Flex and why it's bad

The P-38 has virtually no flex in the crank tube under pedal load. The crank tube is braced back to the stem and with four frame sizes, riders can pick a size that more nearly matches their leg length.
This is, in my opinion, one of the reasons why it has a reputation of a good climber. The other reason it climbs well is the closed rider position and moderately high bottom bracket.

Whenever flex is raised as an issue on Forums, someone will claim that, as the flexed frame returns to it's neutral position, the power lost is returned. This is not correct. When the bottom bracket (BB) is forced down and sideways on recumbents with frame flex, it 'pulls' the chain a microscopic amount. But when the BB rises again, the power side of the chain is slackened. When a BB flexes under pedal load, power is lost.

Some recumbents have as much as 25 mm (1 inch) of flex under static load. My experience and brief tests with a trike before and after a BB brace was fitted, suggest that flex can make a trike 10% slower on average - most of the speed lost is in climbing where the most pressure is applied to the pedals.

Some of the latest carbon frames have very little flex at the BB. They are made with a deep rectangular section crank tube. It's no surprise to me that they climb well.

Riders who are short in stature and who have a high cadence may not notice much power loss form BB flex. Likewise, the very rare riders that have a perfectly smooth pedalling action, with power being applied equally around the pedal stroke, will hold a frame in it's stressed position and no power will be lost.

I have a slow cadence and a very long x-seam (distance from back of seat to pedals.)
As a result crank tubes on my bents are extended a long way forward and unless they're braced like the P-38, they flex a lot.

Lack of flex in the P-38 BB is one of the reasons; probably the main reason, that it climbs faster than many recumbent designs.

There must be some faults with the design?

There are a couple of minor design issues. Whether you call them faults is a matter of definition.

Cut seat mesh
Unless the seat mesh is stretched tight and if the seat is tilted back, the seat mesh can come into contact with the rear wheel. This burns a cut in the mesh.

The fix is very simple. Get rid of the factory seat lacing and replace with cable (zip) ties. Using zip ties gives control over the mesh tension - you can have a very tight back mesh and a slightly softer seat mesh.

To completely eliminate any chance of mesh to seat contact, fit a plastic fender (mudguard) over the rear wheel. The clip on DF variety work well. Cut off the clip at the bottom that normally clips over the DF seat tube. Drill a couple of holes in the plastic and zip tie the fender to the frame cross member in front of the rear wheel. At the top, throw away the sliding metal clamp, drill another couple of holes and zip tie the fender under the crossbar between the seat tubes.

Front brake nipple
Depending on the model and year, the P-38 may be fitted with a caliper brake behind the front fork. If you allow the fork/front wheel to rotate too far to the left, the nipple adjuster on the brake arm will contact the lower frame tube. This may snap off the adjuster thread. Everything still works but it's annoying.

The wheel would never rotate far enough when riding the bike. I've had it happen when placing the bike on a rack. The steering has flopped over far enough to cause the brake to foul the frame.

I haven't had any other issues.

FAQ or Fiction

I'm writing this Blog because I own a P-38 but also because there's some lack of knowledge and confusion about recumbents in general. When you understand why the P-38 performs the way it does, then you're in a better position to evaluate all recumbents.

Recumbents make up a very small percentage of the bicycle market. This means that they are expensive compared to what the recumbent aficionados call DFs, or Diamond Frames. DFs are 'normal' bikes like the bikes ridden in the Tour de France. We often refer to recumbents as 'bents and DFs as wedgies.

Recumbents are still promoted as being faster than DFs. The truth is that the fastest recumbent designs, ridden by fit riders acclimatised to their bike will be faster than an equally fit DF rider. The less palatable truth is that many recumbent designs are at best equal and in some cases slower than DFs.

Where does the P-38 fit on the speed scale of unfaired recumbents? It is my opinion, backed up by riding and testing other designs, including DFs, that on a scale of one to ten, where one is the slowest bent and ten the fastest, the P-38 would rate a minimum of seven and a maximum of eight.

But how can anyone measure the speed of a recumbent when the strength of riders varies so much and we live in a world that's not flat?

That's what this Blog is about. It's one mans opinion obtained at high dollar cost and in future posts, I'll expand on my experience.

Meanwhile, for all things recumbent, visit Bentrider Online. It's a treasure trove of facts, opinions and links. If you have a question about recumbents or bicycles, it will be answered there.

http://www.bentrideronline.com