slick carpet effect

News and information about the Neato XV-11 Robotic Vacuum. All discussion and troubleshooting questions go here.

Re: slick carpet effect

Postby glnc222 » June 9th, 2013, 9:11 pm

Reducing thickness 1/32", with the brush guard 1/32" instead of 1/16 off a hard surface, slightly slowed motion on carpet and required a recharge to complete, compared to single charge run above. Two layers of 1/16 mounting tape plus the 1/32 disk supplied with the sliders. Interesting it is so sensitive to this exact dimension.

Not clear there is a single configuration optimal for both carpet and hard floors with these results, for this cleaning head design.
Funny that when showing the Neato to someone they asked how it as adjusted for differences between these two types of floors, a common issue with full size vacuums.

[edit]using silicon sealer glue in the gap between the slider and the top (covered in scotch tape).
glnc222
Robot Master
 
Posts: 4443
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

My attachable solution

Postby Swede » June 10th, 2013, 3:53 pm

Here is my attachable solution.

I had problems on our bedroom floor which is completely covered with carpet and read about the great solution with magic sliders. Thanks for that! They were unfortunately hard to find in Sweden where I live so I bought a 100 mm cover for house ventilation (not sure about the english word, see picture) and modified it a little bit with saw and sandpaper. I filled it up with floor protection pads for furniture (word?) to get the correct height.

It worked really well on my carpet but I noticed that the performance on flat floor seemed to reduce so today I made a very simple attachable solution with velcro and this suits us perfect. The problem is of course that you need to split the cleaning procedure in two. One for flat floor and one for thick carpets but in our house that is not a problem and the magnetic strips comes in handy for another thick carpet in the living room.

I guess the pictures pretty much speaks for themselves. The reason for taping over the cliff sensors with paper is because it thought the black carpet was a hole and did not want to clean at all. The white paper solved that. In our ONE floor house that is ;)
Attachments
bild 4.JPG
bild 3 (1).JPG
bild 2 (1).JPG
bild 1 (1).JPG
ventil.jpg
ventil.jpg (16.09 KiB) Viewed 8858 times
Swede
 
Posts: 1
Joined: June 10th, 2013, 3:14 pm

Re: slick carpet effect

Postby glnc222 » June 11th, 2013, 4:34 am

Many thanks to member Swede for sharing another example of sliders and fine photography.

Measuring Slider Effect on Neato Position

A 1.75" long heavy duty sewing needle was inserted all the way through carpet and padding to the underlayment to measure effect of sliders on Neato's position on carpet. The height at the peak in the brush guard bulge, aligned with the back of the cliff sensor, is noted as a base reference. Relative differences with sliders is then measured.
sliderneedle.jpg
needle through carpet measures position on carpet

Apologies for difficulty photographing the hard to see needle more clearly.

The 3/8" thick 60mm round slider found effective raised the front 1/16" over the unmodified case.

Also observed was that on carpet unmodified, the wheels are extended out 3/8" from their position on hard surfaces, seated on the weight bearing stop plate; the edge of the wheel arm tip moved up even with the edge of the case on carpet.
This explains the main carpet problem in that the front is tilted down; the rear is raised as much as the front sinks in. The whole balance is shifted detrimental to motion. [edit] Laser measurements in next post show no tilt change; the whole unit is just raised or lowered. The real effect is on wheel contact force.

Effect of the slider on the wheel extension was not measured. Maybe the angle of the case top can be obtained with a laser presentation pointer or picture hanging laser level, locating position on a distant wall to amplify small angle differences. The angle effect was unanticipated, and I don't want to remove the slider again at present -- not fastened with velcro. Typical Neato surprises.

Accommodating both hard and carpet floors at once

The 3/8" slider lifts the brush guard 1/16" off hard floors, requiring removal given the importance of brush action.
An alternative is a thinner slider with a larger area which will be above hard floors but still lift the front when the unit sinks into carpet, engaging the slider for support.

The largest possible full width bottom addition, 5.5" long, was examined 1/4" thick. At 1/8" thinner than the round slider, a clearance of 1/16" is supplied on hard floors, preventing drag and scraping. Corrugated board is used just for measuring positions.
sliderfull.jpg
largest possible slider area


This full size addition does NOT raise the front at all, because it lifts the back tilting the front down again, canceling slider effect.

A half-size 2.75" maximum width area by 8.5" across was then measured, resembling round slider diameters.
sliderhalf.jpg
partial coverage broad area slider shape (in cardboard)


This shape lifts the front the same 1/16 in. as the 3/8 in. thick round slider.
At 1/4 in. thick, clearance would exist on hard floors. The round slider itself is 1/4 in. thick (without fastening material), so maybe the largest rectangular sliders could be cut and assembled into a working unit of this shape and size. The furniture sliders are used for the specially reduced friction of their material, though other materials are shown in user's posts.
No large sliders are on hand to check whether thicker than the smaller round ones.

[edit] funny the full 5.5 x 8.5 shape exactly matches a standard size used for publishing professional journals.

Alternative Method of Shifting the Tilt

The same effect as the slider might be achieved by limiting the 3/4 in. wheel arm extension mentioned which is tilting the Neato forward. A new mod limits this tilt from its standard extreme range, but is not useful for the carpet case. http://www.robotreviews.com/chat/viewtopic.php?f=20&t=17095

Unfortunately, tying down the wheels to the untilted position prevents any climbing over height transitions normally handled by lifting the front or back over obstacles. So in order to adjust tilt for carpets spring tension change would be the only possibility. Whether that would help and how it could be done isn't clear. A non-spring wire extension to the spring might be wound around a shaft at the spring attachment, with a knob or screw to set the tension by shifting the end of the spring back and forth. If I recall correctly the spring attachment is over an open space at the edge of the wheel well, so there might be some way to access it. It might take inserting a crank for leverage with several pounds of tension on that spring. Reminds of a crossbow.

So not worth the trouble with sliders already removable with velcro. It does suggest a possibility in factory designs, if not mods, of a servomotor adjusting tension by winding the spool, maybe using cliff sensor distance to floor data for tilt detection. As long as a factory redesign is involved, completely different cleaning head designs might make it all moot.
Last edited by glnc222 on June 13th, 2013, 12:57 am, edited 3 times in total.
glnc222
Robot Master
 
Posts: 4443
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: slick carpet effect

Postby glnc222 » June 11th, 2013, 3:36 pm

possible 1/4" hard/carpet product for above post, in a single piece
Waxman 4704095N 9-1/2-Inch by 5-3/4-Inch Reusable Oval EZ Sliders
http://www.amazon.com/Waxman-4704095N-2-Inch-Reusable-Sliders/dp/B001W6Q4VA/ref=sr_1_5?s=hi&ie=UTF8&qid=1370978516&sr=1-5&keywords=waxman+super+sliders
Will have to see when arrives. Cut to size; the plastic shell will be thin regardless of rubber filling, likely thicker in large widths. "Magic" brand not made in large widths, made for permanent attachment instead of temporary moving, two classes of slider products.
Beware felt sliders only for hard floors including a 9" x 2 1/2" waxman.

[edit] a 1/4" varnished hardwood version moved on carpet pretty well, but a little slower than the plastic.
Last edited by glnc222 on June 11th, 2013, 5:06 pm, edited 1 time in total.
glnc222
Robot Master
 
Posts: 4443
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: slick carpet effect

Postby glnc222 » June 11th, 2013, 5:05 pm

Height Transition Effect Analyzed

Sliders fail to climb over some sharp transitions because of specific positioning:
stoppage occurs with the wheels up against the transition, with insufficient traction to climb.
Traction is reduced because the wheels are extended out on spring pressure without full weight on the wheels, and reduced running on carpet. The slider is over the transition and raising the front at a steeper than normal angle.
Ramped transitions have less effect on the angle.

[edit] There is a possibility adding weight over the wheels (lead in the dust bin?) might increase traction. Motor torque appears sufficient.

[edit] more position detail: the weight is on the rear end and the front over the transition, with the wheels hanging free. No wonder little traction. Would need a cam or something to push the wheels down to shift the weight onto them.
glnc222
Robot Master
 
Posts: 4443
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: slick carpet effect

Postby glnc222 » June 13th, 2013, 12:54 am

Laser Level Measurements of Slider Effect

The difference between Neato position on carpet and hard floors is measured by a laser level beam extended 20 ft.
sliderlevel.jpg
laser level atop Neato

sliderlaser.jpg
laser level projection from Neato 20ft distant


The effect of the slider is virtually zero compared to the angle difference positioned on hard floors and carpets (1/2 inch beam shift over 20ft). The angle difference between hard floors and carpets is a very small fraction of a degree, and the Neato is vertually level. The effect of the slider previously observed raising the front slightly is mainly to lift the entire unit on the carpet.

The previous observation of wheels extended 3/8" on carpet shows that the unit "floats" on the carpet supported by the front and back, with the wheels sinking into the carpet under spring pressure, with less than the full weight on hard floors. Pressure on each wheel is reduced from 2.5lbs on hard floors to 1.5lbs on carpet (the particular one here -- another here is too soft to even consider) -- a large percentage drop. Hence the lower traction on the wheels. The slider lifts the front slightly to reduce the drag especially from the squeegee scraper all across the front, compensating for the lower traction. Very small differences are involved.

Neato needs a dynamic suspension control to press the wheels down with the same weight distribution on both surfaces -- difficult to engineer. Or maybe higher tension springs and higher unit weight to match. Sport suspension?
[edit] Not out of the question: A 5.25" square steel plate 1/8" thick weighs a pound (lead is 45 per cent heavier); standard hardware pieces. Getting a higher tension spring is more complicated; maybe just shorten the spring, but what if some limit is exceeded? Spring specs unfamiliar territory. Some springs at SmallParts.com.

But more weight on the wheels might just sink them further -- so the solution on carpet may well be runners, or in effect suspension on sliders... As usual, more data needed.
[edit] confusing with all the balanced forces -- no idea yet what this would do.

(This reminds of an old experience with aluminum scuba tanks which float compared to earlier steel ones; the shop fashioned a lead disk inserted in the bottom "boot" rubber cover stand to keep it neutral, as the harness holds it most at the top, and the bottom would flip up in water.)

[edit] When weight on wheels is an issue, for light weight devices the answer could be increasing the area of the drive contact, using belt tracks like a tank, which might actually fit in the space using small diameter cogs. The wide radius of the wheels though, may assist in climbing obstacles.
glnc222
Robot Master
 
Posts: 4443
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: slick carpet effect

Postby glnc222 » June 17th, 2013, 8:16 pm

Cardboard Slider Sufficient -- No Plastic Required

[edit] Botvac versionThe same cardboard works for BotVac, a tad wider, single battery cover. Put holes through cardboard for screw access. Slightly larger brush guard, 1/8" deeper, might require a thicker slider; needs testing. One standard might be preserving the 1/16" clearance under the slider on hard floors, to get the same lift on carpets. Thicken as needed to get down to that. This limits how deep the squeegee across the front up can sink into carpet, reducing drag to compensate for weaker wheel traction on slick surfaces etc. Measuring a BotVac shows the same 1/4" thickness appears to fit the same; some difference in front-to-back length and slanted stance may be a factor.
For Botvac tape to the battery cover must be fastened down on the top, at least on one side, so the back tabs of the cover can insert in their slots, with the cardboard lifted up. The cardboard attaches behind the screws on the front edge, as it is behind the tongue of the brush guard.

cardboard attachment
The cardboard test fitting above 8.5" x 2.75"x.25" unexpectedly turned out a functional slider covered with transparent tape for a slick durable surface, and attaches under the edge of the battery cover without gluing to the case. The plastics are made for heavy furniture weights and need modification, with Neato only 9 lbs. routinely packed in shipping boxes. Plastic hardness might do something for height transitions, hard to tell.

The attachment can be used with any slider, extending thin stiff plastic out to the wheels and gluing to that instead of the case. e.g. slices from thicker bubble wrap packaging, take-out food containers, cutting board sheets. No polyethylene, used for glue bottles. Makes experimenting with different sliders easy.

sliderfinal.jpg
taped cardboard slider attached under battery cover

sliderbattery.jpg
attachment under battery cover

BotvacSlider.JPG
Botvac Slider Addition


Slider tape finish
For 2" shipping tape two strips join across the middle of the 2.75" diameter, wrapped around the edges for hitting obstructions. Heavy duty scotch shipping tape for perfectionists. Extended tape beyond the sides folds under the outside edge of the battery cover for removable attachment, though the battery cover must be opened. Otherwise velcro can be used for frequent changes. 1/4 inch size does not need removal for hard floors having clearance below.

Cardboard assembly
Two layers of thinner corrugated cardboard (e.g. Amazon and Digi-Key boxes) glued together, stiffer than thicker single types; strongest if corrugations at right angles, for perfectionists. Exact thickness matters with a 1/32" difference noted in prior post. 1/4" total thickness including attachment was the design, but I reduced slightly for better climbing heights on top of carpet, unusual. Adjust as needed with paper layers stripped of corrugations, manilla folder material etc. (matching Neato color...even painted). Loose sheets can be temporarily added for calibrating.
[edit] see later thread http://www.robotreviews.com/chat/viewtopic.php?f=20&t=17783; too thick degrades dirt pickup.

1/4" x 3 nominal (2.5 finished) precut lumber also usable for the carpenter in you.

Attachment to Neato

This wider slider shape accidentally supports removable attachment by extending the tape at the outside edges down under the outer edge of the battery cover, with which the slider side is flush. An additional bit of tape can go under the center tongue of the brush guard, covering the vacuum channel, to counter any bowing effects.

The wider slider will have to be removed to open the battery compartment anyway. Velcro can be used cutting out sections deep enough for both velcro sides, for flush mounting. UltraMate style mated pairs total 1/8"; products vary. A small square over or just inside of each battery cover should do.
Finally double stick tape, seemingly superfluous. Or regular tape can be used in alternating reversed strips stuck together immitating double-stick tape. Duck tape is thick enough to effect total dimensions and is less slick.

Performance

1/4" total thickness including fastening handled height transitions better being thinner, and leaves a 1/16" gap underneath on hard floors for standard Neato brushing. The larger area compensates on carpet, working similar to 3/8" total thickness plastic sliders which lifts slightly on hard floors. Carpets vary in properties. This slider even allowed Neato to move on a very fluffy bath rug previously impossible, though still very sluggish. The rear end also drags on carpet. The wider surface may still affect climbing, needs more experience in various environments -- up to the readers.

Effects on Neato movement

Spot-Clean time (in motion) measures travel speed differences, one measure of drag effects.

............................................Beater Brush..........................Pet (bristle)brush
Hard floor .................................. 1:34 ..................................... 1:35
Carpet:
Unmodified ................................. 2:41 ..................................... 2:28
Cardboard Slider 1/4", wide................1:58 .................................... 2:08
--- a second run ............................ 2:25
Plastic 3/8" 60mm ......................... 2:22 ..................................... 2:13

Walls and a furniture bump increased time slightly and the travel pattern does not repeat exactly. The carpet section available was well beaten down but still revealing. The manual mentions a 50 per cent increase in full charge run-time from carpet to hard floors. [edit] faster bristle brush speed on carpet may be from the brush adding to propulsion.

What with tissue paper filters and carboard sliders, next thing will be cleaning a house of cards.

Links to other Mods listed in http://www.robotreviews.com/chat/viewtopic.php?f=20&t=16475

[edit] another member reports success with this slider http://www.robotreviews.com/chat/viewtopic.php?f=20&t=17224
And another success report http://www.robotreviews.com/chat/viewtopic.php?p=122715#p122715
Another member mentioned help with height transitions in rooms, but no further details; might be climbing edges of area rugs?


"An educated consumer is the best customer."
Last edited by glnc222 on November 6th, 2014, 10:12 pm, edited 10 times in total.
glnc222
Robot Master
 
Posts: 4443
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: slick carpet effect

Postby glnc222 » June 18th, 2013, 9:56 pm

Full Width Plastic Slider

Waxman 9.5" reusable slider adapted. Ar Amazon http://www.amazon.com/Waxman-4704095N-2-Inch-Reusable-Sliders/dp/B001W6Q4VA/ref=sr_1_fkmr0_1?ie=UTF8&qid=1371603542&sr=8-1-fkmr0&keywords=waxman+9.5%22+slidr

sliderwaxman.jpg
Waxman 9.5" slider adapted


Too thick rubber filling easily pulled out, replaced by cardboard. Shell slightly over 1/4", easily cut with scissors. Attached over edge of the battery covers. Clear underneath on hard floor.
Spot clean time, beater brush, 2:21 on carpet.

Did not improve height transitions over the cardboard slider.

Rear Slider Improves Climbing

The drag problem is not merely the front; the rear is sunk, too.
Adding a 3/8" thick 60mm round to the rear improved climbing, but lifts the rear rollers off hard floors.
The same only 1/4" with clearance made climbing worse if at all.
A full width 1/4" rear slider with hard floor clearance improved climbing height transitions, at least on carpet -- furniture bases etc.
Spot clean time on carpet same 2:21.
I don't have a lot of hard floor height transitions to test.

sliderdual.jpg
dual end slider pairs, full width


Also in this case the front slider is moved back 1/2". This tested whether a concavity at the front wheel edge would prevent increasing the climbing angle. There was little effect with this setback alone on a single front slider, and moving back another 1/4" worsened climbing. Lot of subtleties to this vacuum thing. It needs something analogous to a wind tunnel for airplanes to do the engineering, best left to the factory.

This seems to exhaust the possibilities for improvement here, in this environment.

[edit] The Rear Slider has been found interfering with backwards climbing, so must be used with caution.
With the back sunk in carpet, the slider blocks the rear, unless shaped to get the back elevated.


[edit]Carpet Trapping Video
Excellent Neato operating video with thick rug problem half way through; but manages a fringed rug.
http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=3&ved=0CFYQtwIwAg&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3Dy021R6c4uPE&ei=xdasUpbPBabQsASMxoHIDA&usg=AFQjCNFsTYFAdPGaAKRpaZRwDBsOAJh-FQ&bvm=bv.57967247,d.cWc&cad=rja

More videos: One failure
http://www.youtube.com/watch?v=LWXzqZsm53E
and one success
http://www.youtube.com/watch?v=okrfS68rlVA
hard to see what makes the difference in being able to operate. The first might benefit from a slider mod.
Have to own a carpet dealership to access all the types for testing.
glnc222
Robot Master
 
Posts: 4443
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: slick carpet effect

Postby glnc222 » September 5th, 2014, 6:44 pm

Waxed Fibers Revealed

A Wall Street Journal review of robot vacuums noting a BotVac slipped on carpet, mentions industry sources said some carpet fibers are made with a waxy material on them.
http://online.wsj.com/news/articles/SB10001424052702304893404579531672245326210
One can imagine this increases resistance to stains and what not. Repeated vacuuming will polish these fibers and increase the slippery quality. One mystery solved.
Slider Mod to the rescue.
glnc222
Robot Master
 
Posts: 4443
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: slick carpet effect

Postby dcjenkins » December 28th, 2014, 11:32 pm

Hello, and thank you all, especially glnc222, for the detailed information.

I've attempted this cardboard mod several times, and my latest attempt seems to be the most successful, but still causes my Botvac to error and stop during a run. I should add that my latest version uses the Scotch brand heavy duty shipping tape, versus the generic brand of my earlier versions.

I'm attaching photos to show the current state of my mod and the specific heights of the cardboard. The photos will show that one end is slightly slimmer than the other, but both are under 1/4". I use repositional Glue Dots, http://amzn.com/B00LHZ0AS4, instead of the mounting tape. These dots are very slim, so they don't add too much to the overall height -- I'd say it's just at, or maybe a bit under, the perfect height of 1/4". Could this potential variance from a true 1/4" be the reason for this mod still causing the Botvac to get stuck on deeper pile rugs? I should note, it's better than the non-modded Botvac, which was hopeless.

P.S. Any thoughts on why 2 1/2" square carpet sliders, side-by-side, with the ends modified to fit the width perfectly, wouldn't' be a better solution than this cardboard mod?

Thanks for the help.
Attachments
carpet-mod-3.JPG
Front height of cardboard mod.
carpet-mod-2.JPG
Rear height of cardboard mod.
carpet-mod.JPG
Cardboard mod with Scotch heavy duty tape attached with Glue Dots to Botvac.
dcjenkins
 
Posts: 7
Joined: December 8th, 2014, 7:18 pm

Re: slick carpet effect

Postby glnc222 » December 29th, 2014, 2:29 am

Some really thick plush carpet cannot be helped even with the slider. The wheels are just not big enough, and the new Dyson uses a track like a bulldozer or tank. I have some fluffy deluxe bath rugs which the botvac can barely move on even with the slider, and is not cleaned by it. At the Neato scale movement is like driving a truck through a cornfield.

The cardboard was found just as a cheap and ready to hand material easily cut to size and also which could be fastened in a simple removable way taping around the edges of the battery compartment covers. Also by using a broad form the lifting effect is achieved with a thinner material allowing a clearance underneath when running on hard floor, so it does not scrape in mixed hard and carpet floor homes.

To adjust thickness of the cardboard I use multiple layers of board from certain types of packages with thinner board than the largest boxes, such as Amazon DVD boxes and Digi-Key electronic components shipping, only 1/8 thick, taped together various ways or even glued, or with double stick tape; around the edges suffices (tape the pack separately together from the attachment to Neato). The Botvac seems to have a thicker brush guard which might need a thicker slider.

The shipping tape is just more convenient being wider and more durable, thicker. The tape just gives a slick surface to the cardboard and protects it from getting torn up. Otherwise cardboard being processed wood, is very durable, and houses are now made with larger forms of composite wood pulp material.
The Neato should run without any tape for testing, especially on carpet which is not grinding the surface.

One effect on Botvac I now use is raising the front some when climbing edges making climbing more difficult. In some cases a quick-removal and install method might be wanted for moving the Botvac into different areas with different equipment.

I would like to see photos of the carpet more than of the slider, and even measurements of the plush thickness if possible. (Photos are much more interesting here than boring text, so thanks for taking pictures). Lay a book on the edge then distance to hard floor. Or press a thin rod, screw driver or something, into the plush pushed up on end instead of crushed, mark the spot to measure by ruler. Little is known about Neato operation on different types of carpet, as well as subtle difference between Botvac and XV performance on different surfaces.

It is a bit tricky to verify the clearance underneath the slider on hard floor after installed, because the brush guard sits down on the floor in front of the slider. A thin stick like a folding ruler may be inserted from the rear and see if it can fit under the slider. Also note the Neato rests on rollers in the front and back on hard floors, thereby locating the floor position on the bottom. A stick placed across these rollers front to back should represent the floor when the Neato is turned upside down.

[edit] You can make thinner cardboard by tearing off the surface of the thicker kinds, and by compressing it as one's full weight on one foot (reminiscent of old fashioned wine making harvests with the grapes).
glnc222
Robot Master
 
Posts: 4443
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: slick carpet effect

Postby glnc222 » December 29th, 2014, 4:16 am

Another Botvac point mentioned early in the Botvac thread I think was the squeegee blade across the bottom behind the brush is deeper than on the XV and will drag more on deep pile. I think someone actually trimmed it back some, as another thing to try. It's function is more on hard floors to direct suction forward, but gets completely buried into carpet. I would check to see if replacements are available first. Remember the warranty...
glnc222
Robot Master
 
Posts: 4443
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: slick carpet effect

Postby dcjenkins » January 1st, 2015, 4:54 am

Thank you for the reply. I didn't realize that anyone had responded. I was waiting for an email confirming any new posts, but never received any. It seems that I won't be notified by email, so I'll just manually check this thread.

I'll have to comb through the information you provided several times, to get it all, but I think I got most of it.

I definitely like the idea of hitting all options at once, being able to vacuum hardwood and carpet in the same run. I realize I may not have a choice, but this will be my goal until I find out it can't be done. So, it would seem I'll have to stick to the slimmer cardboard or carpet slider, in my attempt to do this. I have some other adjustments I've done, in my attempt to have a do-it-all-in-one-run Botvac, but I can address those on the appropriate threads.

I'm providing a photo of the depth of the carpet that is giving me the trouble. I actually had one run, with the cardboard mod in the photos provided earlier, that only gave me 2 errors -- which was really impressive. I haven't have a run that smooth since. I'm now getting clear my path and clean my brush errors, in addition to my wheel is stuck errors, which I realize may have more to do with just creating a successful carpet slider mod. I read on another thread, someone had to have their unit replaced in order to prevent the errors -- but, a new unit did solve it.

I did catch a little bit of info on the Vorwerk Kobold VR200. I wish I had the budget for that, which would seem to be an improvement on the Botvac and Roomba.

P.S. I see all I had to do was check the "Notify me..." box to receive email notifications of a new post.
Attachments
carpet.JPG
dcjenkins
 
Posts: 7
Joined: December 8th, 2014, 7:18 pm

Re: slick carpet effect

Postby glnc222 » January 1st, 2015, 6:09 pm

That certainly looks like one of the more extreme carpets never navigated by these small machines since the first Roomba's. The discontinued Electrolux had larger wheels to handle it, and now the Dyson with tank tracks.

To test the unit for defects it should be run in another area to see whether errors are generated. With such an extreme obstacle it is premature to assume the unit is defective. Getting stuck is going to cause error msgs. It navigates partly by tracking wheel motion assuming it is moving. They are not guaranteed to work on all surfaces.

An extreme measure which can be taken a bit expensive and complicated is replacing the tread on the wheels with an especially grippy type of material used to secure throw rugs to floors. This is harder on botvac because they changed the wheel mounting with the axle apparently secured inside the wheel cover and the screw moved to the side of the drive assembly. XV wheels might work, though I have not taken apart the botvac drive to see the axle (maybe something Vic would enjoy? The whole thing must be disassembled to get the wheels out, compared to the XV design.) XV wheels have been listed some places around $8 apiece, though previously sold only as part of the entire drive assembly. See thread http://www.robotreviews.com/chat/viewtopic.php?f=20&t=17228&p=117712#p117681
Increasing traction might allow the bot to move but it will still face much higher drag, and consume more power moving with reduced run time. Maybe the carpet is really old and you can use a makeover? Some enthusiasts have gone that far.

Since I am not using that material I tested I could send it to you and a couple of wheel hubs which might work, but I don't have axles, and it would be best if you could do some of the fitting work. Contact privately if interested.

Incidentally the Vorwerk is not distributed in the U.S., and not that much outside of Germany. I think the Vorwerk company prefers to sell through its dealerships to provide service and maintain its reputation for a multi-product brand, though just a guess. They do not seem to have pursued global marketing the same as other companies in the industry.
glnc222
Robot Master
 
Posts: 4443
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: slick carpet effect

Postby dcjenkins » January 1st, 2015, 10:06 pm

I do think bettering the traction is the best next step at this point, considering my first very successful run with the carpet slider. My thought was better traction along with the slider could be the answer. I have been pondering this, but as you say, it will be complex and potentially expensive. I do appreciate your offer to provide the materials, but before I do that, let me make sure I can commit to the mod. I'm not completely mechanically inept, but I would consider this a bit advanced for me, and may get me in over my head.

The Dyson does look intriguing, with the tank tracks, but the cost is typical Dyson -- way too high. But, instead of buying new carpets, you could argue the extra expense may be worth it. For now, though, I'll ponder how deep I want to go into this modding, and learn as much as I can with this Botvac.

Thanks again for the offer -- I'll let you know if I decide to go for it. In the mean time, I'll report back with any new successes or failures.
dcjenkins
 
Posts: 7
Joined: December 8th, 2014, 7:18 pm

Re: slick carpet effect

Postby glnc222 » January 1st, 2015, 10:09 pm

One other experiment is doubling the front-to-back length of the slider for more lift, but with a thicket as high as the wheel radius, no telling till tried. Certainly easier than starting a wheel project.
glnc222
Robot Master
 
Posts: 4443
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: slick carpet effect

Postby glnc222 » January 2nd, 2015, 5:59 am

No XV wheel parts and axles can be used directly in Botvac, though the wheels might be usable with a custom made axle carved from hardwood or something. Partial removal shows the Botvac attachment screw is an inch or so long sheet metal screw, not the machine screw used in the XV machined metal shaft. So the Botvac axle is integral plastic molding of the wheel, typical new elimination of machined metal parts. The drive attachment shape, possibly hex or something else other than the XV oblong, could not be seen without completely disassembling the bot. The sheet metal screw must not be over-tightened and fits into a round protrusion of the shaft out the drive assembly side, pressing a flange against a greased plastic surface.
glnc222
Robot Master
 
Posts: 4443
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: slick carpet effect

Postby dcjenkins » January 24th, 2015, 1:23 am

I'm still not getting notified when someone replies to my post, for some reason.

Thanks for the additional info on the wheels. I've given up on the high carpet, for now. I've actually purchased a boat load (five rolls) of the magnetic boundary tape to keep the Botvav off the carpets. I find that the boundary tape is not a perfect situation, but it's a start in the right direction. This little Botvac is frustrating -- I'm now dealing with phantom clear my path and my bumper is stuck errors. I've only had one error free run of my whole house. Looks like my expectations were too high.
dcjenkins
 
Posts: 7
Joined: December 8th, 2014, 7:18 pm

Re: slick carpet effect

Postby glnc222 » January 27th, 2015, 4:07 am

Warnings about limited use on deep pile carpets is a major omission from advertising for these machines. It takes a certain skepticism and some technical knowledge to evaluate such product innovations. You have to ask how does it manage to work at all? Time consuming search of the internet material can be needed even when the problems are pointed out in reviews.

At this point you can see very expensive models like a Dyson with much more powerful drive mechanisms, a tank track, which could raise questions about what other makes have.

You might note that most of the videos and pictures of operation are on the popular hardwood or laminate flooring in use today. So there aren't even that many reviews about operation on carpet.

Besides navigation there is a basic question of how they can clean much on carpets usually done with very powerful full size vacuums. The Neato gave surprisingly good results on that, though a full size is still needed from time to time, depending on the environment. The pet hair problems is separate and gives the machines some utility besides regular cleaning.
glnc222
Robot Master
 
Posts: 4443
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

Re: slick carpet effect

Postby glnc222 » July 5th, 2015, 5:27 pm

Botvac Slider

The Botvac acquired in the Fall with dry weather started slipping on carpet with onset of humid summer weather in the southern U.S. Run time had increased ten per cent and then it was getting stuck in some spots and issuing "wheel stuck" error messages, even though wheels were turning, I think. There was no general failure to move, which internal motor breakage would cause. It appears that water absorption changes the consistency of the textile.
Perhaps the slider will not be needed in the dryer winter weather.

A slider addition improved performance. The Botvac needs a 3/8 inch slider compared to 1/4 inch on the XV, for the larger brush head. Three layers of 1/8 inch cardboard were taped together. Oddly boxes from Digi-Key electronics are exactly the width needed. Fastens around the sides of the battery cover, with notches for the cover screws.

slider.JPG
Botvac Slider Mod 3/8 inches thick
glnc222
Robot Master
 
Posts: 4443
Joined: January 23rd, 2012, 9:19 pm
Location: North Carolina, U.S.

PreviousNext

Return to Neato Robotics

Who is online

Users browsing this forum: amaterasu, Aniseedvan, chefkeith, Dibs, lanchez, Snokia, someoneclever, vic7767 and 342 guests

cron