playing with the lag/smoothing settings.

In raw you can walk from 30 to 85 percent on TPS-lag with some effect that does not disturb everything.

Below the TPS gets too lazy and above the noise fires in from time to time

what can not be controlled safely and the TPS does not react faster.

In the area from 45 to 65 you get the best results because lower lag also means you have to

set the gate to engage accel a bit lower to get the same results.

 

The mix of MAP and TPS accel is also done but the MAP signal ist not the best

due to the 280 degree cam. Quite stable at idle with 4 points sampling over 20 dregrees

It gets some interference in some higher rev areas an as it is smooth at idle

it gets wild in two areas and that has to be taken into account when setting the gate.

Too much lag here means when accelerating the cpu does not see the lower pressure in the manifold fast enough

and hesitates from jumping in the VE-table to the higher values.

You see that in a small jump from the LC-1 over AFR 16 and than come back.

 

The right amount of "driving fuel" from MAP out of the VE table is absolutely no problem.

That works perfect from the first day. Also low revs pulling of the engine is brilliant.

It pulls like a train in 4th from idle without any hickups when throttle not fully opened at 1000 rpm.

That avoids interference fron the individual runners and keeps the fuel in the right runner.

From about 2000 you can hammer the throttle down like you want and get perfect response.

I made the steps in VE-table from 60% to 95% very smooth and get a fine resolution

The LC-1 is very stable, Fuel correction stays within +/- 3%.

I am happy with that!

 

Regarding the ignition I am using the ACCEL 300+ and distributor.

But what is unacceptable for me is that I changed my engine from MS1 to MS2 only by swapping the CPU.

Before that there was no problem with the accel which might be possible that it was just below a gate

that I never recognized ist, anyway..... I did not know anything about noise neither on TPS nor on trigger.

 

Right now I have to take care of this noise and nothing else had changed.

It is only the software and the new cpu.

What I do not understand is that nobody else recognized that.

 

A logarithmic TPS would be fine but with a lot of steps from idle and lazy after 1/3 of movement.

Any idea where to get that from?

Share this post


Link to post
Share on other sites

Hello Nick, James and Gerard,

 

My name is Arjan, i'm from Holland, and currently restoring a TVR2500M.

I was delighted to find your articles and this forum; respect for your project cars! it makes me very enthusiastic to continue mine.

Because i like the challenge of doing something special, and to overcome wearing issues on distributor and carbs, i decided to convert to Megasquirt2.

I'm fairly new to electronic tuning, built my MS with care, and have it up and running on the stim after some hassle with the RPM signal.

I love to read about your experiences, especially wit the triumph straight six, and it's nice to know there are people who already tried the configuration i'm about to use.

My engine is rebuilt, and I adapted a TPS on PI throttle body's as well, along with a Ford Edis6 ignition..

I stripped an Audi 3.0 V6 for the injectors, sensors, fuel pump and fuel rail, 'but have a little doubt about using the TPS after reading your stories.

I also welded a bung in the exhaust with a used 4-wire Lambda sensor from the breakers yard, do you guys thinks it's possible to use this, or should i convert to a broadband like innovate LC1 straight away? (the MS manual doesn't mention this)

I am using a 4 wire stepper motor from a rover for Idle air control although i have got a 2 wire PWM valve from a BMW as well.

What would you recommend, and is there any science involved in wiring up the 4 wires of the stepper, as i don't have a diagram for that..?

As i'm starting from scratch with my tuning, i was wondering if i could use you experience and configurations on the tr6 engine for fuel and ignition mapping.

 

Thanks in advance, i'll try to add some pictures.

 

Arjan

Share this post


Link to post
Share on other sites

Go wideband— it will save you money in the long run.

 

For the air idle— don't get too precious about using the PWM stuff. You only need it for startup unless you have stuff like AC that you need to stabilise with it.

Share this post


Link to post
Share on other sites

Chapter ignition, "noisy map signal"

shows some more information about my IAC.

 

At the TR6 PI engine with six throttle plates

you may use the stock idle advance

lifting the throttle plates by choke cable.

 

If insisting on the IAC go and test the four wires

on the stim and blow into the IAC and see what it responds.

As far as I can see as long as you do not put the wires together

you can not do harm to the stepper control whatever configuration you choose.

Share this post


Link to post
Share on other sites

Hi Arjan,

 

Welcome! Sounds like an interesting project. We like pictures....... :)

 

It is possible to work with narrow band O2 sensor, but SO much easier with wideband and I think it is easily worth the extra money for LC-1 or similar. The fuel you save by being able to tune quickly and easily using auto-tune will mean it pays for itself almost immediately.

 

For extra air I use a simple thermoelectric valve from the Bosch K-jet system (12v only, no ECU connection) which works pretty well. Either of the other methods you mention can work just fine but I have experience of them myself.

 

Cheers

 

Nick

Share this post


Link to post
Share on other sites

I noticed something with my thermoelectric one— it isn't quite working as I expected. It seems to vary based on heat stil, despite the 12v supply.

 

The heat rises in my engine bay quite quickly, from a little over ambient up to 40-50 degrees after less than a minute.

 

I need to look at that when I get back home— currently stuck in SA.

Share this post


Link to post
Share on other sites

Thanks guys for the quick response! i'll have a go in experimenting with my stepper motor today; i would indeed rather forget the extra PWM mods to my megasquirt.

A toyota celica i've owned uses a simple thermo electric valve as well; toyota calls it idle-up circuit for cold starting, and also had a bypass valve acting on the power steering rack..

I do like the idea however for electronic idle control.. As you already mention, i always have the stock idle advance.

In some other topic is read about james' problems to get a stim functioning for the RPM signal, very recognisable, as i had the same problem with the chip on the stim; and also obtained the wrong one.. there is a difference in the stim versions..

Still some problems with uploading pictures; i'll keep trying..

Share this post


Link to post
Share on other sites

I noticed something with my thermoelectric one— it isn't quite working as I expected. It seems to vary based on heat stil, despite the 12v supply.

 

The heat rises in my engine bay quite quickly, from a little over ambient up to 40-50 degrees after less than a minute.

 

I need to look at that when I get back home— currently stuck in SA.

 

It's supposed to vary based on heat - that's what stops your revs being too high after a warm start. Ideally it wants to be bolted to the inlet manifold so it matches that temperature.

 

The Volvo 240 one I have on the PI closes a bit too quick. It's ok if you drive away straight after cold start (will idle at junctions) but cannot be left idling to warm up without stalling after about 90 seconds. Going to try a series resistor to slow it down a bit. The Golf GTI one in the Vitesse is perfect and maintains stable idle from cold to fully warm.

 

Nick

Share this post


Link to post
Share on other sites

After first start of the new 2.7 litre Triumph TR6 engine with 290 degree cam

it is time to come back and report:

 

I did the work on the car step by step, first get the German license plate,

than the fuel injection work properly and legally and now at last the engine.

 

The fun starts with the oil pan baffle made according to Kas Kastners booklet:

 

gallery_1431_55_20916.jpg

 

The block was ultrasonic tested with my old fashioned equipment:

 

gallery_1431_55_23200.jpg

 

 

and than line bored to accept cam bearings I made from bronze

 

gallery_1431_55_12390.jpg

 

The job was done on my old milling machine by employing my own tool with tungsten steel cutters set in.

 

That was the result:

gallery_1431_55_16360.jpg

 

See the cam in the little shiny sourrounding of the bronze bearing in place

 

The whole block already drilled and honed with a piston is in the next picture:

 

gallery_1431_55_56061.jpg

 

Its a famous hypereutectic piston with coated skirt and tolerance of 0.04 mm running.

 

That is the piston with the modified rod to extend working length from 5.75" to 5.85"

(it has only 17mm diametre) to give better high rev reliability.

On the right is the original County piston which is significantly heavier in 74.7mm compared to 77mm of the new one.

 

gallery_1431_55_9976.jpg

 

 

Head work was done according to Neils pictures and my own experience

leading to 37.75mm inlet and 32mm outlet valves with 7mm stem diametre.

 

Head was skimmed to give 10:1 compression ratio

gallery_1431_55_38407.jpg

 

and equipped with bronze guides and oil caps

 

gallery_1431_55_48587.jpg

 

Manifolds carefully grinded to the head and gasket and kept in position by using special

bolts with bigger diametre to fix the manifold properls (In the upper picture lower to the right)

Head already grinded and manifold ready to be worked on.

Shiny end is the valve area.

 

gallery_1431_55_31449.jpg

 

 

Last the damper was modified with a 36-1 trigger wheel and rebalanced static

before the whole rotating equipment was blanced together dynamically on my own equipment.

Hopefully I get it in the range of +/- 1 gramm what I tested with a washer hidden at the flywheel.

 

gallery_1431_55_21787.jpg

 

This is a static balancing equipment made according to the famous Ludwig Apfelbeck.

I use it for the coarse balancing of modified dampers and flywheels

 

And this is how it looks at the engine together with a fan with viscous coupling

made from STAG and ROVER parts:

gallery_1431_55_26536.jpg

Share this post


Link to post
Share on other sites

I see you solved the photo question then...... :thumbsup:

 

Nice work on the engine! Interesting solution on the small-end bush - what are the pistons normally fitted too? Your trigger wheel solution is very neat but timing is dependent on the harmonic damper bonding......

 

Nest question - how does it run? :)

 

Nick

Share this post


Link to post
Share on other sites

very nice. :)

 

Hey, Jon is alive! Long time no see...... are you still on the English Riviera? Was there earlier and it was p*ssing down.....

 

Nick

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...