Novara Pixie 20" project
I bought a used 20" Novara (REI) Pixie bike for my 6 year to replace her 16" singlespeed coaster brake last year. Decent aluminum frame but majority of the components were pretty low end and heavy so this past winter I embarked on a project to upgrade it a bit. Not necessarily a high $$ rebuild, I mostly wanted to use old parts I already had or could find cheap at local bike co-op.
As found, the Novara Pixie weighed around 27 lbs, there were some easy places to shed a lot of weight quickly. Unfortunately, the frame had a one piece crank that weighted in at 1511 gr (including cups) and the BMX style shell somewhat hindered easy of upgrading the crank. Additionally, the crank was 150mm length, a bit too large for the 52" rider. It also had a 40tooth chainring and she never had a low enough gear for climbing.
I made a square taper bottom bracket adapter using an old MTB spindle that I machined to fit within some better quality BMX sealed bearing cups that pressed into the frame. Seemed like a slightly mroe elegant way to deal with the BB rather than using the bolt-in adapter cups and then adding a regular ISO thread BB on top of that. THe adapter BB still weighted a heafty 389 gr, combined with a 34tooth Bulletproof brand 140mm BMX crank (+ outer bashguard) that itself weighted 581 gr, saved a total of 540 gr over the stock one-piece boat anchor.
I also replaced the seatpost with a 230gr post from bike co-op, saving only 25gr over the stock alloy seatpost but got rid of the hideous looking steel saddle clamp. Also changed out the steel rear derailleur for an old ultegra road derailler (218gr), saving 100gr.
On the front end, I swapped the 407 gr steel handlebars for some 145 alloy bars from co-op, got rid of the 371gr steel quill stem for a 285gr alloy stem, changed to a alloy headset and most significantly got rid of the barely functional 1800gr RST fork for a modified, shortened early manitou fork that weighs total of 1330gr and gives a much softer ride actually suitable for a 60lbd rider.
Result as it stands right now the bike had dropped around 5lbs to 22lbs total and I have maybe spent $100 on original price + the upgrade parts, improved the gearing and suspension handeling. The stock V-brakes are decent tektros that will probably stay on it, seat is a bit heavy at 355gr. I have not yet touched the wheels/tires but for the next phase I anticipate that by swapping to 330gr Schwalbe MoeJoes, re-lacing the decent alloy Alex 36 spoke rims with only 18 spokes and perhaps using some better hubs I can pretty easily drop another 2lbs of pure rotating weight and get the complete bike down to 20 lbs.
That's an amazing job. I wish I did something like that with my daughters 20" Nakamura (Sport Check), before I handed it off to one of her friends at school. She now rides a 26" Kona Fire Mountain that is pretty stock now, but hope to have some upgrades in the next little while.
"re-lacing the decent alloy Alex 36 spoke rims with only 18 spokes"
Do you know a site that shows how to do this? Do you simply take out every other pair of spokes? What do you do with the extra holes? Leave them as is? I'd think the sticky rim tape would attract dirt and junk.
Did you take pics of the front fork modifications? A write-up would be awesome!
Nice build BTW.
For a 36 to 18 spoke re-lace, I did 6 pairs of spokes in 2X pattern on drive side that are tension ballanced against 6 radial laced spokes on the non-drive side. The NDS radial spoke is placed between the 2 crossing spokes of each drive side spoke pair. I did this by lacing the 6 NDS radial spokes into rim holes that were originally angled toward the drive side and leaving all the NDS rim holes empty. To get the nipple to seat well and get a good spoke line, I elongated the 6 rim holes for the radial side so the nipple/spoke would angle correctly, used washer between the nipples and rim to distribute the load around the elongated hole better.
Originally Posted by indianadave
For the front, you can go to all radial spokes and just use 1/2 as many. Either lace two consecutive holes, skip two holes or else lace all the spokesfor both sides into every-other rim hole. If you need a crossed spoke pattern for disk brakes on the front, you could likely do a crowsfoot pattern with three sets of three spokes on each side - laced with two spokes crossed and one radial. (2crossed+1radial)x3sets X 2sides = 18 total spokes.
You can always cover any of the unused rim spoke holes with kid stickers if you dont want dirt/mud to get in.
I weighed the spokes for a 24", 36 spoke wheel, concluded it would save 100gr of weight per wheel to reduce the spoke count to 18.
More good kids-bike relacing spoke discussion here;
Why are kids bikes so heavy?
As far as the manitou fork, Credit and inspiration for the idea goes to forum user TigWorld as in thread; Why are kids bikes so heavy? and the great write-up he did at FAQLoad - 20" front fork build
I was just initially a bit intimidated by TigWorld writeup tutorial because of the heavy use of lathe, milling machine tools to accomplish the needed modification. I decided to go ahead and give it a try and accomplished the same conversion using little other than a hacksaw and hand file, the 20" manitou came out great.
My commentary/process corresponding to the steps listed in the TigWorld tutorial are;
Probably hardest part of the entire conversion is just getting the pressed-on cast fork end (dropouts) seperated from the forks lower tubes. Without access to machine tools, I just hacksaw cut the lower tubes about 2cm above the dropout casting (Higher than TigWorld so I would have more of the outer tube to grab), then used a handheld hacksaw blade and to make a couple of slits along the axis of the remaining stub of the outer tube that remained pressed into the dropout (dremmel grinder might also be helpfull here). THe slits helped to weaken and de-tension the super-tight press fit between the tube OD and the dropout casting ID enough that I could eventually partially collapse the tubing stub by proding with a dull chisle and pliers. To help pull last of the tubing stub out, I next I drilled a couple of holes in the tubing, just above the dropout so that I could pass a large nail through the tubing and (loosely) hung the assembly on the through-nail, over the open jaws of a vice and then used a punch to to hammer the dropout down and off of the weakened tubing stub.
For Tig's step 4, I just used a hacksaw to cut the remaining tubing to length (not a lathe) and then used a machinst square and hand file to get the ends nicely square. Note that there is no need to reduce the OD of the tubing any, it is perfect OD for a tight press-fit into the dropouts.
Step 5- I used a spare hub to help align and start the lower tube into the dropout but finished it off by just hammering the lower tube into the dropout as I didnt have a press big enough for length of the tube, worked fine. Along with loctite, I drilled a 1/8" hole in the assembled dropout + tubing and added a pop-rivet to ensure it would never loosen and separate. (The press fit is extremely tight, rivet probably unnecessary, I didnt see evidence that manitou even used loctite for original assembly).
THere a a lot of variations of how the early manitou 1,2,3,4 fork design changed. I think Tig's manitou sport fork was actually based on a manitou 2 or 3, not a 1 as the original manitou 1 (1990-1991) fork (pre-answer design) had a much different looking arch between the lowers. The manitou 1&2 and perhaps some of the later years lower end manitou models put the elastomer stack below the stanchion tube while the improved model 3&4 forks moved the elastomers up to the top, inside the stanchion. For a short fork conversion, stay away from any of the manitou (1 or 2?) forks that used externally butted steel stanchion tubes, the butted section of the stanchion would be too loose when the stanchion is shortened and the butted section re-located up near the forks upper seal/bushing. ( I did manage to use a butted steel stanchion by sectioning a piece from the middle and brazing a reinforcement sleeve inside but it was a lot of work and hassle). The very earliest Manitou 1 fork used non-butted steel and some later models utilized unbutted aluminum stanchions that make a 20" conversion much easier and are approx 200gr lighter.
For any kids manitou conversion, just run a spring on one side of the fork, no need for 2 springs or urethane kabobs with a small kid. You can either stick with elastomers which also provide reasonable dampening but I opted to convert to using a steel coil spring and stretched it to around 60mm travel. There is enough inherent friction in the operation of the fork that it dampens OK with the light spring pressure. On those manitou forks with the spring below the stanchion, if you do go to just using a spring on one side, note that the plastic bushing at bottom of the stanchion is normally held in place by spring pressure so you will need to use a short bolt and large washer to keep the bushing in place without spring pressure. This would not be needed on a manitou 3 or 4 where the spring is located upward, inside the stanchion. THe later forks have bushing holder that keeps them retained without spring pressure from below.
Overall, the manitou was a huge improvement on the too-stiff 4+ lbd junker RST fork that the bike started with and the vintage CNC fork components also have high bling factor!
As an aside, my other previous experience with modifying early manitou forks was exact opposite direction of fitting them to kid bikes, I have modified and setup a couple of manitou forks as fatbike forks for use with 4" snow tires. The versatility of the basic manitou design is great for offbeat adaptation.
FatBike front suspension
Last edited by GrayJay; 04-07-2014 at 02:41 PM.
Sensational job GrayJay, even more so doing that fork shortening using only hand tools. I like the addition of the pop rivet for extra security.
Fantastic end result and bike in general. I hope your daughter loves it and rides it until its worn out.
Very nice job! Love the shock mod.
Tig- thanks again for your encouragement and the original FAQ write-up of the manitou conversion, definitly helped get me started.
Originally Posted by TigWorld
I am also starting work on next bike for my older daughter. My Younger daughter turns 6 next year and already has plans to inherit the 20" pixie when current rider (her older sister) moves onto a 24" bike. For the 24" bike, I am also taking inspiration from your 24" rockshox mag21 conversion. While the mag21 is not terribly laterally rigid for an adult rider, I used one myself for 15+ years and it was always very tuneable, reliable and lightweight. With a simple long-travel spacer modification, these forks get 60mm of travel and the air-spring is very easy to tune for a light rider. For my mag21 24" conversion, I took a different approach, rather than relocating the brake-bosses lower for the 24" rims, I chopped the stock dropouts off the cast lower tubes and devised new dropouts that shortened the lower tubes so the length of the lower tube castings are appropriate for a 24" rim. Potential benefit is that this produces a shorter fork length, exactly matched to wheel size. I will get some pictures and description of my mag21 conversion posted here after I have completed the conversion and had a chance to ride it myself to make sure it is kid-safe.
Sounds interesting. Keep us posted.
I'd rate the Mag 21's as OK but not great. I've done the long travel mod and I've got the stanchions up higher through the crown to get the same A-C as the stock 24" fork. To get full travel out of them my 30kg daughter needs between 25-30psi in one side only. If I pressurize both sides to any degree she can't get full travel out of them. If I presurize one side to more than 30psi, the lack of a negative air spring means that there is too much initial resistance for a plush ride. I just need to find the time to pull them apart and remove all damping from one side, making a larger air chamber for a less progressive air spring and then just run damping in one side. I'll probably make an air cap with shraeder valve as the stock needle fill valve / rubber grommet system sucks. I'm not sure at this stage but I've got a sneaking suspicion that I may need to devise some sort of negative spring (probably elastomer or coil spring) to get the fork working as plush as I like it.
Otherwise I may just go with a 28mm SID.
I've begun some improvements on the Mag 21, starting with the bogus air valve / fill method.
I've got a few more mods in the pipeline, so I'll keep that page updated.
That's an interesting approach GrayJay. I'm not sure how thick the internal casting is between the upper chamber and the drop-out chamber. There is a warning in the Mag21 service manual that reads
"NOTE: PULL UPPER TUBE UP SLOWLY AND CAREFULLY OR A VACUUM MAY
FORM, CAUSING THE BOTTOM PLATE TO BECOME LOOSE. IF OIL IS
VISIBLE IN THE DROPOUT HOLE, THE LEG MUST BE COMPLETELY
DISASSEMBLED AND THE BOTTOM PLATE REINSTALLED."
This seems to suggest that the bottom plate (what you call the separator plug) is removable. If this is the case, then your new dropout may not have anything solid to push against. The epoxy may be enough to hold it in place, but the way the legs are tapered may allow the hole thing to move upwards and leak or fail. Perhaps you can pin it in place or make the dropout so it has a shoulder that rests against the bottom of the leg outer.
With hindsite, I would agree that leaving shoulders on the new dropout to solidly locate against the cut-off end of the casting would have probably been an even better idea. That said, I am pretty sure that all the epoxy will hold it in place fine. The aluminum piece extends 60mm into the void and is completely encased in epoxy and I was carefull to clean the internal surface of the casting to achieve a good epoxy bond. I will test it out throughoutly, ride it hard with my own body weight before I let my daughter use it.
THe 1st generation marzocchi fork (with the thin lower leg castings) would probably have worked better for this shortening approach if none of the swiss-cheese holes in the casting happen to overlapp badly with the new shortenen dropout location.
later edit - I recently dissected a dead '97 Rockshox Indy fork and discovered that the lower portion of the slider casting on the Indy is solid (unlike the hollow mag 21). The Indy forks would be a good candidate to shorten for a 24" wheel. I also completed a similar 24" conversion with a manitou spyder fork, see post 7 of thread 24" air fork
Last edited by GrayJay; 06-05-2014 at 01:25 AM.
Yeah, I was on the look out for quite some time for one of those earlier Zokes but couldn't find one.
My daughter's got a few rides on the modified fork and it's pretty good except for some initial "stiction" which is probably due to the lack of a negative spring. As you've mentioned, the top out springs are too strong to really compress much with only 30 (or even 40psi) in the fork. Some softer, longer springs would be ideal. I also need to reduce the volume in the air chamber so the pressure ramps up faster. For testing purposes, its fairly easy to just add more oil, but some sort of lightweight drop in spacer for the inner leg would be ideal. I'm not sure whether there is some oil resistant closed cell foam that would do the job.
I would think you would still get some pressure compaction from a closed cell foam. There are industrial suppliers for hollow plastic spheres, maybe filling the stanchion tube up with 10mm spheres would sufficiently reduce the compressable air volume.
Have you yet tried gutting one of the valve bodies from one of the two legs, running oil dampening in one side only? Do you still have the top-out springs on both valves?
I dont have the factory approved service tools and couldn get the valve body loosened from the stanchion tube when I tried with regular hand-tools so I have not yet tried modifying/gutting a valve yet.
I am only running damping now in one side. I have left the valve body in place on the air-only side along with both top-out springs.
To take the valve body off you need a good quality pin spanner and some heat. Use the pin spanner in the two larger holes. The valve body is loctited in from the factory and a fair amount of force is required to break it free even after you have heated it up. I left the valve body in place on the now air-only side because if you removed it you would need to make up something else in its place to prevent the fork leg from being able to come out of the lowers. For the air-only side I did remove the valving adjuster that runs through the middle of the valve body.
Those hollow balls look interesting.
REI Novara Pixie drive train component upgrades
Awwww, heck. And I thought I was done.
I recently replaced drive train components on a 2012 Pixie because my daughter was not strong enough to muscle the grip shifters into the two largest cogs. I wonder how many ~6-7 year old girls can. And, she could not make it up the hills on our street with the 28 tooth cog. Doh, dropping weight from the bike didn't occur to me.
The following component changes made the bike more enjoyable for her.
Replaced Shimnano SL-RS25-6 6-speed Revo-shift shifter with Shimano ST-EF40-7 7-speed EZ Fire Plus shifter,
Replaced Shimano MF-TZ20 14-28T 6-speed freewheel with Shimano MF-TZ31 14-34T 7-speed freewheel,
Replaced Shimano RD-TZ50 6-speed derailleur with Shimano RD-TX35 7-speed derailleur,
Replaced KMC Z50 chain with KMC Z72 chain
Replaced Kenda K831 knobby tires with Kenda K841 Kontact tires for paved surfaces.
A frustration for me is that the *retail* price difference between stock and the components I chose is about $20; $20 making the difference between a frustrating bike and an enjoyable bike for my daughter.
What do you think the weight reduction improvements mentioned so far would change the retail price which is currently between $235 and $280 depending on sale discounts?
Some other parts I added are SKS Rowdy Kid's Fenders, Mirrycle Incredibell Original Bicycle Bell, Zefal Bicycle Spy Mirror, Fuelbelt Aero Fuelbox. If the Fuelbox is a bit small, consider the Timbuk2 Goody Box Top Tube Pack.
Last edited by atlr; 08-16-2013 at 06:48 AM.
Atlr- Nice start on your Pixie upgrade. I have subsequently also changed out the grip-shifter for a indexed 6-speed thumbshifter that my daughter has a much easier time operating.
It is disappointing that the pixie 20" didnt come setup with more functional selection of components. THe 3 pound crankset with too large of a chainring and too long of crankarms is particularly low end for the bike. Most of the rest of the components on the bike are easy enough to replace with take-off parts upgrades from adult bikes but the american BMX bottom bracket was a hassle to deal with in order to upgrade the crank. The rest of the pixie frame construction seemed to be of decent quality but in retrospect I probably should have waited and found a threaded BB 20" frame to start with.
I think probably the easiest weight reduction step for least cost on these bikes I would reccomend starting with would be replacing the steel handlebars and stem. The steer-tube on my older pixie was 1" (22.2mm ID) and setup for older quill type stem. I couldnt easily locate a used/cheap 1" MTB quill stem that was short reach and lighter weight so I resorted to using an old road stem setup for 26mm bars and adding a beer-can shim to hold the aluminum 25.4mm bars tightly.
I took my 7-yr daughter out riding in a park last week and without any prompting or direction from me, she hopping off a curb and excitedly yelled out that she noticed the suspension fork compressing and cusioning her landing, made me proud!
Slight trouble I have now is that her little sister that turns 5 soon is very discontent with her singlespeed BMX style 16" bike and wants to inheret the pixie before I have the older kid moved up at a 24".
Very cool that your daughter is becoming one with the bike.
Can you point me to URLs for parts for an "inelegant" solution that results in shorter cranks on this bike? My daughter is about the same height. I think the crank and BB of this 2012 Pixie are still the same parts as the ones you replaced - at least the right crank and chainwheel appear to be one piece. I have not replaced a BB before and do not have machining skills.
Originally Posted by GrayJay
For just going shorter (but not much lighter) you can get a shorter 1 piece crank from a smaller kids bike.
To go to 3-piece crank conversion, you need something like;
Universal Cycles -- Truvativ American to Euro BB-Shell Conversion
(and then add separate threaded MTB bottom bracket)
American 3-pc Sealed BB by PRIMO
(but 127mm is awefully wide spindle for a MTB crank).
Then add short sinz cranks or shorten an adult set.
Sheldon Brown's legacy will guide my hands.
One-piece ("Ashtabula") Cranks
What an amazing resource.
Last edited by atlr; 06-03-2013 at 05:04 PM.
Too bad mass produced short cranks are a rarity in the USA.
Here are links to sources I have found so far.
Sinz - DRIVETRAIN, EXPERT CRANKS, SQUARE
Crank Arm Shortening
Burley puts 140mm Prowheel brand crank arm on the Piccolo. Prowheel offers a variety of sizes and a couple of Prowheel models have double borings for pedals. Alas, Prowheel appears to be only an OEM supplier.
102, 127, 140, 152, 165, 170, 175mm lengths http://pro-wheel.com/main/product-item/a002n/
120mm and 140mm pedal mounts http://pro-wheel.com/main/product-item/a007n/
127mm and 152mm pedal mounts http://pro-wheel.com/main/product-item/a008t/
I am leaning toward replacing the 150mm one piece crank with a 140mm one piece crank. The 140mm one piece crank is also described as a 5 1/2 inch size. I am not particularly concerned with weight.
I found a source for JIS square taper crank arms with two mount points for pedals - Trek. You must ask a local Trek dealer to order these cranksets as a repair part - no online availability. I was charged about $36. I only bought the 140mm/120mm. I have read that the 152/127 is lighter.
Trek part W292771 140mm/120mm pedal positions 32T
Trek part W298798 152mm/127mm pedal positions ?T
Pretty impressive stuff! I'm assuming you already have experience machining parts ect... I'm guessing you have to be pretty good at that to get something like this to work.
Another 140mm square taper cranks option I noticed;
Sram Truvativ Racing Jr Single Speed MTB Crankset 40t 140mm Black Square Taper : Amazon.com : Sports & Outdoors
It comes with a 40t chainring but likely is 110mm BCD so could be changed for a 34t.
My neighbors kid has a trek with those 120mm/140mm dual-position cranks, definitly are made from heavy steel. I would go with the 127mm/152mm model as those are reportably made from aluminum, probably 1/2 the weight.
By sdkjr in forum Clydesdales/Tall Riders
Last Post: 07-01-2013, 09:24 PM
By patineto in forum Shocks and Suspension
Last Post: 03-22-2013, 10:13 PM
By eplanajr in forum 29er Bikes
Last Post: 02-21-2013, 05:34 PM
By m-dub in forum Families and Riding with Kids
Last Post: 02-16-2013, 12:30 PM
By JackStephen in forum Mexico
Last Post: 08-27-2011, 02:59 PM