Hey everyone! I have a few questions about the phoenix (I want to get one really badly, but I want to double-check a few things first)
First off, the chassis of the phoenix looks kinda small…will the 6v 2800mAh battery pack fit in the phoenix? What size are the actual battery cells? (can I use Li-Po cells or can I make my own pack ?) Can I use a 7.2v pack or will that burn something out? (I’m an RC car guy and alot of the RC speed controls are underrated… if it says 7.2v max you can usually run them to about 9.6v before you start to see problems like overheating… 11ish before they fry)
Next, I took a look at the pictures of the phoenix and I have an idea; does anyone know if it would be possible to put a bearing hinge on the back of each servo (knee and vertical hip I think they’re called…the two outermost servos on each leg) and then put another femur on the backside of each leg? (trying to bulk up the legs and make it more stable)
Lastly, is it possible to run different servos? (I know, I know…it says designed to use Hitec 645 servos but does it really need THAT MUCH torque??? My MGT (1/10 scale nitro monstertruck) only needs an 80oz-in servo to dodge parking meters … and it’s 13lb of metal moving at 60 MPH…) Is it possible to run a servo in the upper 80 oz-in range? I figured the movements would be smoother with a high speed servo instead; any thoughts?
There is some space in the body of the phoenix to fit a battery. You can fit anything up to 150 x 45 x 22mm in there. If the battery is bigger, you need to place the battery on top or modify something to get more space. I think that you are looking at LM’s 6V 2800mAh. This one is 1-7/8" x 4-3/8" x 15/16" = 48 x 111 x 24mm which makes it to wide. A single cell should be about 48 x 24mm, a little bit smaller because of the heat shrink.
The voltage that you are gone use depends on the servos. Normally the voltage is a little bit underrated to be sure that they don’t fail. I thought that I’ve read a thread about somebody who tested the max voltage. But I suggest to go with 6V and spare the lifetime. You could go with a higher voltage on your battery though. Just add a voltage regulator. 8)
I don’t know why you would want to do this because the legs are stable enough to carry his own weight (or even twice if you look at Zenta’s video’s). But I’ve seen servo’s with a ball bearing on the back of the servo. I don’t know if it was custom made or just a standard part.
I don’t know if somebody tried 475 on the phoenix. But you do not only need to look at the torque but also at the gearing. The 645 use metal gearing and because it is its only link to the body I don’t think that plastic would do the job. But if you want to solve this with the extra femur part, it would solve that problem. What you could do is get 6 x 475 for the horizontal movements (coxa) and 12 x 645 for the vertical (tibia and femur) movements. This works for sure
From the sound of it, you are really looking for the CH3-R. Because of its shorter femur it will work with all 475 servos, even carring the 2800mAh pack. I have serious doubts the Phoenix would be functional using only 475 servos.
see, that’s the thing; I kinda want the legs of the CH3 (looks more stable…not to doubt the Phoenix legs, but it just looks more robust) but with the body of the Phoenix; maybe it’s just me, but I think that a round body would be better suited to turning where as a straight body would be better at walking… the Phoenix is in the middle so to say, by having the middle legs perpindicular to the direction of travel while having the front and rear legs angled for better turning.
The Phoenix and CH3-R bots have the exact same setup. They both have the legs set on 60° iterations of a circle. Well the Phoenix is an elongated circle. It’s just aesthetics really… With the CH3-R there are always a pair of legs that are “perpendicular” to the direction of travel.
I own both the CH3R and the Phoenix. The Phoenix’s legs are quite robust. Don’t be fooled by the mounting. The aluminum pieces for the legs of the Phoenix are really thick. Even a super strong servo like the 5990TGs can even start bending it.
The challenge of the Phoenix is in the turning. CH3Rs are easy to code for the turning, but the Phoenix presents a problem since the two legs in the front and the two legs in the rear are not within a perfect circle (like the CH3R), so you can’t simply gait all six of the legs with similar code to make it turn. You’ll have to use a IK (inverse kinematic) solution to make it turn.
Luckily, there is an IK-based software generator that Lynxmotion offers, the Powerpod. Jim (head honcho) is planning on releasing a new version of Powerpod soon that will support the Phoenix.
The Phoenix really is an elegant hexapod. The CH3R IMO is a great platform. The Phoenix uses minimal body space to achieve the hexapod with standard servo size, the CH3R’s body is quite large so you can build on it (adding arms, and other goodies) provided that you have servos that is capable of the payload…
You can’t go wrong with either or, they are very good kits…
EDIT:
I did order the CH3R add-on deck kit for my Phoenix. I’m planning to use the very upper deck and cut it to shape with my Dremel. With this deck, it’s possible to add some more stuff to inner cavity of the body of the Phoenix and load the battery from the top (for easier access)…
Zenta and Xan have a very impressive Phoenix control program that’s working great! The turning is solved and works very well. PowerPod support has been delayed even longer.
I took one of the in-house phoenix home over the weekend and played with it for the first time! I had to figure out how to enable the body moves. The cool thing is it can walk with the body left in different orientations. I planned to take video but forgot the camera, I will try to do it this week. 8)
I’m glad to hear that you had some fun with the Phoenix! 8)
You sound surprised about the ease that you can combine different movements I’m glad that you like it!
Hi Xan, A little surprised, but mostly elated! I did see James walking the Phoenix with the body tilted once before, but I was surprised to see the same thing could be done with positioning the body way back over the legs, and it could walk like that too! I have never seen any 3DOF hexapod walk with these contortions!
Here’s an idea. Place a 2 axis accelerometer on the chassis and let it keep the body level while walking over uneven terrain.
You could also monitor the chassis level while operating on a flat surface. If it dips the bot may be close to an edge, and it needs to reverse it’s direction to prevent falling off the table. This may not work that well…
Of course if ground detectors where to be added to the tip of the foot lots of terrain adaptation could be accomplished. Something tells me the Phoenix projects are just getting started.
I defiantly agree! The Phoenix project are just starting! 8)
I think that adding a 2 axis accelerometer will add the possibility to walk on a slope while the body stays horizontal. But I don’t think that it will be able to walk on uneven terrain with just a accelerometer. To do that we need to add pressure sensors on the feet which will make it possible to measure if the feet really touches the ground. Adding both will give a really good “feelingâ€
I will try to do it this week. 8)