Paging Mitch Harris: High trail and bar bags?

[satanas]

Mitch has said more than once that he thinks handling with bar bags and "mid trail" is poor, and that it's better with more or less trail. My forthcoming frameset is going to have to deal with a fair bit of dirt at times, and also a bit of MTB territory; my plan is to have fairly high trail for stability. With a 72° head angle and 45mm offset that would give trail from 60mm with 38mm tyres up to 65mm with 54mm; with 71.5° it would be 63-68mm.

What I'm wondering is what the minimum trail necessary to give stable handling with 38mm Écureuils is likely to be - bearing in mind they'll be on Crest MK3 rims, so gyroscopic forces will be pretty low - 341g rim + 240g tyre + 95g tube = 676g, versus say 950g with an average rim + Hetres. Écureuils are the worst case scenario for instability but I intend to use them at PBP so that combo needs to work; I'm considering getting a medium Swift Ozette, which would be on a custom rack as close to the head tube and as low as possible. (NB: More stability with wider tyres won't be a problem - for me, YMMV.)

After I have the wheels built I can maybe mock something up to do a test, but I haven't ordered the hubs yet as I've been debating what rear to get - it needs to be 28°, 142x12, Centerlock, 11 speed and reliable; quiet and light would be nice too(!). ;-)

Later,
Stephen

[mitch....@gmail.com]

I’ve done two mid-trail bike builds that did well with a front bag-on-rack. One was a standard race/sport geo RJQuinn 73 head angle 45mm offset = 56mm, typical mid-trail. 5 lbs in Berthoud 25 sitting on a Gran Comp Ene mini-rack felt neutral enough and I could ride no-handed, not quite as blithely as no-handing a low trail bike with front load (for me at least; I know people have varied experience with this no-handed question). I rode with as much as 14 lbs in the bag and anything over 10 lbs requires keeping my hands on the bar and tending to steering a bit more than on front-loaded low trail or on mid-trail with no front load. This mid-trial bike did fine with a front bag & 700x26 and even better when I converted it to 584x39mm (first batch BSPs that measured 39 on Compass rims).

I think the key for mid-trail front loading is keeping the weight as far back and low as you can—as close to the steering axis as possible. The Ene mini-rack allowed the bag to sit well back close to the C-P brake cable.

I think weight-close-to-steering-axis is probably always a good guideline but because low trail by definition is less reactive to weight shifts, my experience suggests low trail accommodates larger bag loads and further from the steering axis without much effect on handling. High trail is also good for front loads, not because it isn’t reactive to weight shifts (it is and has high flop numbers), but because high trail at speed develops lots more self/steering stability than mid-trail does. So with high trail at low speed (starts/stops) you may notice a bit of wrestling with a front bag load, but once up to any speed you get the high trail stability that settles a front bag load and it does fine. (Some of this applies to front pannier loads too, but that varies with pannier position and set-up. Front bag loads on a rack above a wheel are different enough that I always specify.)

Your minimum 60mm trail figure is on the border of high trail for me. I know high trail goes much higher than 60mm but I start to notice some high trail stability in the low 60s. With 38s you’ll have some pneumatic trail so I would think you have a good prospect of decent handling with a front bag load.

My other mid-trail build with good front bag handling is my recent disc conversion of my old Santa Cruz Bontrager with RTPs. The disc conversion uses a shorter-than-stock 45mm offset rigid fork that lowers the BB and steepens the angles—I calculated trail at 58. I installed a VO Passhunter front canti-mount rack. This disc fork included removable canti-mounts so I removed them and it allowed me to install the rack 25-30mm further back than a typical canti-mount rack. (Along with crown mount modification to keep the rack level that far back.) The VO tombstone is 27mm from the headtube.

I wondered how this mid-trail bike would handle with a front load and whether I’d be wanting to order a custom high-offset fork if it turned out to be tippy, yank-y, divey steering like I’ve experienced before on mid-trail bikes with a front bag (mounted not close enough to steering axis). Handles great, even with 15 lbs in the bag, even at slow speeds. No handling sensitivity, I can sit up and ride no-handed with no carefulness about the front load. While going slow no-handed on safe pavement, I tried to unsettle the bike, exaggerating my weight shifts back and forth trying to shake the loaded front end off its line and into a veer (and ready to grab the bar), and it didn’t work. Just rides along happily, stably.

For contrast I’ll mention I rode this same Bontrager for a while with a rigid fork conversion using cantis before I did the disc conversion. Same axle-crown dimension and offset. But in this canti version of the Bontrager, a front bag made the bike noticeably more sluggish and slowed steering. The difference? The same VO rack was mounted an inch or so further forward on the end of canti-brakes. Front bag weight was further forward; almost everything else was the same.

This back-to-back comparison is only an n-of-one but it tells me that moving weight back toward steering axis makes a lot of difference for mid-trial.

Mitch
in Utah

[satanas]

Thanks - that's super helpful! ;-)

Since the new bike is going to have discs and will need a custom rack made to fit the mount placement on the carbon fork, I can't see any reason why the bag shouldn't be as low as possible or need to clear the head tube by more than a few mm, but we'll see once I get the frame. It will have a rather long top tube and shortish (80-90mm) stem too, to get the front centre long enough to avoid TCO without slacking off the head angle too much.

There's going to be quite a bit of juggling to get everything to fit, hopefully not literally!

Later,
Stephen

[Alex Wetmore]

I can say from experience that getting the bag close to the headtube takes some planning.  You'll want to run as short of a stem as possible so that the bag can be opened.  A bag may not fit if the headtube is too short as well.  I think that most commercial handlebar bag racks sit far from the headtube just to work with as many stem, handlebar, and headtube geometries as possible.

I would draw it all up before the frame is going to manufacturing.

alex

---

From: 65...@googlegroups.com <65...@googlegroups.com> on behalf of satanas <nsc.e...@gmail.com>
Sent: Tuesday, September 11, 2018 1:23:14 PM
To: 650b
Subject: [650B] Paging Mitch Harris: High trail and bar bags?

 

--
You received this message because you are subscribed to the Google Groups "650b" group.
To unsubscribe from this group and stop receiving emails from it, send an email to 650b+uns...@googlegroups.com.
To post to this group, send email to 65...@googlegroups.com.
Visit this group at https://groups.google.com/group/650b.
For more options, visit https://groups.google.com/d/optout.

[mitch....@gmail.com]

On Tuesday, September 11, 2018 at 2:33:49 PM UTC-6, Alex Wetmore wrote:

I can say from experience that getting the bag close to the headtube takes some planning.  You'll want to run as short of a stem as possible so that the bag can be opened.  A bag may not fit if the headtube is too short as well.  I think that most commercial handlebar bag racks sit far from the headtube just to work with as many stem, handlebar, and headtube geometries as possible.

+1 

Some details about the two mid-trail bikes I described with good front-bag handling: The disc conversion Bontrager I described has more vertical room from rack to handlebar center (about 25-26cm) because 559 tires sometimes allow a lower rack height (this one apparently does, partly because the BB is a a little higher and the saddle to bar drop is the same). A Docena and a Berthoud 28 slide onto the VO tombstone and the bag leans lightly against the handlebar and 10cm Nitto stem. So the height works and the rear-biased rack doesn't require the bag to fit under the handlebar, or require a shorter stem. When I get the decaleur worked out for this bike, it will push the top of the bag forward a cm or two.  

The RJQuinn with a simple Ene Gran Compe mini-rack uses a longer stem and the bag would have had to be pushed too far forward for my taste. I've seen rando bags pushed forward from the tombstone to mate up with a decaleur on a longer stem, and I don't mind a little bit of that--the decaleur doesn't have to be exactly above the tombstone. But on this RJQuinn the stem head was too far ahead of the mini-tombstone. The Berthoud bag I had at the time was a 25, their middle size, and it fit nicely under the forged stem. Although a Matchak plug-in decaleur would have easily accommodated a bag that fits under the bar and stem, this was before he developed it, and I used an inverted VO decaleur. I had learned from BQ that the proper purpose of the decaleur is to stabilize that bag side to side and that it works best if it's not supporting much of the bag's weight. So this inverted prong decaleur was a good test of that principle. I mounted the prong rail to the bag with prongs pointing up and mounted the VO decaluer fairly close to the underside of the stem. The bag went on by sliding the prongs up into the decaleur rather than down into it. Worked fine. A little more fiddly to put the bag on and off, but not bad for a conversion on a budget. The real problem was that the bag's cover was partly obstructed and that made access more fiddly. Not ok for a long-term arrangement, but it was good for me at the time for seeing if a front bag and rack work on the bike I had.

Stephen, given what Alex says about the planning required to get the bag/rack close to the headtube, this is only something I'd consider for mid-trail. My low trail bikes (and even my high trail bikes) work fine with bags in the standard position, an inch or so further out from the steering axis. But if you want to use a mid-trail bike with a rando-style front bag it may be worth it. The comparison on my Bontrager's two recent builds with same mid-trail geometry but the bag 25-30mm further back was very noticeably improved. 

Mitch

in Utah

[satanas]

Thanks Alex and Mitch! I'll try drawing things up first, then mock everything up on something out of the shed; I've got bars, stems and racks to fiddle with. If things look dicey I guess the bag can be a little further forward, but I'd think further back can only help handling/low speed stability, assuming everything fits together okay. I'd say it's unlikely the bag will fit below the bars; I expect their tops will be at about the same heights.

Assuming reach to the bar centre doesn't shrink much - and it shouldn't need to - things should be okay with squeezing the bag in. Current effective top tube is 575mm, with a 120mm (112 effective, C-C) 3T quill stem, so 575+112=687 total top tube + stem. The reach to the hoods is a bit too much with the newer, larger STI levers, but the handlebars will be going from 90/140 (Merckx) to 77/123 (Ergonova), so -13mm reach which should fix that.

With way less offset (45 versus 74mm) and the same head angle, keeping the front centre the same will mean +29 on the effective top tube, so 575+29=604mm, and then 112-29 for the stem => 83; 80 is close enough (or possibly 90) which would move the hoods 16mm (or 6mm) further back; top tube + stem + bars = 575+112+90 = 777mm now, versus 604+80/90+77 = 761/(771) planned. Some tweaking may be needed.

The bag should be able to move ~30mm closer to the head tube before it starts getting closer to the centre of the bars, ignoring the bar top shape which will probably reduce the gap by ~5mm; with a 90mm stem the horizontal gap would be virtually identical to now. I'll try all this out before finalising anything, and in any case am waiting for Salsa to release details of their new Warbird v4 fork (and frameset/bike), due to happen at Interbike next Monday; axle-to-crown and offset might be a little different to Rodeo Labs' Spork, which is the current option.

I'll draw it up in Rattlecad tonight, then try a mockup tomorrow. :-)

Thanks again,
Stephen

[Brad]

Flop is the more informative number.

As you note getting the weight of the load close to the steering axis is critical.

The greater the flop, the greater the effort needed to bring a load far out back to center.  Eventually that will lead to muscle exhaustion.  So how long a ride is also a factor.

Jan's test with front panniers centered found less of an effect with low trail which explains why so many of the touring specific bicycle made after say 1979 were mid to high trail.  They were optimized for a handlebar bag that was for sunglasses and not for Nikon lenses,lunch and tomorrow's clothes plus rain gear and tights and tools. 

[Stephen Poole]

So, with the original GR fork trail was 31, and flop 9 and handling with all loads and at all speeds was IME dire. With the replacement fork that changed to 45/13 and things were much better all round. The first fork was aligned correctly, so that wasn't the problem.

The new frame will be either 72° and 45mm offset, or 71.5°/45, which gives numbers of 60/18 or 63/19 with 331mm tyre radius.

FWIW, my old Tony Oliver was 73° with 45mm offset which gives ~57/16, which was okay with a small Carradice bag attached to the bars. 54/16 was okay on another older touring bike too, with ~72° HTA, 57mm offset and 27x1 1/4" - that was a long time ago(!).

On the other hand, even a very light bar bag on an Alan Super with a 110mm Cinelli stem, 40mm offset and an unknown HTA was dire, and almost unridable out of the saddle as the bar bag wagged the front of the bike horribly.

Since I can't see any trend here it looks like it's "suck it and see," not the most scientific strategy. :-(

And I have Micro 4/3, so don't have to worry about Nikon lenses.  ;-)

Later,

Stephen

[Ken Freeman]

Looking from the side at pictures of low-trail bikes with bags or at least racks installed, it always strikes me that the center of the bag seems to be behind the front axle, if you drop a plumb line to help you visualize.  This also means it is behind the centroid of the contact patch, at least when it is not distorted by motion or force transmitted such as braking or large wheel drag (an aspect of pneumatic trail, I think). Perhaps this is a more precise concept to use than "low trail, mid trail, high trail," and the various comparisons Jan has made between bikes of different geometries based on his annotated riding impressions.

I would hypothesize that you could put a loaded (8# or 4 kg?) front bag on a rack on an Italian racing bike and have no handling problems, if the center of gravity of the added load is "somewhat" behind the centroid of the tire contact patch.  Maybe I'll try that concept someday.

Ken Freeman, Ann Arbor, MI USA

--

You received this message because you are subscribed to the Google Groups "650b" group.
To unsubscribe from this group and stop receiving emails from it, send an email to 650b+uns...@googlegroups.com.
To post to this group, send email to 65...@googlegroups.com.
Visit this group at https://groups.google.com/group/650b.
For more options, visit https://groups.google.com/d/optout.

--

Ken Freeman
Ann Arbor, MI USA

[satanas]

I'm inclined to think that any mass carried in front of the front axle or behind the rear axle isn't going to improve handling.

If the main factor with bar bags was keeping their centre of mass behind the front axle, then an easy way to do so would be to slacken the head angle, but nobody seems to do this. Increasing fork offset will also move more of the bag's mass behind the axle centre, as well as (coincidentally?) reducing trail.

Going to a chopper-like head angle with lots of offset would enable *all* of a bar bag to be behind the axle centre...

As for front panniers, trail and handling, I'm not convinced there's any useful correlation. Back in 1983 I rode the Tony Oliver trans-Am with Tailwind panniers on a Blackburn FL-1 rack and had zero handling problems or soreness, so IME low trail is *not* necessary for good handling with panniers. The only bikes that have had handling issues with front panniers for me had lots of mass there, some of it quite a bit ahead of or behind the steering axis; those had heavy steering at low speeds, but I think that's got to do with leverage rather than anything else. The GR suffered from visible head tube twist and laggy steering on tight low speed turns, and tried to shimmy from 28-30km/h upwards at all times; with front panniers fitted these damped the oscillations, so things never got scary like they did with just a bar bag.

IME, the best way to get good handling with front panniers is to:

1. Keep their mass as small as possible(!)
2. Keep things as close to the steering axis as possible, rather than in front of or behind it
3. Use rigid racks and pannier attachments
4. Have a head tube which resists twisting

Later,
Stephen

[Eric Keller]

On Wed, Sep 12, 2018 at 12:12 PM satanas <nsc.e...@gmail.com> wrote:

I'm inclined to think that any mass carried in front of the front axle or behind the rear axle isn't going to improve handling.

If the main factor with bar bags was keeping their centre of mass behind the front axle, then an easy way to do so would be to slacken the head angle, but nobody seems to do this. Increasing fork offset will also move more of the bag's mass behind the axle centre, as well as (coincidentally?) reducing trail.

I have a relatively high trail bike that I use with a randonneur bag.  I had an interesting incident recently where I started to shiver because of a sudden cold rain squall on a warm day. The bike went into a forced speed wobble. I'm pretty sure it would have been uneventful without the bag, but I had trouble controlling the bike at relatively high speed.  That was the closest I have gotten to crashing in recent years.

I think there is an effect unbalanced mass around the steering axis which should be called "flop" but that word is already taken.  A randonneur bag will obviously want to force the fork sideways to some degree.  Lower trail just affects how far it wants to go.

Eric Keller

Boalsburg, Pennsylvania

 

[Alex Wetmore]

Most 80s through 2000s touring bike designs keep the same fork offset, but slacken the headtube angle, to both increase wheelbase and to get fork-carried loads closer to the center of the turning axis. You don't have to search too hard to find ad-copy showing this.

The three touring bikes sold during this period (big as in you could easily buy them) are the Trek 520, Cannondale T2000, Rivendell Atlantis and Surly LHT (more or less an Atlantis clone), and they call use 71-72 degree HTAs in medium/large sizes.

I still find low trail bikes to handle better with moderate to heavy loads, because there is a lot less wheel flop.  I know that you have different opinions about low trail and don't wish to stir that up again.

alex

---

From: 65...@googlegroups.com <65...@googlegroups.com> on behalf of satanas <nsc.e...@gmail.com>

Sent: Wednesday, September 12, 2018 9:12:26 AM
To: 650b
Subject: Re: [650B] Re: Paging Mitch Harris: High trail and bar bags?

 

[Stephen Poole]

On Thu, 13 Sep. 2018, 03:44 Alex Wetmore, <al...@phred.org> wrote:

Most 80s through 2000s touring bike designs keep the same fork offset, but slacken the headtube angle, to both increase wheelbase and ***to get fork-carried loads closer to the center of the turning axis***

^ I'm afraid I don't understand the latter part of what you've said above - can you explain please?

[Alex Wetmore]

A slacker HTA tends to position most production lowrider racks farther aft on the fork, aligning the bags with the center of the steering axis.

alex

---

From: Stephen Poole <nsc.e...@gmail.com>
Sent: Wednesday, September 12, 2018 10:51:46 AM
To: Alex Wetmore
Cc: 650b

Subject: Re: [650B] Re: Paging Mitch Harris: High trail and bar bags?

[Stephen Poole]

Okay, gotcha I think. If the bag stays in the same place fore aft on the top of the rack, then slackening the head angle effectively moves the axle forward relative to the bag, yes? (Or, put another way, more of the bag ends up behind the steering axis.)

[John Hawrylak]

Ken

Your idea makes a lot of sense and does not require categorizing the trail.  Thanks

John Hawrylak

Woodstown NJ

[mitch....@gmail.com]

Ken said “Looking from the side at pictures of low-trail bikes with bags or at least racks installed, it always strikes me that the center of the bag seems to be behind the front axle, if you drop a plumb line to help you visualize. This also means it is behind the centroid of the contact patch, at least when it is not distorted by motion or force transmitted such as braking or large wheel drag (an aspect of pneumatic trail, I think)...”

So measuring distance of bag-center-weight behind the contact patch(?)...

“..Perhaps this is a more precise concept to use than ‘low trail, mid trail, high trail,’...”

How is that more precise than low, mid, and high trail? If I can feel significant handling differences in a 5mm change of trail and 3mm of flop, that tells me it’s pretty precise. I can’t even get close to feeling a 5mm difference in chainstay length, and know from experience that I often haven’t noticed a 5mm crank length difference, or BB height difference (asterisk). I say this as someone who really likes bikes with each of these trail designs, so this is not a low-trail partisan perspective. I’ve consistently noticed 5mm/3mm trail/flop difference on unloaded bikes so its also independent of load.

Like Stephen, I agree with you that putting weight over or ahead of a front axle generally isn’t likely to be good for handling—any vehicle will carry a load more happily between its axles than beyond them. But I think the more precise effect of front bag load on handling has to do with the leverage that load can put on the steering.

Low trail (and low flop) by definition and geometry allows a front load to have less leverage on steering which is why many people find it provides good handling with a front bag. And it’s why low trail is less fussy about how close you put the load to the steering axis. Mid-trail with higher flop gives a front load lots more leverage on steering so you would expect it to be more sensitive to how close you put the load to steering axis. My mid-trail front-bag builds described in my post above confirmed this for me.

Distance of a front bag load from the steering axis is a kind of lever arm on the steering. A longer one naturally affects steering/handling more. A trail design like mid-trail with more flop that allows weight shifts to have more leverage on steering also increases the effect of a given lever arm on steering/handling. That’s (likely) why I’ve found that keeping the lever arm shorter on a mid-trail bike improved noticeably.

I think this is the same effect Eric describes—too long a lever arm can make a front load flop the steering side to side, especially in a design like mid-trail that helps weight shifts affect steering/handling more. Since the name “flop” is taken, I’d suggest just calling it “lever arm” or “lever length”.

Mitch
in Utah
asterisk: I know others are more sensitive to crank length and BB height so I’m not saying no one can feel 5mm in those. I’ve just been surprised that I often don’t, but still really notice the same few millimeters change in trail/flop.

[Ken Freeman]

I think it's a more precise way of connecting a subjective phenomenon, the perception of a change in handling, with a concrete bicycle design parameter.  To measure the distance of the visual CG of the bag from the visual center of the contact patch is observable and measurable.  To measure geometric trail is also observable and measurable.  But we don't really talk about trail, we try to categorize it as "low trail, mid-trail, and high trail."  But there is no written assumption for those ranges.  So it is highly imprecise, though perhaps our perception of a change in handling is precise.

"Low trail (and low flop) by definition and geometry allows a front load to have less leverage on steering which is why many people find it provides good handling with a front bag."  I see this as a hypothesis about what happens to make riding with a front load easier with certain bikes.  Usually, hypotheses are not considered theories (as close as we can get to fact) until they have predictive value that cannot be contradicted logically.  If a trail value could be qualified as low or high (it cannot without a standard) and flop factor as well, we should be able to reliably predict that any bike with trail and flop that qualifies will behave well with a front load.  Ok.  We can estimate assessments of lowness or highness, but cannot set thresholds.

But then we have someone who seems also to be a reliable observer, who says he can ride with quite reasonable comfort on bikes that he thinks do not satisfy even a subjective assessment of low trail, but are rather, high trail.  So here's an exception.  High-trail/low-trail now cannot be a theory, but is downgraded to a hypothesis.  

I don't know if all of that (I THINK it's all in straightforward English) is clear.  My idea about distance behind the contact patch is an attempt to modify the existing hypothesis so it can admit Stephen's experience with bikes with higher trail values.

"Distance of a front bag load from the steering axis is a kind of lever arm on the steering. A longer one naturally affects steering/handling more. A trail design like mid-trail with more flop that allows weight shifts to have more leverage on steering also increases the effect of a given lever arm on steering/handling. That’s (likely) why I’ve found that keeping the lever arm shorter on a mid-trail bike improved noticeably."  Perhaps so, I don't see it as contradicting me.  With a longer or shorter lever arm for a given mass the CG of the bag with centered steering is going to move forward or backward. 

So I interpret that you are not disagreeing with me, except perhaps related to a concept of precision.  

Ken Freeman, Ann Arbor, MI USA 

--
You received this message because you are subscribed to the Google Groups "650b" group.
To unsubscribe from this group and stop receiving emails from it, send an email to 650b+uns...@googlegroups.com.
To post to this group, send email to 65...@googlegroups.com.
Visit this group at https://groups.google.com/group/650b.
For more options, visit https://groups.google.com/d/optout.

[satanas]

+1

Whilst I'm not going to disagree with Mitch's observations, the apparently somewhat flexible categories of low, mid and high trail have always bothered me, and people argue about which slot things fit into sometimes. And then wheel size and weight (moment of inertia) and pneumatic trail complicate things further, as do various loads.

Since Mitch says he can feel small differences, I think it's more accurate to view trail (and other factors) as giving a continuum where handling comes in various shades of grey, not as three distinct shades. That's my hypothesis anyway.

I suspect a useful description would probably require a 3d graph, including trail, moment of inertia and pneumatic trail - and maybe loading too, but there are only so many dimensions unless one resorts to something like string theory; my understanding of physics doesn't extend that far.

Later,
Stephen

[Ken Freeman]

I agree with Stephen.  I see any aspect of cycling experience as a continuum, within which individuals may have high sensitivity and discernment of small changes, to those who just don't feel it.  All are acceptable in the cycling world.

The three-level trail thing I see is an attempt to classify bicycle design.  But, I think that is also a continuum.  There's no law that says, for example, that 700c bikes must range from 60mm Trail/73 deg HTA, to 50 mm trail/72 degree HTA, to 40 mm trail/74 deg HTA.  Any point on that two dimensional graph can be built, meaning it again is a continuum.  The problem I and Mitch and Stephen are really talking about is, in that continuum of design options, are there regions of sweetness and regions of sourness, and can we define those a priori?  If we could, they might actually show regions.  But they just might not, as well, and if we don't have a clear language for talking about end even evaluating the resulting subjective performance assessments, then essentially zero clear and quantifiable results can be recognized.  We're still in the boat of "I like it" and "I don't like it, so that other guy must be wrong."  And if you read the recent blog recounting the decades of Nimrod bicycle reviews in a British periodical, you might be convinced we have not progressed since perhaps 1920.