In my experience, very minimal difference, even less than the 2 to 3 watts you’re referring to. Keep in mind these wheel manufactures test wheels individually - not on a bike with a rider. I don’t know anyone who’s tested these things super accurately the way you’re describing in a wind tunnel or better yet, velodrome. In my on road field testing it’s definitely within the margin of error. I would save your money. I’m running 2007 808s with 23mm tires and found absolutely NO disadvantage over 808FC in my field testing.

Where did you see that those older wheels were optimized for a 21mm wide tire?

Not sure how scientific this is:

https://www.slowtwitch.com/Products/Things_that_Roll/Race_Wheels/Hed_revises_iconic_H3_4242.html#:~:text=The%20original%20Hed%203%20is,width%20to%20retain%20aero%20performance.&text=According%20to%20Hed%2C%20the%20new,with%2021%20and%2023mm%20tires.


"The original Hed 3 is 19mm wide, requiring the use of 19mm tires for optimal performance. "

Used to be on Zipps website.

I don't know HOW MUCH the effect is, but the rule of thumb that seems to go around and promoted by the guys at Silca, ( they don't sell wheels or tires, but work with optimizing top teams ) is that you want the widest part of the rim to be at least 105% of the measured width of the tire for optimal aerodynamic profile of the wheel.

Rule of 105

So if your printed 23 mm tire measures out at say 25 mm on the rim, you'd want your rim at it's widest point to be at least 26.25 mm

https://biketechreview.com/...ce-of-tire-selection

---

=================
Kraig Willett
http://www.biketechreview.com - check out our reduced report pricing
=================

Remember that Continental 5000 GP was designed with newer, wider rims in mind. If you run them on old, narrow rims then you are going to have a narrower tire. My 5000 in 32 measures 28 with a 14mm internal. It is just under 32mm on a 21 internal and probably right at 32mm with a 25mm internal. I would imagine the 23 would be closer to 21 or 22 on a narrow internal rim.

I tested my old 19mm Bontrager Aerowing against a newer Conti GP 4000S in 23 mm on the same 14mm internal rims. The results ended up being exactly the same, so either the difference was not determinable by my testing, or (IMO) the better rolling resistance of the Conti tire was counteracted by the more aero-ness of the narrow tire so the speed was a wash.

...as well as wind direction. According to my seat of the pants testing, a 23mm 5000 on a 19mm wide x 80mm deep rim (on the front) is definitely faster than a 23mm 5000 on a 22mm wide x 50mm deep rim. I'm sure that 19mm width hurts, but it can't be by very much.

Confession time, my race set up is an old hed 3 with a 23mm GP4000S. There is definitely some mushrooming of the tire over the wheel rim. However, there is method to the madness. I have a speed concept which has wide fork legs. The idea is that air passes between the fork legs. The Hed 3 is great because it has a lot of open space for air to pass through the fork without interfering with the wheel. So the wheel itself may take a hit aerodynamicly, but the fork- wheel interface makes up for it.