Your controls wishlist

[swalker]

Burgerman, thanks for cleaning up this thread. It really needed it. Those criticizing your efforts might not have had the chance to read those messages and see how utterly off topic they were. And those from a poster who had a total of 7 posts (3 in this thread) since he joined in April of this year. He never contributed anything, but sure criticized a lot.

Steve

[Burgerman]

Thanks. He was actually several users... Just a common troll.

I almost deleted him when I saw the IP adresses he used to register in april, and email. But wanted to see what happened. Theres a couple more silent ones on here that may be similar that I am also watching!

[steves1977uk]

And with a name like rtardxpress, it's obvious he was a troll and hate disabled people which is petty! I've come across a few of them in my time. People like them are not worthy belonging here.

Steve

[Bubbernator]

I've come across that attitude occasionally. It's rare but it does happen.
When I encounter it in public I make sure that I "accidently" run over their toes. At least now they have a really GOOD reason to hate me.

[slomobile]

They wont do it however. Because there are a set of settings, and limits, that allow each chair to pass a bunch of tests. For e.g one test is anti tip stability with a test on a slope of X degrees and full forward acceleration***. Allowing you to mess with speed inhibits, forward acceleration on some chairs, motor compensation changes these results. Those are all regulated and tested in the eu and US. braking speeds and run on when stick is released, and more is all legislated too. Its all bollox but they will lock it all down.

I will make it better if I can. We are about to go through that round of testing and I'm about to sign an NDA so I won't be able to discuss it later. If anyone has any insight on the testing process or ways to pass it without making the chair bad I'd like to hear them now. Please.

I've got a call with Curtiss - Wright applications support today. Is there anything in R-net you would like addressed? I just want to be sure I am prepared with all the information.

Sorry about criticizing the spam cleanup. I never saw the posts before they were gone. There is a general paranoia in the US these days about information being witheld. I need to watch my step and not go off half cocked.

[Burgerman]

Well you wouldnt have wanted this information!

You already know my views.

Remind tham that most users on this forum for e.g. will only ever buy, or advise others to buy R-Net equiped chairs precisely because we CAN get hold of OEM programming tools one way or another and we NEED that! And these rather a lot of us. So thats great for their sales.

[woodygb]

@slomobile
Have you obtained and read through the various ANSI/RESNA-ISO documents pertaining to safety / construction etc of mobility vehicles?

[Burgerman]

As woody says. Its one reason they force us to have unusable chairs. Both CG, programming etc.

For e.g. one test is how fast it slows as you release the stick. I HATE that. So another rule gts broken as I set back stick braking high, and set forward deceleration as low as poss after removing some walls. At high and low speeds. So it rolls on untill I CHOOSE to brake. Like my car. Much smoother and nicer to drive.

Theres another huge bunch of the daft rules that ignore and I change CG etc, within days of getting any new chair. They want to act like big brother. They dont want me doing this. And this type of stuff is your problem.

[swalker]

The rules you will have to comply with represent a very conservative view of what a wheelchair user is capable of. They assume not only limited physical capability but also limited mental capability as well.

Thus, they are built around someone that cannot readily shift their weight, but would not know that a weight shift was necessary.

Therefore, developing a wheelchair that will comply with those rules will result in a chair that will not be of much value to most of the posters on this forum.

The only way around that is to make a wheelchair that, while complying with the rules when delivered, can be easily and cost-effectively modified to violate those rules by the end user.

That is why there is so much emphasis in the responses on allowing access to OEM level "programming". But, it goes beyond that. We want to be able to change the center of gravity, have higher speeds, etc. Unfortunately, doing all that will likely result in a wheelchair that is too expensive for Medicare and other insurance providers.

Thus, I think the practical reality is that yet another group 3 wheelchair will result from their efforts. It will likely be slow, have lead acid batteries (limited range), have anti tip wheels, and weigh close to 400 pounds. Anything you can do to prevent this outcome would be of value.

Steve

[slomobile]

Have you obtained and read through the various ANSI/RESNA-ISO documents pertaining to safety / construction etc of mobility vehicles?

No! Do you have them? I've read large bits available on the internet but I'm not sure it was complete based on some references I couldn't find. The RESNA publication is $1800 if memory serves and I don't have a budget yet. ISO is freely published online if memory serves, but not required for FDA. ANSI refers to RESNA. publications if memory serves so the publicly available docs are incomplete.
I've never been through the process so I don't know exactly at what point the things you all don't like get introduced. If I see it coming, I'll be better prepared. Has anyone here been through it? Pitfalls to avoid?

@Burgerman I do want your opinion and have adopted much of it. How does the general public go about obtaining R-net programmers? If they are legally available everywhere maybe this isn't the issue I thought it was. Maybe I'm designing a solution in search of a problem. R-net is perfect? My understanding was that the OEM programmers are still restricted somehow. I just found out that Curtiss - Wright advised that there was no need for the OEM to purchase an OEM programmer unless they needed to change the basic program. I'm thinking an accountant saw that as an opportunity to save $700. Is that the problem? The cost. Oh, and you have to sign something? What do users need to sign to get an OEM programmer?

My suggestion of allowing only small programming steps per day or per boot(thanks LROBBINS) was limited to programming from the chair by user without any programmer. Full OEM programmability from a PC is still a requirement to implement, because the OEM requires it. We can't take 20 days changing a step at a time to ship a chair. Saying it that way, maybe the user cannot wait that long either. The point was to allow the average user without a programmer access to more of his chairs capability and require him to drive at least a little with each change so he doesn't foolishly make a bunch of rash changes that throw him.

So I pose the question again. R-net as it is, or identical chair with more (but still limited in some way) programming available from the seat but not R-net as it will need to be implemented from scratch (or from Lenny's template more likely), plus it will have a PC programmer available very cheap, probably just an off the shelf CAN adapter, and cable connectors that you can actually buy from digikey, and 48v.

I'm trying to gauge if the effort is worth it. Sounds like no, but I wasn't sure if the proposition was understood as I intended.
R-net must be the immediate solution because we could not complete EMC testing of a custom solution in time, I'm trying to see how strong the desire is for something else.

Actually, is ~48volts 24 lead acid cells the target you are looking for? Or is it just higher voltage generally? Is there a specific chemistry / cell combination you are after? There are many 35v rated components that are either unavailable or way more expensive in 60v variants. A 12 cell LTO solution would have a max charging voltage around 33.6v, 27.6v nominal and 18v minimum cut off. High speed charging, 10c discharge and very high cycle life. It would be compatible with most existing 24v parts but much higher current and a bit higher voltage. 1.25kg per 40Ah cell in a cylindrical shape that can fit inside frame tubing. But if 48V is really what you want, I'd like to know specifically why over 12 cell LTO.

[Burgerman]

Several ways. PGDrives will sell to people like me that they believe understand what they are doing to build my wn chairs.
And from a place in taiwan (from memory) and from a place in the US although they dont say that its OEM level. If you PM me I will give a link.
It is OEM level. And several have been sold on ebay and on this forum.

And a dealer dongle is relatively simple to get. And that can write an OEM level file, to a chair. So you just need the OEM software. Which is easy to find...

Dont go telling this to Curtiss Wright!

I am going to delete this post in a few mins.

LTO is too low energy density and far less size options. Will not allow us to fit a 200Ah lithium ion phosphate pack in the place of a set of grp 24 lead bricks.

[woodygb]

Do you have them?

NO ... They can occasionally be found via the internet but get removed as soon as the powers that be become aware of the leak ( loss of revenue ).

[Burgerman]

They were posted on here once. I got a nasty letter Aabout the range measureent ones! That was years back, but now those would all be out of date anyway. They get updated quite often it seems. Of course they onli apply to the US again.

[slomobile]

Thus, they are built around someone that cannot readily shift their weight, but would not know that a weight shift was necessary.

Sorry to revive an old thread, but this just came up in a design discussion. I'd like to gain some informal intuition about the percentage of wheelchair users that are able to shift their weight. And among those, how many would prefer a dynamic seating solution that encourages a driver to shift weight in turns and off camber (hill) situations?

Think of a seat on a post. The base of the post connects to the drive base through a non rotating Cardan(universal) joint and is held upright or allowed to pivot by 3 actuators arranged radially around the seat post. Accelerometers detect the direction of largest acceleration (down with gravity, or toward the outside of a curve) and move the actuators to keep the seatpost vertical with respect to this vector.

Even if you can only shift your weight a little, we can sense it, and amplify it with quick response actuators. The same system will buffer some of the rapid accelerations that come from curb drops and high speed bumps. But if you aren't paying attention and shifting your weight in anticipation, just along for the ride, it can feel wobbly. How wobbly is a function of the response rate of the sense/control/mechanical system. Is this something you would want?

This is a feature that I can sell as providing exercise and balance rehab opportunity and tip risk reduction for aggressive drivers. But it also requires the user to periodically refine the response of the system to suit their individual needs and preferences, for safety as well as performance. If this is determined to be a desirable feature for a wide enough market, it will open a permanent user window into the settings we hate being locked down.

[Burgerman]

That would be my worst nightmare even if it didnt weigh a ton and wasnt wobbly and flimsy like a typical poerchair with seat elevator.

Be OK if weigfhtless, didnt add any height, and was solidly rigid and could be turned off.

If you turn it on it would need to be FAST. And imensely powerful. And it will suck if it isnt. So that it could keep a seat totally flat as I drive at full speed over rough ground. Or our footpaths. Same thing. No use if it detects it and tries to catch up later. And thet means weight, lots of it. And battery power wasted.

[slomobile]

I think you have some incorrect assumptions about it. If you switch the electronics off, you still have a solid seat post buttressed (triangulated) against movement by 3 actuators. Any seat movement (beyond a small amount of unfortunate unavoidable backlash) would need to backdrive 3 actuators. I described them as being placed radially, but its also part axially, more like a teepee. I can't think a proper name for that arrangement. Those 3 actuators also provide the tilt function, so you are only adding 2. There is a way to make it also do the recline and leg elevate function for no added weight, but that is untested. Height is normal. Battery mass is distributed around perimeter frame with individually repairable cells. Structural 4680 cells if they are ever a real thing. This is necessary because we are occupying the central area where batteries usually set, but perimeter mass works better for the crash structure anyway. As far as speed and wobble goes, expect it to be as fast and stable as a Segway scooter or a racing sim seat. But backlash is a real thing I didn't want to gloss over.

Really? not useful at all? Like the lateral lean of a motorcycle and the forward/back lean of a Segway with the ability to maintain CG within track and wheelbase even on high degree slopes. Essentially tip proof. This is front of mind for me after being dumped over twice in 5 minutes on my new m3.

[foghornleghorn]

As a regular driver on slopes and rough ground I can see this being a disaster for too many reasons to explain.

Out of curiosity though - what is the duty cycle of your actuators?

[slomobile]

@foghornleghorn I am really interested in understanding your reasons for believing it would be a disaster. I'm not trying to convince anyone, just understand your perceptions. Do you have trunk control? Is limited trunk control among your concerns?

I have to admit I am a former motorcycle rider that still has leg control (just not enough spine stability to make much use of it). I may underestimate the amount of confidence I have due to the ability to kick out a leg and catch myself if I am falling. When I go over a curb in the chair, I feel extra out of control because of the way a rigidly mounted chair flings me forward, then the chair back. It happens too quickly to put my leg out or do anything about it. I thought by softening that lurch with active electronics, it would feel much better.

If I didn't have the ability to kick out my leg, maybe I would feel that my best chance is to get maximum feedback from the chair by being rigidly mounted. ? I don't know. Is that something like what you were thinking?

Actuator duty rating is trying to be 100% but heat rejection is a problem in very dynamic environments. Duty cycle? not sure how to define that, we are using FOC brushless gimbal motors. Initial model was with ball screws but one design study uses toothed belts and rotor inertia like reaction wheels. Around 150mm travel, up to 1500mm/s on low inertia setups, 600mm/s is more realistic. Nothing on a real chair yet, just a simulation and benchtop model.

[Burgerman]

I think you have some incorrect assumptions about it. If you switch the electronics off, you still have a solid seat post buttressed (triangulated) against movement by 3 actuators.

THOSE actuators and seat ost have some flex in them. And some free play. And that mens that just like the seat on my powerchairs, where there is tilt, recline theres WAY too much flexibility and its like driving around on a jelly. Compared to my non tilt/recline.lift chais. Those are very rigid and much nicer to control and abuse.

Any seat movement (beyond a small amount of unfortunate unavoidable backlash) would need to backdrive 3 actuators. I described them as being placed radially, but its also part axially, more like a teepee. I can't think a proper name for that arrangement. Those 3 actuators also provide the tilt function, so you are only adding 2. There is a way to make it also do the recline and leg elevate function for no added weight, but that is untested. Height is normal. Battery mass is distributed around perimeter frame with individually repairable cells. Structural 4680 cells if they are ever a real thing. This is necessary because we are occupying the central area where batteries usually set, but perimeter mass works better for the crash structure anyway. As far as speed and wobble goes, expect it to be as fast and stable as a Segway scooter or a racing sim seat. But backlash is a real thing I didn't want to gloss over.

And part of the problem. As described already above. And flex in metel too. Remember I can take an engine block and bend it enough with my hand to measure wth a micrometer... And actuators are not engine blocks. So some amount of freeplay and flex will manifest as a not very solid seating system...

Thats the least of its issues. To be useful its got to have the torque to move a 20 stone user, fast enough that it can keep up with running fast on an uneven track, bouncing over 1 or 2 inch curbs/ramps etc without slowing. Right now I choose to wheelie up curbs so no pitch shock reaches me. Your seating then would need actuators that are far more powerful than a wheelchair motor, to be able to do this with enough torque and speed to cope. Or its already transmitted the movement to me seat.

[slomobile]

Ok, I get it. Nothing will ever be as good as what you built.

But your inflexible chair can't do this https://www.youtube.com/watch?v=61MD3bv1qeQ&t=284s

well, neither can mine. I want that on a drivebase now.

[Burgerman]

Its nothing to do with "what I built".

Its the REASONS. The physics. The thing that determines the result. I have all kinds of chairs. I can see the problems and the physics and the differences.