FSSB resistance

Discussions about Cockpits, Joysticks & Setups
User avatar
Lawndart
Virtual Thunderbird
Posts: 9290
Joined: Tue Mar 29, 2005 8:36 am
Location: Mooresville, NC

Post by Lawndart » Thu Sep 27, 2007 11:57 pm

Gunner wrote:With our nose-down trim setting, I'm pulling a constant 8lbs to achieve neutral pitch.
...and yet you're complaining. Wimp!!! :lol: :lol: :wink:

Just kidding, that's great news though regarding the accuracy. I wouldn't expect anything less from our friendly European engineers! :)

So, Gunner, when are you going to increase your x-axis resistance to the highest setting too?

Re: realism; the real plane has 25/17 (y/x)... a 0.68 ratio. With some axis shaping I think this can be achieved through HCCP, but only at the lower force settings, starting at 10/8 for example. After this season, I'm definitely going to experiment with this!
User avatar
Gunner
Virtual Thunderbird
Posts: 1190
Joined: Thu Jun 01, 2006 10:01 pm
Location: Phoenix, AZ

Post by Gunner » Fri Sep 28, 2007 6:16 pm

Lawndart wrote: ...and yet you're complaining. Wimp!!! :lol: :lol: :wink:
I have NOT complained...only whined a little :lol: :wink:
And once you've gone heavy, and gotten used to it, you'll never go back :wink:
Yes, I will play around more with roll settings AFTER this season...I do have my curves set so it feels right even in the heavy pulls.
User avatar
Lawndart
Virtual Thunderbird
Posts: 9290
Joined: Tue Mar 29, 2005 8:36 am
Location: Mooresville, NC

Post by Lawndart » Fri Sep 28, 2007 10:02 pm

It's amazing what you can do with a little software tweaking, isn't it! Making that .68 ratio work... 8)
User avatar
Burner
Virtual Thunderbird Alumnus
Posts: 1420
Joined: Tue Mar 29, 2005 5:08 pm
Location: New Orleans, LA

Post by Burner » Fri Sep 28, 2007 11:36 pm

Gunner you might try setting your upper and lower deadzone of your y-axis to 10%-20% in the HCCP. This will give you a closer 0.68 ratio of x/y axis. If its a better setup my experience is that you acclimate to it quickly b/c in the end a better setup means it makes the plane easier to control.

I plan to set my x-axis to 8lb and do the same deadzone trick to get 7lb x-axis force. Making the 7lb/10lb a 0.7~0.68 ratio.

This is the entire method for adjusting the FCC control stick as it does not have variable resistors like the FSSB.
Image
User avatar
Gunner
Virtual Thunderbird
Posts: 1190
Joined: Thu Jun 01, 2006 10:01 pm
Location: Phoenix, AZ

Post by Gunner » Sat Sep 29, 2007 9:56 am

At this point, I would rather increase force in x axis than lower the y. I really am flying better and I'm kinda into realism :D :D :wink:.
User avatar
Lawndart
Virtual Thunderbird
Posts: 9290
Joined: Tue Mar 29, 2005 8:36 am
Location: Mooresville, NC

Post by Lawndart » Sat Sep 29, 2007 11:25 am

I think what Burner is trying to say is there are more ways than one to simulate realism, in this case by altering the "feel" in order to acheive that magical 0.68 ratio - even if it means somewhat lesser force compared to the real airplane. Or you could go the way you're going about it, setting the hardware settings as close as possible (at least in the y-axis) and then compromising on the x-axis as needed. You should be able to shape the x-axis down from the highest setting to about 14 lbs (0.68 ) with a some simple math based on the "available range of shift" (i.e. a different percentage number than the ratio you're seeking) in HCCP. Having a solid pit and mounted force sensor obviously makes a big difference as most people who have tried 21 lbs on their desktops would know!!! :D

I'm looking forward to trying out your setup next week Gunner! Are you using our default Kitty profile or do you have your own by now? (Just curious so I don't mess up my slow roll with an audience... :lol: ).
User avatar
Gunner
Virtual Thunderbird
Posts: 1190
Joined: Thu Jun 01, 2006 10:01 pm
Location: Phoenix, AZ

Post by Gunner » Sat Sep 29, 2007 2:45 pm

LD-
I am so looking forward to having you fly my pit - it's almost scary to think how good you'll fly with an accurate setup :shock: .
The only changes I've made to our Foxy setup are a couple views, and the trim setting is -31 instead of -33. I had to take that to -24 when I first increased the resistance, (wimp) and have been steadily increasing as I've gotten stronger 8) . It will probably be up to the -33 next week.

My curve settings are as follows:

Image

Image

Image

As you can see, I'm compensating for the x/y ratio with the LOMAC curves - with this setting, it feels about right (I'm using the same scientific method as you, LD, the billions of sensors... yada yada) :wink:
User avatar
Burner
Virtual Thunderbird Alumnus
Posts: 1420
Joined: Tue Mar 29, 2005 5:08 pm
Location: New Orleans, LA

Post by Burner » Sun Sep 30, 2007 1:45 am

Here's the results of my latest stick setup trials as we discussed in Ven.

Y-Axis = 21lb FSSB Resistor Setting
X-Axis = 10lb FSSB Resistor Setting

Here are my CCP settings, that by my calcs give me 9.6lb X-axis, and 15.6lb Y-axis for a 0.62 ratio. I then set my LOMAC curves to straight (50% Shift, 0 Deadzone, 0 Curve) for the X-axis and Rudder Axis. I let LOMAC handle the Y-axis curve and deadzone as its uniform for the whole team.

X-Axis - Upper and Lower Deadzone set to 8% decreases total required input strength by 4% (see calc below). Notice also the 6 deadzone and -6 curve. In LOMAC there is no deadzone or curve for this axis
Image

Y-Axis - No Deadzone or Curve, just a 29% reduction in required pull. In LOMAC the curve and deadzone are the usual values for this axis.
Image
And the Y-Axis LOMAC Settings:
Image

Rudders - Just to round out the post in case anyone is interested in what I use. No curve or deadzone for this axis in LOMAC.
Image

Here's how the percentage math works to determine pull strength.
Actual% = CCP% / 2
Pull Strength (lb) = (1 - Actual%) * FSSB Resistor Setting (lb).

Ex.
X-Axis = 8%
8% / 2 = 4%
(1-0.04) * 10 = 9.6lb

Y-Axis = 58%
58% / 2 = 29%
(1-0.29) * 22 = 15.6lb

9.6 / 15.6 = 0.62

PS - I like to favor left roll over right roll when I do the manual cougar calibration. So I take the stick all the way to the left and pull the trigger then try to match how that left input felt on the right input, though its almost always 80-90% of full deflection. This gives a nice symetrical roll response and a smooth uninterupted y-axis pull. [/img]
Last edited by Burner on Sun Sep 30, 2007 2:53 pm, edited 1 time in total.
Image
User avatar
Lawndart
Virtual Thunderbird
Posts: 9290
Joined: Tue Mar 29, 2005 8:36 am
Location: Mooresville, NC

Post by Lawndart » Sun Sep 30, 2007 6:58 am

Still not over 15 lbs Burner! Psst... 21 lbs! ;) Dividing by 2 comes from the assumption that the deadband range is 50% or half of each side of the axis, correct?. Good info, thanks for posting it up!

Y-Axis = 58%
58% / 2 = 29%
(1-0.29) * 21 = 14.9 lbs
User avatar
Burner
Virtual Thunderbird Alumnus
Posts: 1420
Joined: Tue Mar 29, 2005 5:08 pm
Location: New Orleans, LA

Post by Burner » Sun Sep 30, 2007 2:21 pm

Gunner's results showed 22lbs at middle finger & ring finger contact point. The resistor is set to a labelled 21lb setting but that setting produces 22-19lb depending on hand position. I set my hand very low on the stick which means the least possible torque of the possible hand positions, meaning maximum possible stick force => 22lb. That's why I use that number, but if you have you middle finger right underneath the trigger use 19lb for your calc. To be more precise you have to account for any deviation from 325 on your cougar calibration as well. Any deviation will produce a stronger or weaker pull strength from the 19-22lb range. It also could have affected Gunner's readings, though I remember him saying that he keeps things locked at 325/325 with his new mounting so that's probably not the case.

Once you start testing I think you'll see that feel is the driving factor and actual pull strength is less important than the x/y ratio- which is unitless.

To expond further on the x/y ratio and Gunner's measurements-
So long as Gunner measured pull strength at the same point on the stick for x and y axis his accuracy of pull strength is unimportant (ie- closeness to the real values). So long as his precision is sufficiently good to get a reliable +/-0.25lb reading which is probably the case his x/y ratio will be reliable within +/-0.04. So the deciding factor behind the accuracy of the x/y ratio is the precision of Gunner's spring scale. I calc'd 0.62 ratio for my settings but with the possible variation I could be 0.62+0.04 = 0.66 or 0.58 on the low end.

So as you can see there are a whole host of ways our calculations can have unaccounted for variation, from hand position, to Gunner's measurements, to cougar calibration point, to variations in resistors from cougar to cougar, and the list continues onto the amount of area your fingers occupy on the stick or if you're wearing gloves.

Setting up a pull strength that works for you is going to be akin to setting up a race car, a little math to get in the ballpark and alot of testing and feel. I found a runway takeoff and split-s with some trim-on flying at 400knts gave me a nice range of pulls and rolls at various airspeeds for quick testing. Once I was happy there I fine tuned with phantom flying. I must have gone into phantom testing and scraped my settings about 4-5 times, going back to the drawing board b/f I was satisfied.
Image
User avatar
Burner
Virtual Thunderbird Alumnus
Posts: 1420
Joined: Tue Mar 29, 2005 5:08 pm
Location: New Orleans, LA

Post by Burner » Sun Sep 30, 2007 2:39 pm

PS- the sure fire test would be for Gunner to set his CCP to 100% Upper and Lower Deadband and retest with his spring scale. Assuming a linear response of the FSSB (which is one of their selling points on their website) you could calc how much 100% in CCP represents in real life. If its indead a 1/2 ratio he should get 11lb using the same test point as the 22lb response.

Again- I don't feel that getting an accurate number for pull strength is important to setup.
Image
User avatar
Lawndart
Virtual Thunderbird
Posts: 9290
Joined: Tue Mar 29, 2005 8:36 am
Location: Mooresville, NC

Post by Lawndart » Sun Sep 30, 2007 3:00 pm

I was just curious as to why you called it the 21 lbs setting early on in your post, but then used 22 lbs for your calculations. I'm sure Realsimulator showed the FSSB to be 21 lbs at a specific point of measure during design testing, much like Gunner tested his to be 22 lbs at home - I was just curious why you used 22 lbs as your basis to compute your force?

Since the number is only an approximation for the many myriad of reasons you already brought up, I just find it easier to stick to the manufacturer's rated force setting when talking about it and comparing settings, since all stem from the switch position labelled 21 lbs.
User avatar
Gunner
Virtual Thunderbird
Posts: 1190
Joined: Thu Jun 01, 2006 10:01 pm
Location: Phoenix, AZ

Post by Gunner » Sun Sep 30, 2007 3:59 pm

Burner wrote:PS- the sure fire test would be for Gunner to set his CCP to 100% Upper and Lower Deadband and retest with his spring scale. Assuming a linear response of the FSSB (which is one of their selling points on their website) you could calc how much 100% in CCP represents in real life. If its indead a 1/2 ratio he should get 11lb using the same test point as the 22lb response.
And the results are:
0% (it was set at 5% default) 22.5lbs
100% 13lbs
This is with the strap halfway betwixt the trigger and pinky switch. (It pretty much has to be there 'cause of the taper of thumb switch)
User avatar
Burner
Virtual Thunderbird Alumnus
Posts: 1420
Joined: Tue Mar 29, 2005 5:08 pm
Location: New Orleans, LA

Post by Burner » Sun Sep 30, 2007 4:09 pm

Lawndart wrote:I was just curious as to why you called it the 21 lbs setting early on in your post, but then used 22 lbs for your calculations. I'm sure Realsimulator showed the FSSB to be 21 lbs at a specific point of measure during design testing, much like Gunner tested his to be 22 lbs at home - I was just curious why you used 22 lbs as your basis to compute your force?

Since the number is only an approximation for the many myriad of reasons you already brought up, I just find it easier to stick to the manufacturer's rated force setting when talking about it and comparing settings, since all stem from the switch position labelled 21 lbs.
We are using the resistor label setting in a sense- we're using the 10lb x-axis as being spot on at the point of measure- so that's the chosen base line. Then measuring the corresponding y-axis force at the same point yields whatever it yields and that's our ratio. We could just as easily have placed the base line where the spring scale reads 21lb on the y-axis and measure the corresponding x-axis force to come up with ~9.5lb.

Gunner remeasured today realizing he had a 5% upper and lower deadzone in CCP during original measurements and got 22.5lb. He also tried a 40% upper and lower deadzone and got 18.0lb which corresponds to the estimated 1/2 ratio between CCP% and Actual%. Further testing showed 13lb at 100% which gives ~0.57 ratio. I would chalk the difference up to any non-linearity in the strain gauge response of the FSSB. If we use 0.57 to predict the 40% setting it yields 17.3lb which is not what Gunner recorded in earlier tests (18lb).

I'm all for refering to it as the 21lb and 10lb setting, just be sure to remember the measured values when doing any calculations.
Image
User avatar
Burner
Virtual Thunderbird Alumnus
Posts: 1420
Joined: Tue Mar 29, 2005 5:08 pm
Location: New Orleans, LA

Post by Burner » Sun Sep 30, 2007 9:02 pm

Here are my latest preferred settings. The new x-axis roll curve is definitely better than the previous one, you can really lock in with the Boss and make fine corrections while still having crisp roll authority at higher deflections. The Y-axis is lighter than before and that may, or may not, be increased to a harder pull in the future. We'll see how my arm feels after an acclimation period of a few days :oops: Though it is a step closer to the 0.68 ratio than the previous setup.

The Numbers:
X-Axis - 9.6lb (the new roll curve makes the difference)
Y-Axis - 14.6lb
Ratio - 0.66

Image
Image
Image
Post Reply