Making Monsters out of Midgets
The Flaming Cockroach getting ready to run. My buddy, thinking he was
delivering an insult, once referred to the Midget as a cockroach. The more
I thought about it, the more I came to realize that's exactly what it is - fast,
annoying and always under foot, but hard to kill.
There's not much point in running a stock MG Midget on a dyno. The numbers will be depressing as hell. Unless of course, you've just installed $3,500 worth of supercharger on your $5,000 car. In the latter circumstance, a trip to the dyno may be considered obligatory. The good folks at Magnuson Products, having just completed the installation of a shiny new Mustang Dyno, were kind enough to let us come in and break a few parts.
I had always hoped some day to put a blower on the car. As luck would have it, circumstances conspired to make this possible. After the successful completion of kits for the MG TC/TD and MGA/B, I used a little pull to get the Spridget next up on the R&D list. The basic Moss kit, designed for 1275cc A-Series engines, uses an internally bypassed Eaton M45 supercharger on a bespoke manifold casting. A 1.75" SU HIF44 carb feeds the mixture, and a serpentine belt drive conversion provides a better and more compact drive compared to traditional v-belts.
Six years ago, I gave an engine to a well known vintage race car builder and asked him to build me a bullet proof lump. The resulting engine, finally assembled just in time to get the blower, was built with an eye to running boost.The motor is punched out .060" to 1330cc. It has an off the shelf alloy head, forged pistons, 3/4 cam and an aluminum flywheel. Compression is set at 9 :1 and a total of 11 ARP studs hold the top down. The stock supercharger kit targets 5.5-6 psi, but being an American, and living by the motto more is better, we've givin' it a little extra dose. At sea level, boost typically runs in the 7-7.5 psi range.
We had previously dynoed the stock kit on our development mule at 66.6 (how convenient) hp. I knew from driving my car there was more on tap, but the real test comes when rubber hits rollers. Empirical evidence suggested the flaming Midget had some snot. On an early trip out to Buttonwillow Raceway, The car was able to hold an '05 Mustang GT at bay through first and second gears. Of course, the Mustang had the legs from third gear on, but its driver was more than a little impressed when, foot to the floor, he couldn't reel me in off the line.
The car has yet to receive an oil cooler. So, with 70 miles of freeway driving and traffic under its belt before arriving, we only had one glory run before temps started robbing power. The dyno guy backed off at six. Knowing what we know now, I would have asked him to twist it to seven on the first pull. It was close to tipping over at six, but not there yet, and 99-100hp would not have been out of the question. The dyno graph below shows the difference between a stock 1275 Midget and what can be achieved without going totally crazy on the engine build. For those of you from Rio Linda, the blue lines are horsepower, and the green are torque. Gotta love that flat torque line from 2500 to the rev clip!
Tests performed on a Mustang Dyno. Your results will vary.
Not long after posting this article, I got a call from Sean Brown at Flowspeed. Sean primarily does MGB cylinder head modifications, but was interested in reworking a head for my car. With visions of three digit horsepower readings dancing in my head, we pulled the Midget's head and sent it off to Oregon. The head came back shortly thereafter looking a lot sweeter than before. Flow bench numbers are posted here.
After buttoning things up again, it was immediately obvious the car was off the boil. Material removed from the combustion chambers had lowered the compression to a recalculated 8.8:1. Although counter to proper testing procedures, we also replaced the rusty old 3 into 1 header with an LCB. The new header has larger primary pipe diameters and combined with a newly free flowing head, contributed to lower boost. We have since tried a couple pulleys lying around the shop, but it will take a skim cut on one to get the boost right where we want it. At the moment, I'm back to 7 psi with more lead in the distributor timing. The car feels good, but wants a bit more massaging before it heads back to the dyno. When it does, I'll post the results here.
Update 2: Better Late Than Never Update
Guilty as charged. There was always going to be at least one more chapter to this story, but it has taken rather longer than expected to get here. Having a new custom pulley made was going to cost some dough, so I begged and borrowed a random pulley which just happened to produce about 8.5 psi in good air - a tad more the I had wanted, but the engine seems content with this much boost. The problem was after the changes the car then ran quite fat. You could feel it and see it in reduced fuel mileage. I did a little experimenting and ultimately settled on particular needle with some hand filing done to get more fuel on tip in.
Although I have spent the last couple years simply enjoying the car, there are a few gnawing issues which have never been addressed. The small blower pulley and subsequent boost require lots of fuel at tip in. Lacking an accelerator pump, the SU struggles to meet this demand. Up until recently I had been running my custom ground upon jet needle, but it was never completely satisfactory. Also, the top end power just hasn't impressed the ole "butt-dyno." However, feel is a very poor indicator of actual performance, so its time for some instrumentation.
Trust but Verify
With the help of the good folks at Innovate Motorsports, we hooked up their LC-1 wide band O2 sensor controller and a G3 AFR gauge. Although it may be nice to tune on a dyno, there are none near where I live, and frankly results achieved on a 4,000 pound dyno drum don't always translate to a 1,700 pound car on the street. Innovate's wide band O2 sensor gives instant visual feedback of air fuel ratio. When used in conjunction with one of their analog AFR gauges, you not only see when things happen, but also to what degree. In other words, I knew I was lean on tip in, but not exactly how lean, and for exactly how long after stomping the loud pedal.
Once the Innovate was installed, we took the car out for some tests. First up was a simple drive around to see how the AFR looked. Then it was time to see if we could "move the needle," which in this case meant how much change can we cause with any given adjustment. My buddy helping me through this is a highly experienced carburetor and fuel injection guru. We went looking to see how much impact can a series of changes bring about. We ran with no dashpot oil, and with 90 weight. We played with the piston return spring. We removed the air bleed bushings from the bottom of the piston, and tried various combination of the above.
The interesting and unexpected conclusion of this experimenting was the realization that the only way to significantly affect how the SU runs is by either changing needle profile, or by resetting the mixture adjustment with a given needle profile. Mucking around with the dashpot piston can destabilize air velocity across the bridge, but it won't dramatically change the inherent tune. Only needle profiles, and jet height will bring on significant changes.
Just Breathe Baby
This probably won't come as much of a surprise, but we learned the awful truth about the air filter. Moss' Spridget supercharger kit includes the same air filter as their original MGB supercharger kit. If its adequate for the MGB, it should be enough for a 1275, but mine is a heavy breathing 1275. We pulled the filter for one acceleration run (Don't run without a filter, Moss won't be responsible for what your blower swallows) and huzzah! The car went absolutely berserk above 5,000 rpm. In fact, the first time I took off from a stop I hit 7,000 so fast I didn't realize what was happening.
Backing up for just a moment, on paper, the cylinder head work, new exhaust header, and increased boost should have added up to more performance, even if the jet needle wasn't quite right. I wanted to make another trek to the dyno, but just didn't feel as though the car was performing well. Although we had yet to make any permanent tuning changes, without an air filter the car suddenly felt as should have. Moss has recently reintroduced the MGB supercharger kit. Along with the redesign comes a new larger air filter. A quick check showed it fits on my Midget, so I'm now running a bigger filter. I've also contacted a buddy about making some stub stacks, which will fit inside the filter, to smooth airflow into the carb mouth. There may be a little more to be had there too.
...And Hydrate Too
Through all of this, the lack of tip in fuel remains an issue. Using the Innovate, were able to choose a better suited stock needle for idle, cruise, and full power runs. We had previously learned that small idle mixture adjustments could have a big impact on cruise and full power, so chose to ever-so-slightly sacrifice perfect idle mixture for better cruise mixture. You spend more time cruising than anything else anyway. Now there only remains the one glaring issue of not enough tip in fuel.
When you crack the throttle, that blower needs to drink big time. An SU carb simply cannot supply sufficient tip in fuel while maintaining any sort of reasonable cruise AFR. The problem is you're using the same portion of the needle for both activities. Using small piston air bleeds, heavy dashpot oil and a stiff return spring can help keep the slide down, thus causing a momentary increase in velocity, but it's a delicate balancing act, and ultimately a not satisfactory fix.
The Double Secret Fix
Based on a suggestion from my tuning expert buddy, I have designed a little device and am in the early testing stages. Although it's 45 degrees and raining, there may be news to report in the next couple weekends. With it's new found top end, the Cockroach is behaving as expected, and I'm itching to head down south for a dyno test, but not until I have that tip in cured, or at least greatly improved. There should be something to report in time for spring driving season.