It is hecking monday, 17:30, 20/02/2023
So...
Sigh...
There is a problem.
On Project Log 16, I somehow got to the conclusion that each muscle would only need 0.0753 liters.
Which made my sigh in relief and made my really happy with the result...
... And since then I never got out of my mind that something wasn't right, it couldn't be that easy to make a mech.
And guess what? Unfortunately, I was right.
And yes, one of the reasons I didn't do much these last months was exactly because I was trying to figure out what I did wrong. Redoing calculations and so on.
The difference in volume was based on the idea that the muscle didn't change its diameter too much, and only shortened in length.
So...
As demonstrated in the weird looking comparisson I've made on paint bellow:
Each different color is comparing sizes.
So, red line is the initial length of the first McKibben, in orange, the final length.
The purple line is the difference between the final and initial length, the pink is just separating the purple lines.
In resume, the length of a McKibben muscle generally contracts to just 20%, and since it took 4 purple lines to complete the initial length, which it is exactly 25% of the initial length (100 divided by 4 = 25).
(I did the same thing with the second mckibben one, it gave the same result).
Now, if we take the same approach for the diameter, the light green line shows the initial diameter, and the dark green line shows the final diameter.
Now, the orange line shows the difference between these two diameters, and since it takes 2.5 orange lines to complete the final diameter, I can assume it is 40% of the initial diameter (100 divided by 2.5 = 40).
Of course, I tested it with other McKibben muscles, and generaly, it goes around 30-40% in difference.
Where I'm going with it?
Well, if we take the differente in volume again, just like I did on Project Log 16, it will give a completly difference result.
For example, if I take the exactly same equation I used on that Project Log:
- Initial length = 30cm
Final length = 24cm
Difference in Length = 6cm - Initial Diameter = 14cm
Final Diameter = 19.6cm for 30% and 18.2 for 40%
Difference in Diameter = 9.8cm for 30% and 8.4cm for 40% - Volume of uncontracted Mckibben = 4618.14cm³
Volume of contracted McKibben = 7241.25cm³ (or 1330.02)
Difference in volume = 2623.11 - 0.9g/ml x 2623.11cm³ = 2360.799
- This gives 2.3607 kg
Which is 2.68 LITERS.
*Not* 0.07 liters.
I also used this online pipe volume calculator and I got the same result.
What this means?
Well...
It means I screwed myself.
So, let's now calculate how much fluid flow this would need.
So, I need 2.68 LITERS per McKibben muscle.
I have 72 muscles in the mech body.
This means that I would need 192.96 liters in total to fully actuate all muscles.
The unactuated McKibben muscles have 265.68 liters (332.496 kg) already.
This actuation fluid would add 416.88 kg.
Giving 682.56kg of weight in total.
Now, the pump would need to give 578.88 liters per second.
And I had all this problem for just 900 liters per MINUTE.
I would need a pump that puts out...
34732,8 liters per minute.
That's like... 38.5 times more power than 900 liters of the would pump.
So I would need 38.5 times more energy, which would make the batteries 38.5 times bigger. Which would increase costs by 38.5 times.
I needed 750 watts for 900 liters per minute, so...
I would need 28875 watts.
This is a 30000 watt (30kw) Diesel generator.
This is the POWER SOURCE.
Of course, I could simply attach the diesel engine to the hydraulic pump.
This is a 30hp diesel engine, which is a little more than 28kw watts (1 horsepower = 750 watts).
Also, the 200 liter oil tank will be more or less the same size.
Still...
I feel stupid right now.
I don't doubt the mech would be able to keep going with 600kg of extraweight, there are 12 5 ton muscles on the thighs alone.
However... I feel like that it won't be as easy/efficient as before.
Argh, I don't know what to think...
I just feel stupid...
I was thinking on maybe just... Using more pressure and less fluid, but I feel like I will need to use even more energy for that.
In fact, I don't even know how to calculate that...
Anyway, that's the project log.
Discussions
Become a Hackaday.io Member
Create an account to leave a comment. Already have an account? Log In.