Slow Car Fast
THE MG
January 5, 2011 - While I had the car on the lift, I decided to drop it down on to the scales.
After all, it's just about completely done from a mechanical standpoint. There's the bodywork and paint to do, but that's unlikely to add any weight. So we can call this a pretty accurate final weight.
The car has nearly a full tank of gas and even carries a spare tire. So this is a full-on legitimate street weight, not a setup for the lightest possible option. Still, I'm pretty happy. When I first weighed the car a year ago, I estimated 2400 lbs. Looks like I was right!
Check out the balance. 51.9/48.1 front/rear. Even better, perfect corner balancing and side/side distribution. There's no reason for that, I just got lucky. But it's still worth bragging about!
entry 529 - tags: weight
January 9, 2011 - A few interesting numbers.
I was looking at various specifications, and came across the original MG ones. Or at least, a reasonable facsimile thereof - it can be hard to find sources for some of this.
Apparently the MGB GT weighs 2443 lbs and has 52% of its weight on the front wheels. The V8 version that appeared in 1973? 2390 lbs - the stock four-cylinder was really heavy. So my car is basically the same weight as the factory V8 job. Of course, it had 137 hp while mine is somewhere around 350-400, depending on the modifications that were made to the engine before I got it.
I was also looking at the hot new 2012 Boss Mustang. It's rated at 440 hp, but weighs as much as the MG and my Seven combined. More eye-opening is the size - it's 35" longer than the MG. Nearly three feet! Wow.
More work to come on the MG soon. I can feel the final bodywork looming. It's going to be a dusty time.
entry 530 - tags: dimensions
January 14, 2011 - I spent part of the day today talking paint and bodywork with the local paint shop.
Hightower supply has been very helpful to me. I came home with some new primer and a big bucket of body filler to make those fender flares all pretty.
I also want to try some rubber motor mounts in the car. I think some of my noise is vibration coming through the mounts. I put some competition motor mounts on Janel's street Miata and got a similar noise. Now, the mounts I used were generic small block Chevy (pre-LS1) mounts from Energy Suspension. I went to NAPA, opened up the Big Book Of Motor Mounts, and spotted this on the first page. I don't know what the application is - probably every V8 from General Motors from 1960 to 1997 - but I think that's it. I just hope the softer mounts don't let the exhaust and other bits and pieces lean up against each other.
entry 531 - tags: motor mounts, bodywork, paint
January 14, 2011 - Big news - I'm going to the Targa Newfoundland.
Long-time readers may remember that the Targa caused a hiatus in MG development in 2008. Actually, it was during that pause that I decided to go with the LS1 instead of a Ford 302. But anyhow...we're going back.
You can read all the details on the Targa Miata website. It's actually going to get an engine swap, and the subject of the swap should be pretty familiar. Yup, there's an LS engine going in. Not an LS1, but the L33 I'd been considering years ago. The power/weight ratio will actually be very similar to that of the MG, so I guess we could call this little car practice. What I learned building the Miata in the first place helped the MG, and I've picked up a few things with the MG that will make the V8 Miata work better. Everybody wins!
However, I need your help. We need more funding in order to make this a reality. If you're enjoying the build of the MG and my other projects, please consider helping out a bit. Of course, the build and the race will be documented just like this one has been. Details
entry 532 - tags: targa newfoundland, help
January 16, 2011 - I swapped the motor mounts out today.
It was a reasonably easy job, probably much easier than most production cars. It'll be even easier next time, since I turned one of the long bolts around so it runs back to front instead of front to back. Seems simple, but next time I won't have to pull the alternator...
I believe it was a success. The car doesn't boom and vibrate as much. It's still vocal but you can tell the noise is coming from the back of the car through the exhaust pipes instead of simply through the chassis. A bit more work with sound deadening in the back half of the car and I think it'll be quite good.
entry 533 - tags: motor mounts
January 18, 2011 - The flares look pretty good, but there's one line that bothers me.
The trailing edge of the front fenders doesn't blend into the body as well as I'd like. It's not something you'd notice unless you spend hours looking at the car and working on small areas of it, but they don't look factory.
I do have a couple of stubs that were cut off the leading edge of these flares as I was building them. They have a better shape.
entry 534 - tags: body, fender, flares
January 18, 2011 - Much better!
The flares now blend into the body perfectly, even following a curve at the bottom of the fender when viewed from the side. I'm really happy with how this turned out, they're even better in person.
It's unlikely anyone would have noticed unless I'd pointed it out, but now I'm satisfied.
entry 535 - tags: body, fender, flares
January 18, 2011 - Since I had the welder out and the front wheels off, the obvious thing to do is to add a front sway bar.
The car needs more roll stiffness, and I've been looking at how to add a front bar for a while. Of course, I'm trying to use a Miata part because I have a huge collection of them.
I built a couple of plates with studs in them and welded that to the bottom of the frame rail. I considered simply installing some rivnuts, but the sway bar mounts see quite a bit of load and I can see myself ripping them out of the frame. So I have a nice solid mount instead.
The geometry would have worked better if the bar was on top of the frame rail instead of underneath, but the packaging would simply not allow that. Even if I could cut a big enough hole in the panel above the rail, there isn't enough room for the bar to go between the engine and radiator. We'll see how it works. On a short drive, the car did feel more settled. Too much traffic to try out the handling balance. I'm still dialing in the spring rates so I'm not sure exactly what the sway bar sizes will be yet. This one's off a 2004 Mazdaspeed turbo Miata. Why? Because that's what I had on hand! If I want a softer bar, I can use a non-turbo version. If I want an adjustable bar, I can use one of the Flyin' Miata ones.
And yes, I have an idea for a rear bar. We'll see if it will package.
entry 536 - tags: sway bar, handling
January 18, 2011 - The sway bar has an upward angle to it due to the bar placement.
The stock end link - the black one - made this as bad as it could be. So I decided to go with a shorter unit. I have boxes full of old Miata suspension pieces, of course, and the heim-jointed end link in the middle is from the "end link" box. Seriously, I have a box for Miata end links. I also have a box for bumpstops, a wall full of springs, a cord of sway bars, random steering column parts, etc. And at least three boxes of wires. But I digress.
The unmodified heim-jointed link was still too long. So I cut it down, as you can see on the right. There's still a lot of thread engagement so strength is not a problem, but I've made the angle of the bar much better.
It's not the only suspension work I did today. When I was first doing the suspension work on the car, I didn't pay too much attention to the control arms and suspension uprights I used. All the important dimensions are the same. But in my fevered excitement to assemble the car for the First Drive, I left those parts on the car. And that was a problem.
Mazda tweaked the Miata's suspension in 1999. The control arms got beefed up a bit, although the dimensions were basically unchanged. The one change was on the taper of the upper ball joint. The steering knuckle/upright was also very similar, with the steering arm moved up by 7mm and the appropriate taper on the seat for the upper ball joint.
Well, when I looked at what was on the car, I realized that I had the new design upper control arm on one side, but the older steering knuckle. On the other, I had the older parts. This is less than ideal, so I swapped everything out for the newer designs. It was actually a pretty quick job, the guy who built this car was careful to leave enough room to do things like remove suspension bolts. Whew!
entry 537 - tags: suspension, sway bar, end links
January 18, 2011 - Time to play with tires.
When Janel drove the car on Saturday to see how the various modifications were improving things, she discovered quite quickly that the existing tires were poor. A serious liability, actually. The car's like a loaded gun, with an extreme power/weight ratio and some pretty poor tires. That's the current tire on the left - it's the Falken Azenis RT215. The 215 was discontinued years ago, and while this tire appears to have lots of tread it's actually been heat-cycled to death so it's a little black rock. They hold air, that's their biggest redeeming factor.
The middle tire is a Toyo RA-1, a race tire that can be run on the street. It was pulled off my Seven, which is sitting in the trailer. It's mounted to the same wheel as my Falkens, but is a slightly wider 205 instead of the previous 195.
The right tire is another RA-1 with a little more wear. That's off the Targa Miata. These tires actually last an amazingly long time for a race tire. I pulled it out because it's mounted on a 15" wheel with a different offset, and I wanted to see how it would fit. Also, it was in my way.
entry 538 - tags: tires
January 18, 2011 - Here are the 14 inch RA-1s installed.
The left rear is the most problematic tire, and it's not quite tucked into place here. The others are fine, with a perfect fit up front. Really, I should have made that rear axle about 1-2" narrower so I could run bigger tires in the back. That may be a future project. For the time being, I want to see how this setup works, as it's a simple bolt-on that doesn't cost a cent.
On the street, the car is far better. It'll still overwhelm the tires, but only under real duress. I'm not sure it's considered socially acceptable to light the tires up all the way through second gear, but it's in the name of Science!
A nice side effect is that the ride has improved considerably. The Falken Azenis has a sidewall that is slightly stiffer than concrete. The RA-1 has a stiff sidewall as well, but it will flex. So I'm quite happy with this.
entry 539 - tags: tires
January 23, 2011 - During construction, Janel often referred to the MG as her little grocery getter.
And you know what? We're actually using it for grocery runs. She's pretty happy with how it's evolving, between the better/more compliant tires and the sway bar it feels much better planted and less skittish. She also really likes the current suspension compliance.
I've known since I put them on that the current springs weren't long enough - they coil bind before full compression. And I suspect that's actually happening once in a while. So I swapped in a set of 10" springs with a 250 lb rate instead of the 8" 225s that were in there. Why the rate change? Because that's what I have! I'll see what change this makes to the dynamics, and then order some springs of the appropriate rate and length. I've said this before, but it's so easy to change the springs on this car!
entry 540 - tags: suspension
January 26, 2011 - One last job before I start in with the body filler.
I've decided to remove the side marker lights from the car so it looks like the earliest GTs. I've been driving around with black duct tape covering the holes, but something more permanent is required. Step one, clean up the edge of the holes.
entry 541 - tags: bodywork
January 26, 2011 - Step two: cut out a patch.
I used the leftover bits of metal from the Rabbit flares. First I used the hole as a template to trace out the shape of the piece needed, then cut it with a set of snips. Much fine-tuning followed to get the perfect shape.
entry 542 - tags: bodywork
January 26, 2011 - Then the patches were carefully welded into place.
Of course, none of the holes in the car (I also patched the antenna hole) were on a flat panel. So I welded in one or two sides, then reshaped the patch with a body hammer to make it conform. It's still going to need body filler of course - especially on this side, where there's a fairly thick skim layer on the panels laid down before the black paint - but that will be kept to a minimum.
Now, on to the fenders!
entry 543 - tags: bodywork
February 1, 2011 - Big work!
I had a few other things to do over the weekend, but I couldn't leave the MG completely alone. I have a few toys that have come in for the interior, so I decided to play with those. Here's the original state of the door.
entry 544 - tags: interior
February 1, 2011 - The final result!
Okay, it's not a big change. The most obvious is that door pull. The plastic MG one really bugged me, and I wanted leather. This particular strap is actually a door limiting strap for a classic Chevy truck - but it's got the right vibe to it and should wear in nicely. It isn't a perfect match to the interior color, and that's exaggerated by the flash - but it's better. I'll live with it for a while and decide what to do.
The window winder was originally black plastic, and I replaced it with a cast metal unit that both looks and feels much better.
We have replacement door panels for the car, but this is still the dirty and wobbly original part. Why? Because I'm still considering moving that door pull, and I don't want to put extra screw holes in my good door.
entry 545 - tags: interior
February 1, 2011 - A fancy new paint gun!
Well, not new. But new to me. Since I want this car to look good and black paint is very unforgiving, I've decided to step up from my cheap and cheerful Harbor Freight guns to a good one. It's still mostly up to me, but this guy should get rid of some excuses.
entry 546 - tags: paint
February 1, 2011 - I've been working on gauges recently.
The parts just arrived to mate my mechanical oil pressure gauge to the engine. The biggest problem was how to deal with the usual oddball British flare fittings. After doing some measuring, I was able to sidestep the problem by cutting the flare off and using a 1/8" compression fitting. It's a perfect fit. That goes to an AN -3 flare, then it's all just off-the-shelf race stuff. I still have to get behind the dash to make sure the connection to the gauge is tight, then we'll see how it works!
The coolant gauge has been more of a hassle. The one for the engine computer works fine so I can monitor things that way, but the dash gauge reads really low. At 180F, it's just come off the C peg. After some putzing around with the gauge and checking wiring, it looks as if the culprit is the temperature sender in the head. The GM part measures 3800 ohms at 63F and about 150 ohms at 185F. A parts-store piece I also have on hand is the same at 63 and 200 at 185F.
The MG sender apparently measures about 800 ohms at 68F and 33 at 190F. So I need to find one that's somewhere in that ballpark. It doesn't have to be perfect, just close enough that the gauge stays near the middle at normal operating temperature. I'll stop by NAPA tonight and see what I can find.
entry 547 - tags: gauges
February 1, 2011 - My little science fair project.
I started by stopping off at NAPA on the way home and going through their Big Book Of Temperature Sensors. Not all of the sensors had information about their resistances, but I was able to pick out a few likely suspects out of the catalog. Two of them were in stock, so I brought them home to test.
First, I used the multimeter to check the resistance between the body of the sender and the stud. The MG part apparently reads 800-830 ohms. NAPA sensor 6046 read 631, sensor 6178 was 273.
Then I boiled some water and risked my tender fingertips by dipping the sensor in it while I measured the resistance. The cooking thermometer told me the actual temperature of the water. At 195F, sensor 6046 read 48 ohms and 6178 read 31. Apparently the MG one should be 31-35 at 190F.
So, 6046 will be fairly close when the car is cold, but read a bit low when hot. However, it's also changing quickly in the normal operating range. 6178 will probably read right about C when cold, and should be pretty close when hot. I'll give them both a try in the car and see how they behave for real, but at least I know what to expect!
entry 548 - tags: science!, cooling