DIY G Seat by inflatable cushions

blekenbleu · Oct 18, 2021

Folks on xsimulator and elsewhere simulate acceleration and cornering effects using wedges.
Compared to more mechanical (and challenging) approaches,
inflatable cushions are potentially seat-agnostic.
However, initial testing with air wedges:

.. suggests issues:

effects from my largest (25x12 cm) wedge are too subtle
3 mm air lines slow inflation and deflation.

While less sturdy, these 34.5 x 24 cm inflatable cushions should enable stronger effects:

https://www.amazon.com/dp/B098DT619M

Alvaro de la Orden · Oct 18, 2021

looks interesting for flight sims!

blekenbleu · Oct 18, 2021

Alvaro de la Orden said:
flight sims

It's been far too long since I flew in a plane cockpit
and have done nothing nearer flight sim than Google Earth with my SideWinder,
but slow changes are more easily accomplished than are sudden changes.

Predistortion is one trick for improving transient response;
supposing that cushion response to a square wave driving servos looks like:

.. then briefly overdriving changes by e.g. 20% for delta t ~ 2 should help.

blekenbleu · Oct 20, 2021

Cushions are actually 32 x 21 cm:

.. and fill quite rapidly after inflation nozzle is cut off:

Now, to order some 10mm o.d. vinyl tubing and cement...

dwightyone · Oct 20, 2021

Crossed my mind something like this with air bladder bags - the uber budget version of a G Seat just for braking. I was also thinking can something be designed with air bags that simulates braking g force when you press the brake pedal in hard using high air pressure somehow - more specifcally actually physically connected to the brake pedal i.e no telemetry .

Also a cable/piston system fitted to the brake pedal and then on to the seat belt on the the rig itself - a uber budget version seat belt tensioner just for braking.

My ideas never get off the ground but that xsimulator site has some wacky and really good ideas at times that do.

NeoGrotesk · Oct 20, 2021

Wasn't there a commercial g seat that did something similar? I can't remember what it was called or much about it, other than the fact it was apparently really noisy because of the air pumps.

Insert Coin · Oct 20, 2021

Latency is often a problem: inflate / deflate through a narrow tube.

blekenbleu · Oct 20, 2021

dwightyone said:
air bags that simulates braking g force

Harness tensioning works well for braking and fairly well for cornering..

NeoGrotesk said:
a commercial g seat that did something similar

Could be; there was Realteus ForceFeel for tactile;
I hope that you can find something about such a commercial pneumatic seat cushion;
air pumps and valves do tend to be noisy.

Insert Coin said:
Latency is often a problem: inflate / deflate through a narrow tube

Latency is certainly an issue with the 3mm i.d. tubing used by air wedges,
which are after all designed for higher pressures than needed for acceleration simulation,
but was still considered worthwhile by several hobbyists at xsimulator.net.

For these low pressure Amazon cushions,
approx 8mm i.d. tubing has 7x cross-section area
and should flow about that much faster.

We shall see...

NeoGrotesk · Oct 20, 2021

blekenbleu said:
Harness tensioning works well for braking and fairly well for cornering..

Could be; there was Realteus ForceFeel for tactile;
I hope that you can find something about such a commercial pneumatic seat cushion;
air pumps and valves do tend to be noisy.

Latency is certainly an issue with the 3mm i.d. tubing used by air wedges,
which are after all designed for higher pressures than needed for acceleration simulation,
but was still considered worthwhile by several hobbyists at xsimulator.net.

For these low pressure Amazon cushions,
approx 8mm i.d. tubing has 7x cross-section area
and should flow about that much faster.

We shall see...

I think it might have been the Frex Air G-Force. I can't see much about it online though so I imagine it didn't really go anywhere....

blekenbleu · Oct 20, 2021

NeoGrotesk said:
Frex Air G-Force

I found a Facebook video for it:

https://www.facebook.com/video.php?v=1085019251531142

Puresim also offered one:

blekenbleu · Oct 22, 2021

Vinyls tubing and cement arrived today;
To minimize disappointment, tubing and inflatable pillow nozzle scraps were tested:

Meanwhile, at the other end of tubing, an easily removeable connection is wanted.
Tubing is a loose fit in bellows inflation fitting;
a 6mm i.d. o-ring was stretched over the tubing for an air-tight fit,
then a few wraps of half-width electrical tape were wound around
the last cm of tubing to discourage that o-ring from pulling back through:

This lash-up should facilitate evaluation of cushion locations
for best G simulation effects, after glue test succeeds,
then tubing-to-cushion adhesive cures.

dwightyone · Oct 22, 2021

There are inflatable kayaks - not toys for the family down the beach one and more like for going serious touring with anyhoo, they comes with a puncture kits and the glue that comes with it is designed to still hold firm way over an inflatable high pressure of 6.5psi - with someone in it, waves bashing it, sun directly on it. Other inflatable kayaks operate at even higher pressure and they need glue to match. Anyway, what I'm saying is if your glue fails there is tougher plastic/cement glue which is more than able to do the job.

Thinking of your project I wonder if high air pressure is possible as it would move air far quicker with much more impact. Don't know how to do it though or if it's even possible...

blekenbleu · Oct 22, 2021

dwightyone said:
inflatable kayaks

I am aware; they use PU-coated rip-stop nylon fabric,
which are materials for which mutually effective cements are rare.

Strongest polymer bonds result from welding,
either by heat (for thermoplastics) or solvent.
If, as I suspect, my inflatable cushion nozzles are vinyl,
then a vinyl solvent would make strong bonds to the tubing,
but bonds also need to be air-tight,
and organic solvents are generally low viscosity,
so not ideal when gaps need filling.

Loctite adhesive should have fully cured in another 12 hours;
the bond strength issue will be evaluated then.

Some folks at xsimulator tried high air pressure from compressors for air wedges
to overcome slow airflow in their 3mm i.d. tubing
That gets noisy and expensive, with pressure sensors and controllable valves.

Meanwhile, inflation/deflation response seems acceptable for these small pillows
when using 8mm i.d. tubing and small foot bellows. Preliminary results
encourage belief that foot bellows actuation will suffice.

Removing return springs from foot bellows non-destructively may be doomed,
but wanted for bellows actuation by hobby servos.

dwightyone · Oct 22, 2021

Yeah I get what you're saying about high air pressure being noisy and expensive. Really interesting project

blekenbleu · Oct 23, 2021

blekenbleu said:
Loctite adhesive should have fully cured in another 12 hours;
the bond strength issue will be evaluated then.

Loctite seems strong enough; clearly not so strong as welding,
but test pieces can only be peeled but not pulled apart.
Consequently, one vinyl tube has now been glued using it into one air cushion:

Cotton swab is an unsatisfactory applicator, leaving too many fibers.
This time tomorrow...

(Amazon US) parts costs (so far)
16.4 ft 7/16" o.d. vinyl tubing $9.99
Loctite Vinyl Adhesive $3.48
4x inflatable cushions $10.69
Foot(let) bellows $11.84

Larger foot bellows are wanted; one is on order @ $10.58
Hobby servos will eventually be wanted to actuate bellows;
RDS5160 will be tested, since one is already available.
These are $54 @ Amazon and only 180 degrees;
$32 and 270 degrees from AliExpress
< $28 @ (x4) and 270 degrees from ebay...

These will be driven by a Black Pill (STM32 version of Arduino)
already acquired for speed curving fans and stuck to my seat
using double-sided carpet tape, also already acquired for other projects
with power from a supply already deployed for ($50) fans and harness tensioning.
Total cost for both harness tensioning and G seat should be around US$300.

Checking this morning, the joint was obviously NOT air-tight;
applied more adhesive, using a toothpick

blekenbleu · Oct 24, 2021

Three additional adhesive layers later, that joint is now airtight.
Must work more adhesive between tubing and cut-out nozzle of the next cushion.

A larger Texsport bellows not only more nearly fills a cushion in a single stroke,
but also includes a more substantial and air-tight inflation hose
that needs only a short length of vinyl tubing for cushion attachment:

dwightyone · Oct 25, 2021

It's great someone from RD is showing their progress of it here - looking forward to seeing how this pans out and of course your end review of it in action.

blekenbleu · Oct 25, 2021

blekenbleu said:
a more substantial and air-tight inflation hose

Sadly, that coiled hose generates obnoxious whistles
but also flows air usefully faster; 14mm i.d. vinyl tubing is on order.
That size is relatively unusual;
both cheap and quick delivery were not possible.

Toothpick adhesive applicator was a success,
facilitating adhesive insertion between tubing and cut-out cushion nozzle vinyl

dwightyone said:
how this pans out

A key technical risk is extracting coil springs from bellows non-destructively...

If xsimulator folks were satisfied with hobby servo + air wedge implementations,
this should be a slam dunk:

much faster response
positive extraction
more volume change
- lifts a carton 5cm

blekenbleu · Oct 29, 2021

While awaiting delivery of hardware for attaching servos to bellows,
air leakage remains a concern.
If cheap low pressure air relief valves could be fabricated,
allowing bellows' inlet flappers to replenish,
then just blowing off any excess would be more robust
and easier than chasing slow leaks.

Applying side pressures for cornering force simulation
does not on my office chair have the same constraints
as does back pressure for acceleration;
space is available for servos behind side paddles...

dwightyone · Oct 30, 2021

Is your glue holding up?

Aside from kayak cement glue I've also used this stuff to install my camper van windows and gawd it has taken man/kid manhandling abuse and the windows are still installed 10 years later. Saying that, I think it is the same one from their range, anyhoo, its imported stuff and has every skull and bones warning signs that exists - and cancer warnings etc... so its a glove and mask job. Gotta be honest I can imagine it's a fiddly job to get right especially if it needs to be able to take sudden strong'ish air pressure.