dbakl

I remember this coming up in a thread recently. What do they do?

Anyway, spotted 2 on ebay if someone needs:

https://cgi.ebay.com/Simplex-Demultip...0#ht_500wt_922

ColonelJLloyd 

Cue Grand Bois in 3...2...

-holiday76

thanks for reminding me I needed to buy that $20 one

Charles Wahl

If I recollect accurately: the Simplex barcons wrap so much cable that it's difficult to find the friction-shifting "sweet spot" unless the demultiplicator is used, so that there's more rotation on the levers between each gear.

sailorbenjamin

Yeah, the incoming cable joins the lever at a different spot than the outgoing cable so you get a different amount of pull.
The levers don't give the same amount of pull that the derailler wants so this converts it for you.

tugrul

Does anyone know the ratio?

jrecoi

Doesn't that work in a similar manner to a Shiftmate? If it does, would a shiftmate be useful for the same purpose as the original Simplex demultiplicator, as an in-line kludge for shifter-derailer cable pull issues?

-holiday76

I'll find out soon enough, but what I don't get it how the simplex units pull any more cable than say a campy bar end or even a suntour. I have all of them and mechnically they really seem to be the same.

-holiday76

and btw that guy must have a lot of them because he just relisted another one for $19 right after i bought mine

https://cgi.ebay.com/Simplex-Demultip...item4aa9726c4c

Charles Wahl

Yeah, so you can get the demultiplicators in bulk, but who can find the Simplex barcon levers that go with them?

sailorbenjamin

Maybe it's the derailler that needs less than average pull?

-holiday76

wasnt too hard to find the set I have if I remember correctly. They arent the retrofriction ones though.

noglider 

I understand why it does this but I don't understand how. Please explain it so even I can understand.

Michael Angelo 

Mechanical advantage. At the pivot point you have a lever with two cable stops with two different lengths. The rear lever(shorter) moves farther than the front lever (longer)for the same amount of movement.

qd-s

perhaps best to comprehend by illustration, see sketch (third picture):

https://velobase.com/ViewComponent.as...m=103&AbsPos=5

Grand Bois

I've given up trying to explain it.

dbakl

Does it DECREASE or INCREASE the pull? Not playing dumb, just never seen one!

Michael Angelo 

It pulls more cable at the back side, so it increases the pull.

Ken Rasmussen

I have a Bike Friday that uses a Sram 3 x 7 hub, which is a three-speed hub that accepts a 7 speed cassette. I’ve actually got 8 speeds on it, by using a nine speed cassette with the largest cog removed. Coupled with a double crankset, it makes a 48 speed wide range gear system—ideal for a folding bike. The problem that I had, is that the Sturmey Archer barend shifter that I used does not index properly with the 3 speed hub. The shifter pulls .3” of cable between first and second gear. The hub requires .2” of cable pull to accomplish a well-centered shift into 2nd gear. Plan B was to make a visual sight gauge that would allow manual shifting using an un-indexed bar end shifter. However, after I had accomplished Plan B, I remembered that I had once found an antique part of unknown function, that appeared to have been made for the purpose of changing the cable pull chararacteristic of a shift lever. The part was made by Simplex. It had a glide for a front derailleur cable on the left side, and the cable pull alteration mechanism on the right side. I modified and adapted the part, and mounted it upside down and backwards on the bike below the front derailleur—(my shifter lever was a left side bar end shifter, and I needed to cross to the other side of the bicycle. My cable routing on this bike is all improvised to take maximum advantage of an inadequate number of cable braze-ons). By playing with the angular attitude of the pivoting lever, I made it accomplish a properly indexed shift. It’s a terrific discovery, because using a 3 speed hub with handlebars that use racing handlebar diameter tubing makes it impossible to use a proper indexed shifter. This solves the problem very elegantly. I subsequently found your forum posts on the subject, which clears up the mystery of what the part was originally intended for. I would have posted photos, but I’m new to the forum, and I was blocked.

machinist42

OP's Album.

gaucho777 

OP member dbaki was a good dude. RIP.

Welcome to the forum, Ken. I can't decide if this latest example is an elegant configuration and Rube Goldberg machine. Seems that a lot of thought and experimentation went into that setup. Nonetheless, those Delrin Simplex shifter units are atrocious seem like a weak link in the system.

My machines tend to be simpler, though I am curious about the intricacies of cable pull. I've never paid much mind to this measurement. Is cable pull something than someone measures or is it simply build into the specs of the components? I wonder if careful and judicious filing down or adding material to the groove in a shifter unit wouldn't also accomplish the same goal.

Ken Rasmussen

I’m not offended by the Rube Goldberg comparison because this bike certainly has the look. However, the difference between Rube Goldberg and my bike is that Rube Goldberg takes a very indirect route to accomplish a task. I believe strongly in the “form follows function” design approach. In this instance I had design parameters, and I was looking for the best and most practical solutions to problems. If you think about small wheeled, folding bicycles, there are many drawbacks. They are ill-fitting. They are harsh riding. The gear range is deficient in either low gears, high gears, or low and high gears. They are slow. And so on. I’ve made a fair job of addressing ALL of those criticisms as regards this particular bike. In addition to that, I used the Rotor RS4X crankset to smooth out the dead spot in the pedal cycle. It does that by means of yet another Rube Goldberg looking mechanism involving an eccentric and linkage. The cranks run in and out of line, and the chainrings speed up and slow down. Those effects are timed to make the dead spot much easier to pedal through. I’ve become quite fond of those cranksets, and I use them on the five bicycles that I routinely ride. These cranksets are heavy, expensive, hard to get parts for, and require more maintenance than ordinary cranksets. A rider trying one won’t like it until they’ve pedaled it long enough to adapt. After that they’ll love it. HOWEVER they accomplish their objective very well, and that more than compensates the rider for using them.

As regards the Simplex Delrin shift lever that you see attached to the stem riser, that doesn’t shift the hub. It shifts the front derailleur. I absolutely agree that it is one of the less likable shift levers that one could use. However, when I went searching through my boxes of junk to find an unindexed shifter and bracket that would fit around a 1 1/8” tube, that was the best that I could come up with. It has several good characteristics: It works well. It’s light. It’s a five speed bracket—no ugly boss sticking out the other side. And so I used it. One of these days I’ll find a Simplex solid aluminum or laminated steel shift lever and replace the Delrin one. I like the bracket very well. I’ll stay with that. By the way, it’s 1” bracket that I distorted and modified to wrap around the 1 1/8” tube. I had to use a bit narrower screw that was longer.

When I read on this forum what the purpose of the Demulticator was, I had your same response: Why didn’t they just design a shift lever with a smaller barrel? As regards your idea of filing a deeper groove into an existing shifter to accomplish my purpose, it might have worked. It’s hard to say. The ratio between the cable pulled, and the requirement of the hub was 3:2. If that means that I would have had to file 1/3 of the way through the barrel to make it small enough, I wouldn’t have enjoyed it! Especially with having to install and test it periodically to see how I was doing. Installing the Demulticator was relatively easy, and by angling the arm back and forth, I was able to make large differences in the amount of cable pulled until I found the proper angle. It works exceptionally well.

As regards cable pull, I don’t know where you’d find specifications. They have to exist, but they aren’t routinely published. Sheldon Brown has some information on his site. It didn’t shed any light upon what shifter you could get to operate a Sram 3 x 7 hub with a racing handlebar. Bike Friday was using an Ultegra 9 speed set of brakeshifters on the bike originally. They somewhat worked—much like the Sturmey Archer lever that I used. They really weren’t a great match for the hub. They required Travel Agents to adapt to the brakes. I think that they were as good as could have been found, and brake shifters enhance the sales appeal of bicycles in the marketplace. I don’t like Travel Agents. When the pads wear beyond a point, the brakes fail dramatically without warning. I just measured the cable pull myself. It was easier than trying to look it up.

Green Gear Cycling/Bike Friday have done a lot of interesting things. Making a good quality bike, sized to the customer, that fits into a suitcase, with a suitcase that converts into a trailer that the bike can tow is quite an accomplishment. I’ve taken a high quality, 20 year old Bike Friday, and thoroughly redesigned it for my own taste and purposes.

Ken Rasmussen

I suspect from your comment that your attention was misdirected in the photo that prominently featured the Simplex Delrin shift lever. I notice that the camera mistakenly focused on the shift lever, leaving the item that I had been trying to focus upon a bit fuzzy. I was trying to show the improvised sight gauge that I made out of a ball point pen, a bead from a fishing lure, and two adjusting barrels. The sight gauge ultimately became unnecessary because I succeeded in adjusting the cable pull ratio so that I could use the indexing shift lever. However I left it in place because it was a useful reference when I was working by trial and error to adjust the angle of the Simplex Demultiplicator. The original thought was that the sight gauge would be an essential element because I’d need to use an unindexed lever. I think that it’s worth calling attention to it, because it’s an idea that could have relevance for other mechanics faced with similar problems. Anyone wishing to duplicate the gauge might be interested to know that Cyanoacrylate adhesive—i.e. super glue—was very successful in attaching the bead to the cable, and that the adjusting barrels tended to screw themselves out of adjustment from road vibration until I stopped them with some light Locktite.

Ken Rasmussen

The Demultiplicator is a pivoting lever with two cable attaching points. The one nearer to the pivot will tend to move less cable than the one further from the attachment point. It isn’t quite that simple though, because the lever isn’t straight, and when it rotates very much, the ratio of cable movement between the one attachment point and the other fluctuates hugely. I mounted mine upside down and backwards on the wrong side of the bike, and set it at a bit of an angle so as to direct the cable across to the other side of the bike. This was because I was making the best use of existing braze-on cable stops on a complicated folding bicycle frame. However it created a very strange set of leverages, and by luck, I was able to find an angle of the lever that resulted in exactly the characteristic that I was looking for. Had I not found it, I’d have tried to mount the Demultiplicator on the opposite side of the bike, which would have given me different leverage relationships to experiment with. In its normal position on the right side of the bike, mounted as intended by the manufacturer, more cable pull at the shift lever will be converted to less cable pull at the rear derailleur. In my case, with the three speed hub that was being shifted from the left side of the bike, the same thing was true, but there were some potential leverages available to me that were extremely wonky due to the bend in the lever running backwards to what the manufacturer had intended.