[DeeEight]

To expand on the other thread about older suspension bike/fork setups, I thought I'd write this up... in the Feb 1992 Bicycling (an interesting issue to be sure as they had a Univega Team Shockblok featured also and its obvious now that it was the Boulder Defiant frame with new decals/paint) tested the five most popular spec'ed suspension forks of the model year... Scott Unishocks (the original white version that Ruthie Mathes won her world '91 XC title on) with 2" of travel, Answer Manitou 1 with 1.6" of travel, Trek DDS3 with 1.8", Specialized Futureshock at 1.75", and Rockshox Mag-20 at 1.84". They built this machine called the MONSTER that'd record the G forces transmitted to the rider at the handlebars by the fork over a trail ride (same rider/tire pressure, same line around the course, same speed), a 3-bump test (3 2x4s 20' apart in a parking lot, 12mph speed), and a 6" curb at 5mph. All forks were run at factory recommended setups and the ones with damping adjustments were tested across the full range to factor into the numbers. They also tested a conventional steel rigid fork (curved blades) to compare against.

- The rigid fork averaged 5Gs on the trail, 4.5 on the 3bump and 7.7 on the curb impact.
- The Mag-20 was 4.1, 4.0 and 4.3 respectively
- The Futureshock was 3.1, 3.7 and 3.3 respectively
- The DDS3 was 3.3, 4.1 and 2.9 respectively
- The Manitou was 3.5, 3.9, and 2.9 respectively
- The Unishocks was 3.2, 3.8, and 4.8 respectively

In other words, the Mag-20 was beaten in every category and by every fork EXCEPT the unishocks in the curb impact, and the dds3 in the 3-bump test. Yet somehow it was considered the top fork of the year by most magazines. Makes you wonder how many magazine editor bar tabs it cost paul turner to pull that one off.

In August 1993 they did another fork test, this time with 10 forks, but this series they changed the MONSTER tests to using a series of wood blocks on a flywheel and the bike with the forks was fixed to it and preloaded to simulate a 150 pound rider. They used 1", 2" and 3" wood blocks and recorded the G-forces the suspension fork transmitted to the bars when it encountered the different size blocks. They also measured the fork travel for each size impact. I'll try and get the g-forces right, they used these goofy bar graphs in this article, and didn't print the actual numbers. Again they tested an oversized unicrown rigid fork also, and got 3G for the 1", 7G for the 2", and the 3" there was no figure because the fork kept bouncing erratically.

- Rockshox Mag-21 scored 1.5G/11mm for the 1", 3.9G/15mm for 2" and 4G/20mm for 3" blocks.
- Rockshox Quadra scored 2.3G/6mm, 3G/19mm, 4G/28mm
- Trek Mogul Black Diamond scored 2.5G/8mm, 5G/16mm, 7.5G/24mm (now that's consistent)
- Manitou 2 scored 2.5G/7mm, 5.3G/19mm, 7.8G/25mm
- Future Shock Adjustable scored 3G/14mm, 6.8G/21mm and ?!?G/31mm (they didn't list a G figure for some reason for the 3", maybe it went off their chart that only went up to 8G)
- Schwinn Paiolo scored 2.5G/6mm, 6.7G/16mm, 6G/17mm (go figure, bigger impact, 1mm more travel, and lower G score)
- Tange Struts scored 2G/10mm, 2.5G/18mm, 7.5G/26mm
- SR DuoTrack scored 2.3G/7mm, 5G/27mm, 7G/28mm
- Marzocchi XC400 scored 2.5G/13mm, 5G/18mm, 7.5G/25mm
- Scott Unishocks LF 2.5G/6mm, 5.3G/12mm, 6.6G/17mm

Now later on in the same issue it has the included Mountain Bike magazine (this was back before MB split off into their own seperate publication) section where they tested that Unishocks LF at greater length and they listed for a 2" bump on the monster (but not exactly what sort of 2" bump they used) a score of 3.3G on setting 1 (of the compression lockout dials) for a Mag-21, 3.3G for a Manitou 2, and 3.5G for the Unishocks LF. So whatever the bump was this time, it wasn't the same as for the test series bicycling ran in the same issue.

Skip ahead another year to October 1994 and they test 10 more forks with the same 1", 2" and 3" bumps as the 1993 test setup. Except this time they don't tell us how much the forks actually compressed and they didn't actually test either the girvin vector nor the Amp F-1 as neither they claimed would fit their apparatus on account of the rearward extending linkages. This is amusing as the kind of square edge bumps they tested with coming head on at the wheel is EXACTLY the sort of impact these linkage forks excell at, and beat telescopic forks hands down with. So I suspect they didn't print the test numbers because neither company spent as much on advertising as Rockshox and Manitou did, and they couldn't risk printing how much better these forks were for the bumps they were testing the forks against. Who'd buy a Manitou 3 if the Girvin or Amp is shown to score 1/2 as many G's for an identical bump (especially when the Amp was $30 cheaper and only 1/10th of a pound heavier). Anyways, the test numbers were...

- SR/Suntour Duotrack 9001 2G / 4.8G / 5.2G
- Rockshox Quadra 10 2.8G / 4.2G / 7.4G
- Tange Pro Struts 1.9G / 5.2G / 5.3G
- Specialized FutureShock Adjustable 1.5G / 3G / 5.3G
- Marzocchi XC500 4G / 4.7G / 6.9G
- Manitou 3 1.6G / 3.7G / 3.8G (best in test they claimed)
- Scott Unishocks LFR 2.6G / 3.2G / 4.5G
- Rockshox Mag-21 SL Ti 1.9G / 5G / 6G

Edit : As i read thru old issues i find Mountain Bike tested the original Mongoose/Amp Research Amplifier fork in their April 1993 issue and gave a 4.0G figure for a 2" bump, with a Mag-21 at 3.3G and a Halson Inversion at 3.5G for comparison. So how come Mountain Bike managed to test the linkage fork on the MONSTER but Bicycling magazine couldn't ?!?

 

[grawbass]

Interesting article. Thanks for the writeup. :thumbsup:

but...

"In other words, the Mag-20 was beaten in every category and by every fork EXCEPT the unishocks in the curb impact, and the dds3 in the 3-bump test. Yet somehow it was considered the top fork of the year by most magazines. Makes you wonder how many magazine editor bar tabs it cost paul turner to pull that one off."

I have a feeling that low G forces was not the most important factor with most riders of the day. Coming off rigid forks, many people wanted a stiff fork with minimal bob and the Mag 20/21 excelled at that. (and by stiff, I mean compression wise)

 

[Sideknob]

Hmmm. I used to have a Mag 21, a Halson Inversion and have ridden a few of those early bumper forks. The Mag always felt better than the others. I'm suprised that horrible Quadra did anything at all - they only ever seemed to move about an inch, then bounce back about three inches at a rapid rate.

 

[uphiller]

I remember those tests, and am not so trusting of the results. Square-edged blocks of plywood are not irregularly-shaped trail obstacles.
I rode a Halson Inversion, and it always seemed like on smaller stuff it performed almost as well as a Manitou 3, on medium sized bumps it was king, and because of its long spring stack, it didn't pogo like some other older elastomer-based forks. Its spring stack was 7" long and had 1.9" of travel, whereas the Manitou 3 had 6" spring stacks and 2" of travel, i.e., it compressed the elastomers to a greater degree than the Inversion at full compression. (DeeEight may correct my numbers)
The AntiGravity fork I believe had a massively long 10" spring stack, but it was a serious noodle, worse even than a MAG21.
About the Quadras- as flexy as they were, and as bad as they were at actually absorbing bumps, they were better than many forks of the day, and I have seen people put them through some SERIOUS abuse that I don't think any other fork of the day would live through.

 

[DeeEight]

The manitou 3s actually ran two stages of compression bumpers. The primary upper spring stack was 6" but there was another 2" of bottom out elastomers in the lowers. You'd remove 1" of each bottom out stack and stick another 1/2" elastomer on the primary stack to extend the fork travel to 2.5" on those forks. When I reassembled my ChrisChance painted manitou 3 sliders onto M4 stanchions and shafts (my black M3 stanchions were just so worn out) I forgot to remove the bottom out elastomers so I dropped the travel on it back to 2". Maybe I'll get around to modding it back to 2.5" travel this winter sometime.

 

[Tracerboy]

Thanks DeeEight.I have the Scott LF Unishock on my Frishi Scott CST and I've been looking for information on the LF fork so thanks for posting this.
I'm curious about the recall. There's no details on why and where the forks failed. Was it because riders abused the fork or did they fail at the crown or sliders?
The steerer and crown look solid. The sliders are carbon fiber and are thin.At first glance,they look like plastic tubes. The fork dropouts appear to be magnesium bonded to the fork lowers.
I would imagine the forks would break just above the dropouts.
Can anyone enlighten me on the recall details? In the meantime,I'm watching out for the white Unishock,which is what the Frishi Scott originally had,going by a photo on Frishi's site.

 

[Zanetti]

In other words, the Mag-20 was beaten in every category and by every fork EXCEPT the unishocks in the curb impact, and the dds3 in the 3-bump test. Yet somehow it was considered the top fork of the year by most magazines. Makes you wonder how many magazine editor bar tabs it cost paul turner to pull that one off.

You, of all people, should know that the Mag series was highly tunable. Properly valved and set up with oil viscosity/height and air pressure, the Mag 20 (and more so the 21) could emulate the feel of any telescoping fork of the time.

 

[ssmike]

Hmmm. I used to have a Mag 21, a Halson Inversion and have ridden a few of those early bumper forks. The Mag always felt better than the others.

Which just shows that these tests cannot measure rider perceptions. And these tests assume that the "rider" is completely stiff-armed. They don't take into account the amount of "suspension" the rider has in their arms.

 

[DeeEight]

The mag series was highly tuneable for travel sure (and nobody tuned them for more travel than I did at 78mm on a Mag-21), but in no way could altering the oil deal with the fact that the damper didn't actually adjust the compression or rebound damping at all, it was strictly an adjustable blow-off lockout circuit. They also flexed like crazy and there was only so much you could do to compensate for that. I had to resort to a significantly thicker/heavier brace, a totally different crown, and a hybrid setup between Mag-20 and 21 bushings. MBA may claim to have killed the aftermarket brace industry with a single test which was rigged (by how they were testing) to put the stock cast Mag-21 in first place, but they didn't. What killed it was all the major manufacturers switching to intergrated braces.

The MBA test for brake stop flex had nothing to do with the flex the real world knows happens at the brake studs. In the real world, you'd be turning the bike, or swinging it side to side when sprinting, and the wheel would buzz the pads. This is the fork legs shifting up and down independantly of one another allowing the wheel to tilt enough to change its position. What did MBA test? They measured how much the brake cable housing stop flexed forwards when you applied the front brake which they explained would lead to a pulsing feeling as the brakes applied in a turn. Totally different thing and totally going to skew the results based on how thick the brace is, and more importantly how HIGH the cable stop is on the brace (taller stops mean more leverage to flex it forwards which means more deflection is possible). Most aftermarket braces were rather tall because they also were after increased tire clearance over the stock braces.

They also tested brake boosters seperately to see how much flex occurred in spreading the studs apart when one tester squeezed the brake lever. Any variance in how much he squeezed the lever would change the test results and they of course as with the brake stop test of the braces, they made no attempt to measure how much force he was applying to the brakes each time. The smart thing would have been to rig a hydraulic ram to the lever to push it back an identical distance and force each time, but MBA ain't Bicycling magazine and their test rigs were never as advanced, which is odd given Hi-Torque publications resources you'd think they could find some better equipment.

They also tested torsional flex by mounting the deflection gauge they used to the fork CROWN and then a torque wrench to the top nut of the headset, which they then tried to turn with 5, 10, 15 and 20 ft-Ibs of torque (tightening or loosening off the headset as it were.... instead of ohhhh I dunno... using a stem connected to a torque wrench which might have been more appropriate. But again, you wouldn't see much change in crown flex since the crown is the same in all the tests, and since the torque wrench's gauge wasn't anymore sensitive than 5 pound increments, any slight variance in the amount of torque would skew the results. Also they for some reason tested clockwise and counter clockwise torque forces (as if the brace would really care which way you turned) again ignoring the fact they were tightening and loosening the pressure on the headset more than anything else. Yet somehow this was important to a brace test and how it affected the ride of the fork.

 

[DeeEight]

Which just shows that these tests cannot measure rider perceptions. And these tests assume that the "rider" is completely stiff-armed. They don't take into account the amount of "suspension" the rider has in their arms.

MBA in their brace test mentioned in one small paragraph most didn't seem to read apparently (if MBA's claims the test killed the aftermarket brace industry are to believed anyways) than a national MX champion came back from testing a new shock, told the technicians it sucked, and the techs said the computer data showed it was the best shock ever tested on that course. So the racer goes "then let the computer ride the race".

 

[DeeEight]

The recall on the scott forks which affected the LFR, LF, VR, and S models only (the original unishock without a submodel and the unishock TX - which was the 700C model... with clearance to run as a 29er today - were not affected by the recall) and had to do with the dropouts potentially coming loose. All unishocks had aluminium dropouts that were pressed in and bonded to either aluminium or carbon fiber sliders. The intergrated braces were also pressed and bonded to the top of the sliders (only the LFR which appeared for 1994 as a "racer only" type fork had a bolt-on brace, which are was quite beefy looking). My friend emily (who is also the only person I know to personally wear out the brake tracks of Mavic M230 rims to the point of catastrophic failures) had the problem with her unishock LF. But everytime they'd loosen, her bf would just whip out the composite epoxy (he'd had to fix the aluminium inserts on several trimble frames before, as well as having made composite skateboard decks, so he knew what he was doing) and rebonded the things. When the recall was announced around 1995-96 range, Scott replaced the forks with Manitou Magnums. The original unishocks presumably weren't affected by the recall because either they were manufactured differently in some way (different factory or different epoxy used) or the CPSC never got a complaint on them. The Unishock TX's had cast sliders like with say, rockshox mag/quadra forks where the dropout is part of the casting (they still had the bonded on unibridge, but those never had anything to do with the recall).