Jump to content
GT6 Nick

Nick’s Green GT6 megasquirt - take 2

Recommended Posts

Hi y’all,

After an age mucking around with my GT6, it’s getting close to running again.

The short version: it has MS3-pro running the injection and wasted spark ignition. It ran for about a half hour before I left Brisbane two years ago. When I first started it, it ran beautifully. Then after pulling the engine to repair a bad oil leak, it developed a misfire, like it was running on four cylinders. Then it started blowing the ignition / fuel pump / CPU fuse.

 

Instead of driving it to my new home in Narrabri, it was trucked down. That went badly when the towie dropped it off a tilt tray, and a lot of time was spent straightening the chassis. I’ve also worked through the de-snagging list that its first drive in Brisbane threw up. Driveshaft touching the floor, screws in a brake caliper touching a disc, disintegrating urethane bushes, an oil leak behind the front engine plate...

 

And so so onto the final and hardest part, figuring out the electrical issue. I’ve pulled apart and rebuilt the wiring loom, isolating and testing each circuit. Today I managed to power up the MS3 Pro and link it to my laptop, proving that it hasn’t been fried. Huge sigh of relief. The fuse blows, however, when I connect up the coil pack and driver. The wires between them seem fine, so I figure that either the driver (a Bosch 0 227 100 203) or the coil pack, a 3-coil wasted spark unit, has been damaged. The driver seems more fragile, but if it has died (maybe not all at once, explaining the poor running) then it’s likely to be because either it isn’t compatible with the coil, or I had the settings wrong.

 

What coil pack and driver combinations have other megasquirters used?

Share this post


Link to post
Share on other sites

Hello Nick

                  This is the type of coil pack I am using on Spitty as the driver is built in and they are cheap and common! (the copy ones not paying Bosch prices!)

https://www.ebay.co.uk/itm/NEW-AUDI-SEAT-SKODA-VW-IGNITION-COIL-PACK-032905106B-032905106E-032905106-UK/271395612357?epid=1144571949&hash=item3f307066c5:g:Z-4AAOSwxH1T1pG9

I am using Microsquirt V3 and you can wire direct from ECU to coil pack(save one more connector!)

You can see it on the right.

Roger

DSC06724.JPG

Share this post


Link to post
Share on other sites

Roger - that's one advantage to the MS V3, it has internal coil drivers I think?

Craig - Wow, 8 coils and leads for so little! Hmmmm. I've ordered a Quadspark ignition driver from Autosport labs, who make the MS3-Pro, for less than that, so that I could keep my coil and very expensive leads. If I'd seen that add first I might have gone the 6-coil route. If the Quadspark solution doesn't work, I'll use your idea.

Such a relief though, to power up the GT6's megasquirt CPU on Saturday and see that it isn't fried, and it could see the engine sensors. No brain damage!

Share this post


Link to post
Share on other sites

We have made fire. And noise. And smoke. But mostly noise. :banana:

Yesterday I substituted a Quadspark coil control module for the Bosch coil controller. With power to the Quadspark and coil, but the ignition system not connected to the MS3Pro, it was fine. But as soon as the MS3Pro was connected to it, "Pop" went the ignition fuse. Darn. Head scratching. Hmmm, it's a 5A fuse. Why do I have that there? Surely that's not enough to power the coil? Plug in a 20A fuse and... no pop. Phew. Ten seconds later... "Pop". Darn. 20A should be enough though, I can't imagine each coil draws more than that? Maybe the MS3Pro powers the coils before the engine starts cranking? So I delved into its programming and checked things like dwell and spark duration. I fitted a 30 amp fuse (blow that yer bugger!), which is probably as much juice as the wiring can handle anyway.

I reset a few other things too. I'd changed the timing wheel, so played around with the degree figure in the setup menu. Time to hit the starter button. No blown fuse but no sign of starting either. Hmmm, it wasn't recognising the timing wheel. Changed to falling edge and now we have a crank signal. Hit the button again - kaboom! 

After a few misfires as two year old petrol was flushed out of the fuel line, she started and idled. A few more tweaks to spark duration and base timing saw it hold a steady idle, but stall when the throttle was opened. Since she last ran in 2016 I'd fitted a MAP sensor taking vacuum from all six runners, so I changed the acceleration setting on the laptop to ITB (independent throttle body), which uses both throttle sensor and MAP to determine engine load. Much better, no stalling. And so off down the road for a test drive.

Observations

  1. The driveshaft's rubbing on something intermittently. I thought I'd trimmed enough floor away, but maybe it's rubbing on the handbrake mechanism.
  2. The throttle return springs could be snappier, as the revs take a couple of seconds to fall after blipping the throttle.
  3. The motor's noisy, possibly because of wide roller rocker gaps.
  4. Darn she's long-legged, compared to the Herald. Second gear just goes on giving.
  5. The brakes are heavy compared to the Herald. They're Canley vented discs and four-pot calipers, and a small booster. 
  6. The fan is so close to the crank pickup that it's creating interference. There's probably no simple/cheap way of fixing that. 
  7. The oil leak from behind the front plate is fixed.
  8. Once it's running properly, this thing is going to be a rocket ship.

Engine.JPG

Fan clearance.JPG

Noise.MOV

Edited by GT6 Nick
Edited for spelling and coherency

Share this post


Link to post
Share on other sites

That's a gloriously technical looking engine bay!  Well done on getting it running.  Not sure why it needs a 30A fuse when the engine isn't even running..........

I forget which cam you have  - are the wide tappets due to that or roller rockers?

2nd gear...... is that standard GT6 gearing with a 3.54 Subaru?  Should be good for about 70 in second - with fairly bright 0 -60 times possible.

Ar you talking about mechanical or electrical interference between trigger wheel and fan?

Nick

Share this post


Link to post
Share on other sites

From memory, the valve clearances are due to the roller rockers. I'll close them up a bit and hopefully that will quieten the din. The camshaft is Gareth's GTT43 courtesy of Andy Thompson.

The gearing? Between the 3.54 Subaru diff and the W58 Supra gear ratios, 2nd should be good for nearly 100km/h. Compared to my Herald, or the diesels I'm used to driving, that's amazing. My motorbike can't do that!

Unfortunately the fan's interfering electrically with the crank sensor. With the engine not running but the ignition on and fan running, the injectors clatter away erratically. The best solution will be to fit a different radiator that doesn't have the fan anywhere near the crank sensor.

By the way, I just found an episode of Car SOS where they rebuild a GT6. Nice!

Share this post


Link to post
Share on other sites
1 hour ago, GT6 Nick said:

Unfortunately the fan's interfering electrically with the crank sensor. With the engine not running but the ignition on and fan running, the injectors clatter away erratically. The best solution will be to fit a different radiator that doesn't have the fan anywhere near the crank sensor.

By the way, I just found an episode of Car SOS where they rebuild a GT6. Nice!

I take it there's no chance of squeezing a fan in front of the rad?

Share this post


Link to post
Share on other sites

There’s room, but I’m worried about blocking airflow. Still, if the choice is that or a misfire, it’s worth a try. More tinkering tomorrow then.

Edited by GT6 Nick

Share this post


Link to post
Share on other sites

20 uF capacitor (suppressor) between fan supply and ground might be worth a try.  Is the wire to the VR sensor shielded?

VR sensor itself is rather close to the fan motor. A metal sleeve over the body of it (earthed) also might help.

Nick

Share this post


Link to post
Share on other sites
8 hours ago, GT6 Nick said:

 

Unfortunately the fan's interfering electrically with the crank sensor. With the engine not running but the ignition on and fan running, the injectors clatter away erratically.

Hm wonder if I'm going to have same problem with MJ sensor?

Alan

5ac11c43ca213_AcusumpTriggerWheel006(Large).thumb.jpg.a1c004076daa545d8c1a780b48f8c1b8.jpg

 

Share this post


Link to post
Share on other sites

Once it's running properly, this thing is going to be a rocket ship

 

Unfortunately so are some of the white van men, had a drag race with one the other week trying to over take me in a stupid place, i won in the end but afterwards i felt disappointed as thought the GT6 was more rapid.

Share this post


Link to post
Share on other sites

@ Alan, If the sensor is right next to it...... maybe.  I don't, but my sensor certainly isn't as close as Nick's.  Can't tell from the pic whether yours is right along side it or just above it?

@RR, wouldn't worry about it.  Very few things on the road are as quick as a white van driven by a wild eyed lunatic....  I remember an epic duel with a newish Renault Traffic van somewhere in the approaches to the Alps (heading for Col De Bonnett IIRC) very early in the morning on the 2013 10CR.  We'd been driving all night and came upon this van just as it was getting light.  He wasn't going that quick initially and we nearly got past past him before he realised he was about to be overtaken by a couple Rosbifs in a funny old car, at which point he put his foot down.  We chased him for about 60kms along some very wriggly roads.  He couldn't get away and we couldn't get past.  We could see he was trying pretty hard as he was lifting rear wheels in corners and sliding around all over the road.  We were pretty much flat out too.  If he was delivering croissants, I reckon they arrived in a battered state.

We had a similar duel with a 1300 Twingo in the Pyrenean foothills in 2009.........

Nick

Share this post


Link to post
Share on other sites

You could revise the position of the sensor, put it as far away from the fan as possible.   It ca be anywhere around the trigger wheel, as along as the missing tooth is 60 degrees ahead when the crank is at TDC.

As documented elsewhere, my Megajolt installation ad it's problems, but didn't suffer from fan interference despite a similar fan to yours, perhaps because the sensor is on the opposite side of the crank.

John

Share this post


Link to post
Share on other sites

Just a thought - its possible the interference from the fan motor may not be electrical - it could be magnetic in which case suppression capacitors and earthing won’t help. The pickup is very sensitive to rapidly moving magnetic fields and the fan motor has electro-magnets whirling around inside. Screening that may prove difficult. Ideally you would need a soft iron plate between the two, though it might be enough if the sensor body was enclosed in a steel box or tube. (ali won’t work, it has to be ferrous).

Rob

Share this post


Link to post
Share on other sites

I’m pretty sure that the interference is magnetic, as Rob says, caused by the magnets in the motor. I could move the crank sensor, but sitting it above the pulley means that I can replace the fan belt without disturbing the sensor. So, I’ll move the fan. In the short term though, it doesn’t seem to be causing poor running, and the effect is only noticeable when the engine isn’t running.

Today I checked the valve clearances and inserted spacers between the body and chassis to provide more clearance between the driveshaft and the plate under the handbrake lever. Something else I found today is that the AP calipers have small screws retaining wear plates, and that those screws can loosen, hitting the disc. Noisy!

De-snagging continues...

Share this post


Link to post
Share on other sites

Rob, unless the outer shell of the fan motor is aluminium, won't that provide some sheilding?   Or just not enough?

Intrigued by the problem - possible problem - I found this article online, about using 'magnetic shielding foils' to inhibit magentic field interference.     https://interferencetechnology.com/magnetic-shielding-basics/#

A leading manufacturer is "MuMetal" and this nickel alloy is often so referred to, but eBay and Amazon suppliers of any such foil only offer it in much larger quantities than you would need , Nick, and at some cost!

JOhn

Share this post


Link to post
Share on other sites

Aluminium can help for higher-frequency fields if it is thick enough and forms a complete Faraday shield John, but this is a low-frequency thing and here you really need a 'magnetic' material. Mu-metal is a pretty expensive option but is about the best. µ is the permeability factor and this stuff has a very high figure - hence the name. It has been used for decades in high-quality electronics and telecoms applications as your link explains but its really overkill for things like this.

It's possible a bit of steel conduit tube around the body of the sensor would do the trick, or even a flat steel plate between it and the motor though that wouldn't do much for the efficiency of the cooling fan.

Share this post


Link to post
Share on other sites

thanks, Rob!    Should a magnetic shield be earthed, like one for radiated interference?

John

Share this post


Link to post
Share on other sites

No John - earthing of a magnetic screen isn't necessary. What is happening is that a varying field induces currents (eddy current) in the metal and these currents act as small electro-magnets to effectively cancel out that applied field for anything within the screen. It all happens within the thickness of the metal.

Share this post


Link to post
Share on other sites

Thanks, Rob! Learn something every day!

And glad the whole thing isn't necessray, Alan, packing the fan with foil sounds very fiddly!

John

Share this post


Link to post
Share on other sites
19 hours ago, RedRooster said:

Once it's running properly, this thing is going to be a rocket ship

 

Unfortunately so are some of the white van men, had a drag race with one the other week trying to over take me in a stupid place, i won in the end but afterwards i felt disappointed as thought the GT6 was more rapid.

Not all white vans are made the same...

 

Share this post


Link to post
Share on other sites

De-snagging continues, mostly getting it running properly and eliminating nasty noises. I've ordered new, lower gearbox mounts from Carbuilder Solutions to prevent the driveshaft rubbing on the floor. The rear mudflaps have been trimmed - those 6" wide wheels fill the arches! I thought I'd cured the front brake squeals, but they're persistent. The reason is that the 4-pot calipers don't sit centrally on the discs, because the caliper mounting brackets are about a mm too big and need skimming. It won't be hard to do, it's just a pain to have to strip the front suspension down again and get them machined.

I had the wheels aligned today: at the front it has 3 degrees castor, zero camber (the factory specifies 2 degrees positive!) and 0.1mm of toe in. At the rear it has 1mm of toe in. I would have preferred a degree or so negative camber, but it feels planted and tracks perfectly as is. Considering that two years ago it had a chassis like a banana, I'll call that a win. We went for a cruise just to be sure.

The engine's a grumpy bugger though. I think the idle's lumpy because of the camshaft, and because I haven't balanced the throttle bodies yet. However, I suspect that the snatching, misfiring and spitting back are because the engine's running on Megasquirt's 'getting started' map, and I need to start mapping it properly. Engine mapping is a dark art to me so far. Anyone got a suitable 2.5L map I can shamelessly copy?

Test ride.jpg

Edited by GT6 Nick
Edited to uncorrect Autocorrect

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

×