Prove me wrong, measure the heat coming from a one disc and a two disc at stall RPM. I'm pretty sure the one disc is going to show more heat than the two disc.
It is not my intent to prove you, or anyone wrong...but I can offer my experience. If the clutch is set up incorrectly or overpowered, then what you say would be absolutely correct. In the OP's case of a very limited power engien, I am not so convinced, and here is why.
Keep in mind that this is very difficult to quantify and repeat (at least with my limited abilities.)
Here is what I can provide for you to consider:
My claim: A one disc Bully clutch has sufficient surface area for a small plate (purple in this case.)
The temperature produced from slippage is nearly identical. On two back to back runs from 2000 rpm idle to WOT with the same amount of load on the dyno, each clutch (fresh single disc and fresh 2 disc Bully clutch (same oem clutch linings and surface finish/grind used)) produced the same amount of increased temperature (15* F) after a 20 second pull. With three add'l down pulls, the clutches each produced only slightly more heat (raising a total of 20* F) more than what they started at (ambient air temp at 78*.) Actually, the single disc clutch was a degree cooler than the 2 disc but I would assume that is within reasonable error for my Goodson temperature gun. Still both were cool to the touch after being run - that's what is important to me.
Clutches that build heat are slipping power away from the rear wheels. If they get hot, (I don't care how many discs they've got), they're robbing power. The more discs (same size diameter and surface width) a clutch has, the better it's ability to disperse heat across a greater surface area. For that reason, higher HP (more specifically torque since that's what we're concerned with clutching,) engines require more surface area in the clutch, ie more discs.
For purple plate animal, a single disc will be sufficient. If you find it getting hot, it may be out of spec or in need of being rebuilt.
On the other hand, a 2 disc really won't hurt your performance (seat of the pants anyway.) Interestingly, the weight of the clutch also has some input on where the peak total output of the engine (torque) lies. That's why you often times see builders recommending an engagement speed slightly different than the engine's peak torque on the dyno. For myself and my customers, I recommend to engage the clutch (single or 2 disc) AT peak torque.
FWIW: On our dyno, the clutch I typically use for engine testing has a belt drive and a very low engagement rpm so I don't miss peak torque numbers when making pulls due to clutch disengagement. When breaking clutches in though, we use a chain drive and normal engagement at peak torque.
For the OP, your problem is not the number of discs in the clutch, it is likely how it is set up or in need of being rebuilt.
It could be something entirely unrelated to the clutch as well (driver not anticipating the starts, engine power is down, dragging brake, incorrect gearing, driver, etc.) as well. If you'd like to give us a call here at the shop, I can offer my recommendations on your clutch settings. If it varies considerably, then we may have just found the problem. If my springs, spring heights, weights, air gaps, etc set-up is similar to yours, then it just might need rebuilt or start looking elsewhere for the problem.
-----
Thanks and God bless,
Brian Carlson
Carlson Racing Engines
Vector Cutz
www.CarlsonMotorsports.com
27 years of service to the karting industry
Linden, IN
765-339-4407
bcarlson@CarlsonMotorsports.com