Track and Canyon 3000GT: excessive Toe In/Out under braking/accel

Hi everyone, long time fan of GRM, but unfortunetly it took seeking knowledge beyond my own to get me to sign up on the forums...

I've been spending an irrational amount of time and money over the last decade turning Mitsubishi 3000GT's into something...well...they should never really be turned into within reason..

(With the exception of the raw grip from the 275's and 2700lb weight, along with the high speed stability of the long wheel base, I know I could put $50K into this 28 year old mitsubishi, and it will never hold a candle to my $4,500 95% stock Mazda in the corners)

My most recent adventure was installing an aftermarket tubular front subframe, with provided LCA's.

Since installing I've been fighting problems with toe angles... The toe gets all kinds of out of shape under acceleration and breaking, even after a little bit of mild street use... What makes the problem even worse, it seems to almost get stuck that way until I come to a stop, and go lock to lock with the steering. I installed a fresh Reman rack, with new inners and outers, and have not been able to come up with any thing while troubleshooting that the rack is faulty.

I'm trying to figure out if maybe its just bad geometry, or an obvious faulty design I cannot see... I've spent about 30-40 hours just this week working on it every day, trying different things, constantly test driving it, re-aligning the car,checking nuts, and bolts for proper torque, increasing that torque, prying on things to check for slop or play, or anything left loose. Tried disconnecting the sway bar, pulled steering and tore it apart to see if there were any obvious problems, etc.

here are some pictures for reference, these are not of mine, but they show a better look then I can provide, just unfinished..

I do want to point out, while not shown well in the pictures, the rear most mounting points of the LCA's is moved forward about 2-3 inches from the stock design, wondering if this is allowing the control arm to pivot more then normal maybe? The Geometry is supposed to be based on all OEM mounting locations except for that, one mounting point, but it was also built primarily for drag racing, not road racing. I'm sure the geometry is not that great, maybe even trashed, but I was hoping I could make it suitable for the purpose of the car, and perhaps modify it down the road.

I can weld, and do moderate fabrication, but before i got too carried away, I wanted to reach out to people who may have a more knowledge then me on the subject.

Thank you in advanced, and I'm glad to be here!

~ Chris

First of all, THIS is perfect GRM first post.  Other newbs, take note! 

blackmount1 said:

I've been spending an irrational amount of time and money over the last decade turning Mitsubishi 3000GT's into something...well...they should never really be turned into within reason..

Regarding the toe isses, can you be more descriptive of the symptoms? 

How do you know there are toe issues? 

With one side, or both?

Is your wheel center moving?  

Does it pull one way?

Probably not the issue, but one time the strut bolts on our Lemons car came loose and created havoc.  Everything that you would think would control toe was fine, but the camber moving created a corresponding toe change.  Thankfully we caught this right away.  Good reminder to mark your bolts and check your torque before each session!

In order to change the total amount of toe, the positions of the outer tie rod ends needs to move relative to the hub.  I assume it's a fairly standard strut upright that the tie rods attach to?

I would look closely at the bushings and the attachment points for the control arms and steering rack onto the subframe.  Are they close fits to the bolts, or are they slotted?  Is it possible that the two parts can slide relative to each other under load?

Got a gopro or similar? I think mounting one looking at the front lower control arm mounting location and doing some hard driving might be informative or at least rule out what i think is most likely.  I think the whole crossmember is flexing. If you have a decent way of measuring between the two forward-most tubes of the crossmember, see if that measurement changes. Alternatively and probably easier, measure distances from each tire to fender front and back and see if those change after hard driving. This probably needs to be accurate <1/4", preferably way more accurate, so you might have to put some thought into making sure the way you are measuring is very repeatable. 

If you think about how this part is typically called a K member and look at yours, it is more like a capital I. 

ProDarwin said:

First of all, THIS is perfect GRM first post.  Other newbs, take note! 

blackmount1 said:

I've been spending an irrational amount of time and money over the last decade turning Mitsubishi 3000GT's into something...well...they should never really be turned into within reason..

 

Regarding the toe isses, can you be more descriptive of the symptoms? 

How do you know there are toe issues? 

With one side, or both?

Is your wheel center moving?  

Does it pull one way?

 

Probably not the issue, but one time the strut bolts on our Lemons car came loose and created havoc.  Everything that you would think would control toe was fine, but the camber moving created a corresponding toe change.  Thankfully we caught this right away.  Good reminder to mark your bolts and check your torque before each session!

Thank you. I've checked any and all hardware, I even swapped in a few different shock/spring and coilover assemblies I have laying around to see if maybe it was the nature of the beast. Such as to take advantage of it the car needed to be slammed, and stiff with lots of static camber, or closer to an OEM Ride height with more travel available and a somewhat existent camber curve... unfortunately the difference between the two was minor, but it was happiest low, with alot of static camber and very limited available and physical travel.. maybe just because things couldn't move...

Aside from physically watching it happen as I may or may not have hung my head out the window and accelerated hard followed by braking hard to see the Toe change an excessive amount.. When it goes way out the steering becomes very sensitive, wonders all over the place.. (it usually goes further out, then it does in) It can get so bad, it will a constant squealing, at even 10-15mph, and intensify if I gently apply the brakes, even with the absolutely tortured 120,000 miles OEM bushings this car did not have anything like this issue before the subframe was installed....

it seems to happen to both sides, there is also quite a bit of creaking like a bushing or control arm is moving, the control arm bushings are poly and the brackets are a snug fit around the control arms bushings when bolting to the subframe.., but the brackets are also quite rigid, wondering if maybe I cannot get enough clamping force on the bushings which allows them to move and, and possibly cause the control arms to have a slight pivot like motion, could explain why the toe seems like it gets stuck out of spect sometimes until I go lock to lock with the steering

.... I have stuck a prybar inbetween the arm and subframe to pry, and it does move slightly with a decent amount of force behind it, but I feel like it is not enough to cause my issue, but I could be completely wrong, I plan to focus more on the control arms tomorrow.. 

codrus (Forum Supporter) said:

In order to change the total amount of toe, the positions of the outer tie rod ends needs to move relative to the hub.  I assume it's a fairly standard strut upright that the tie rods attach to?

I would look closely at the bushings and the attachment points for the control arms and steering rack onto the subframe.  Are they close fits to the bolts, or are they slotted?  Is it possible that the two parts can slide relative to each other under load?

 

They are not slotted, and they do have room to move, and rely primarily on clamping force to hold them in place.. I plan on taking a closer look at the control arms tomorrow... I think it was a missed oppertunity not to slot them and use eccentric bolts to allow further adjustment... or just heim joints with a threaded bar.. but again, this was built primarily for drag racing.

I think Vigo nailed it. I am pretty sure your caster is changing as well as the wheel base each time you experience the change. When you stop the car and exercise the steering rack with no cornering loads the (formerly K) cross member gets bent back close to its beginning shape. This unit was never intended to take any cornering loads, let alone the high ones you can make with sticky tires. It will likely fail from fatigue too. Sell it to a drag racer soon.

blackmount1 said:

.... I have stuck a prybar inbetween the arm and subframe to pry, and it does move slightly with a decent amount of force behind it, but I feel like it is not enough to cause my issue, but I could be completely wrong, I plan to focus more on the control arms tomorrow.. 

Any idea how much force you are putting on it at the end of the prybar?  Typical braking forces are going to be >1g on street tires and even at a 2700lb curb weight, you are talking a substantial force on that cradle.

This is obviously very concerning, but what is really concerning to me is that it stays in the messed up toe condition until you move the steering around to reset it.

Nothing seems to be connecting the front mountings of the LCA's together on that new tubular frame. Under loads, the whole thing might move and those 2 mounting points might end up closer or farther from each other, changing the toe.

I'd say measure the space between the LCA's, go for a run and when you get the "special condition", measure again to see if it has changed.

That front subframe is probably more flexible than a go-kart chassis. I bet you are plastically deforming it - bending the whole thing - and it is taking a "set" at the toe angle than you're noticing. You will want many more tubes in there to triangulate it properly so that it doesn't bend.

Thanks everyone, I've been considering throwing the stock subframe and control arms back in the car with some fresh bushings, but wasn't quite ready to write this subframe off as a complete loss yet, perhaps just a victim of shiny-car-part-syndrome.. I did have a bit of a unsettling feeling once I recieved it and compared it to the stock one..

I've got a few ideas I am going to try tomorrow before pulling it all out with the help of a second person which was something i had not had available all week 

In reply to blackmount1 :

I'm no engineer, but the fact there's nothing directly preventing the front control arm mounts from flexing inward/outward seems like a potential source for problems. 

blackmount1 said:

 

 

 

 

 

 

 

You've replaced a railroad tie with a toothpick. No chance it survived anything more than hard parking competition.

Pete Gossett (Forum Supporter) said:

In reply to blackmount1 :

I'm no engineer, but the fact there's nothing directly preventing the front control arm mounts from flexing inward/outward seems like a potential source for problems. 

That's where those bosses for the extra braces are.. The subframe is not just mounted by itself, it has extra bracing that bolts onto it to connect to the front cross member, I also made a brace that connects those two points to eachother at the forward mounting points for the LCA's . I'll have to get a picture for you..

Here is OEM setup for reference, 

All I can add to this conversation is that I had some really nasty toe deflection under hard braking in my GTI when the front suspension bushings were shot, but it looks like you're already looking into that.

I mostly wanted to say that I love your taste in cars and I wish I had the stones to own a similarly awesome collection.

Cool cars. A 2700 lb 3000GT! Wow, congratulations. I have a soft spot for 3000GTs even though they are so heavy and overly complex.

I have nothing really helpful to add, Unfortunately. 

Fire up the welder.

My first thought when looking at your original pics is that you're flexing the entire thing. It looks weak from a design standpoint.

Maybe I missed it, but what is the perceived advantage of this over the stock subframe? Unless it's something substantial, I'd skip the fabricating and sell it to someone else, and use the stock subframe. 

Like the comments above, I think the subframe is deflecting quite a bit, likely in more than one plane. The stock piece is very stout. If it’s like a DSM, then there should also be heavy plates bolted underneath too at the corners stock, which also looks to be missing in the new subframe. 

Some good news.. I got the car on a lift today, and was able to get a much, much larger prybar under there with a pipe slid over top of it as oppose to when it was on jackstands.

For the most part all i got was some scratched off paint, The Frame itself still does not seem to deflect very much if at all, even at the front - it does when I take the braces off, but even then it was hardly much at all - with them on if it moves at all, its invisible to my eyes, and the other persons eyes who's lift i was borrowing.

BUT

I did find the control arms and their bushings moving in and out excessively, I can pry on it, with not much very force, it will move, and stay there. Pry it back, same thing. It moves, and stays there... about a good 1/8" of movement and it translates to alot more then that in Toe at the wheel, and causes the control arm to pivot one way or another making the toe angle change drastically... I'm thinking this is why it gets stuck, and going lock to lock pushes the bushings in a different direction... I've got 75ftlbs on each mounting bolt, being only 3/8ths they wont go much more then that before snapping or stretching..

At this point I'm more just trying to help out the guy building these things. Because he's a good guy, and how can you fix a problem if you dont get any feedback about it?

CyberEric said:

Cool cars. A 2700 lb 3000GT! Wow, congratulations. I have a soft spot for 3000GTs even though they are so heavy and overly complex.

I have nothing really helpful to add, Unfortunately. 

Thanks man, glad to hear it! Yeah I've been obsessed with them ever since I was in elementary school and you could still buy them new. I've had about 8 or 9 now, healthy mix of VR-4's and Non Turbos. Personally I prefer the Non Turbos. I just love simplicity,

It use to be almost 2600lbs but I street drive the thing so freaking much, that I decided I was getting old, and wanted a bit more interior, and a quieter ride, as well as exhaust..

replacing turbos with a 8500RPM Redline, and Big Cams is a good way to shave off 200lbs, and weight in general comes off pretty easy on these cars, all the Active Aero is gutted and permanently fixed in the deploy'd position, No Radio, No AC, just heat.. Still has power mirrors, windows, and locks though.

I try to keep it a fully functioning street car, but a half cage, and FIA legal buckets with harnesses are coming this fall.

chaparral said:

That front subframe is probably more flexible than a go-kart chassis. I bet you are plastically deforming it - bending the whole thing - and it is taking a "set" at the toe angle than you're noticing. You will want many more tubes in there to triangulate it properly so that it doesn't bend.

That's what I am thinking, the control arm pivots are probably rotating around a central axis, since they are pretty much unsupported, so the ball joint can move fore and aft.  This will cause a toe change because the tie rods are not going to be equal in length to or parallel to this axis of motion.

A brace between the forward piivot locations would be the most beneficial.... If you can even get something there with all that drivetrain and stuff in the way!

So is your car n/a and AWD, or FWD?  how much power are you getting out of it?  If you have the awd car down to 2700, that's pretty impressive!

blackmount1 said:

Pete Gossett (Forum Supporter) said:

In reply to blackmount1 :

I'm no engineer, but the fact there's nothing directly preventing the front control arm mounts from flexing inward/outward seems like a potential source for problems. 

That's where those bosses for the extra braces are.. The subframe is not just mounted by itself, it has extra bracing that bolts onto it to connect to the front cross member, I also made a brace that connects those two points to eachother at the forward mounting points for the LCA's . I'll have to get a picture for you..

 

Here is OEM setup for reference, 

 

Ah ol, it's pretty much a scaled up version of a DSM, which makes sense as I've been discovering quite a bit of similarities & parts-bin sharing between the two. I don't own a 3000gt but I picked a VR4 for a friend/fellow GRMer SaulGoode & have it stashed next to the garage. 

blackmount1 said:

BUT

I did find the control arms and their bushings moving in and out excessively, I can pry on it, with not much very force, it will move, and stay there. Pry it back, same thing. It moves, and stays there... about a good 1/8" of movement and it translates to alot more then that in Toe at the wheel, and causes the control arm to pivot one way or another making the toe angle change drastically.....

 

Are the bolt holes slotted/bigger than the bolts? If so-why? Is it for camber adjustment? If so it needs some sort of cam washer and better way to lock it down.

Edit-can we get a pic of where it moves?