Orders can still say Pending even after you have received it. Don’t worry because you will get your stuff as soon as Lynxmotion can get it shipped which is typically 3 - 5 days for US shipments. Well, that’s how long it usually takes to get stuff I order, but your milage may vary. 
I am not connected with Lynxmotion, except as a very satisfied customer and continuing customer. 
8-Dale
After building a handfull of humanoid robots, you will want to have every servo in the robot to have the same torque rating. Bipeds fall down. There is nothing you can do to prevent that from happening. After a while, you are going to want to program the robot to get on its feet on its own. If your arm servos are too weak, it won’t be able to get up. If you compare the price difference in using cheaper servos in the arms to the over all costs of the robot, to using good servos in the arms, the overall cost difference is small. Much smaller than having to buy new servos later to replace them.
The only exception to this would be the head servo.
Pete
Well, I’m going to buy a couple dozen 5955’s to replace the servos i got. I was about to make that purchase until last weekend i noticed a flat tire on my car and the machanic told me all four tires had bad sidewall damage. Time for new tires($400)…not to mention auto insurance ($700). So i guess I’ll have to wait a few weeks till i get a couple more paychecks in the bank.
~Mike
I don’t agree here. I don’t think it’s necessary to have all servos have the same torque ratings. Servos should be chosen for each joint according to how much of a load is going to be handled. For instance, I wouldn’t use an HS-5955TG unless I really needed one at a particular joint, because these are very expensive servos.
I’d use HS-5645MG servos everywhere except in cases of extreme loads, such as the ankles and hip vertical joints where these have to support the the entire robot, for instance. I’d use an HS-5955TG at the ankle and hip vertical for sure.
Neither of the biped projects I am working on now have arms, but I am allowing connections if I want to attach arms later.
I still say it depends on how much of a load the servo is going to have to handle. If the servos are placed intelligently, there should not be any problems. Considering the HS-5645MG at $54.95 and the HS-5955TG at $114.95, I’d say that is quite a difference in price. There is no way I am using an HS-5955TG unless it is absolutely necessary because of extreme load.
8-Dale
My point is when choosing servos, try to imagine everything you might want the robot to do. Like walk, bend over, squat up and down, stand up. Then for the more challenging tasks, walking up and down stairs, or holding one leg out front and squatting.
These different activities have different load demands on the servos. Walking on a flat smooth surface or standing still are the easiest things to have the robot do. Squatting has a lot more demands on the thigh, calf, and foot servos. Ankle servos are not used as much since both feet are on the ground.
Then when you get to having the robot walk up some stairs. Much more demands on all of the leg servos for one leg. Not only is the one leg trying to move the robot forward, and balance, it has to dynamically lift the entire weight of the robot.
Then lets look at costs the cost risks. If your robot has 10 servos for the legs, and you want to use the HS-5955TG for the hips and ankles (4 total) and the HS-5645MG servos for the other 6 servos you would spend $459.80 for the HS-5955TG servos and $329.70 for the HS-5645MG servos. Total $789.50 Then after you build the robot, get it up and running, and then start having it do some of the more difficult activities and learn that the HS-5645MG servos don’t have the power to do all the things you want the robot to do, then you end up getting 6 more HS-5955TG servos and spend another $689.70. In the end, you spent $329.70 more than you needed to spend.
Sometimes you don’t have the budget to spend all this at one time. so you go with the two servo size route and when you have the extra funds, get the better servos.
Believe me, the more power you have up front, the better off you have because you limit what you can do to compensate for lower powered servos.
This is an expensive hobby.
Pete
I have been considering what activities my robots would be doing as I choose servos for them. However, I don’t need to consider some activities right away. I don’t necessarily want the robots to be able to do every type of action. The legs on my bipeds are hybrid designs, and allow some things to be done easier and with less power required than might be needed with a standard design. The BiPod has 4DOF hybrid legs and Mega Scout has 8DOF hybrid legs.
Even if I did go this route and end up having to upgrade one or more joints at some point, I wouldn’t look at it like I had spent more than I needed to. There are always other projects to use servos on. I already have three custom robot designs, one of which I am building right now - The BiPod. I won’t be using standard servos if my experimentation with Open Servo project works out as I hope.
I could buy maybe two HS-5645MG servos per month (possibly three), or one HS-5955TG per month. However, for the first two servos I get for The BiPod, I am getting two HS-645MG servos and having them made into Open Servos. This should get me a servo at least as good as an HS-5645MG as far as the torque rating, and it will be much more flexible and accessible. The HS-645MG has the same components as the HS-5645MG, except for the electronics.
Yes, it is an expensive hobby. However, it is possible to do quite a lot on a limited budget. I have learned so much about robots and robotics just tinkering with WALTER, it is really amazing. I’ve learned how to work with several different kinds of sensors, experimented with different configurations of sensors, including two different designs of pan/tilt turrets, etc. WALTER is an excellent experimental platform. I am just starting into the custom designing with The BiPod, and I believe I have figured out how it can walk in both modes now. It is going to be interesting, to be sure.
8-Dale
I’ve decided to wait a little longer. I need to learn this software and get something simple walking. I’m also planning to get a PIC controller but i think I’ll wait till the “new version” comes out that is in the works.
Hopefully I will get the raise I was promised next month and will have more money coming in.
Now…where can i find some more of this Spare Timeâ„¢ stuff?
~Mike
What “new version” PIC controller are you referring to?
8-Dale
Could be the Explorer 16 system
microchip.com/stellent/idcpl … e=en024858
I’m holding back from it at the moment to see how others get on with it.
Sorry, I meant the future Bot Board being discussed the the Bot Board section.
I may even build my own, I have experience with Atmel chips.
An update on this project. I had put this away for a few months while I saved enough for new servos. UPS will be delivering 22 5990tg servos tomorrow. Hopefully before I go to work, but I doubt it.
I can’t wait to finally get started on this project again!
~Mike
Got my Servos today. I don’t want to enlarge the holes in my brackets just yet so I’m going to change the hinge backs for the flat backs they came with. Now I have to go get some M2 size screws for the metal servo horns.
~Mike
pictures please?
I just got off work and am very eager to start building on my bot. I’ll post pictures when i have something together…right now, it’s just a pile of servos, screws, brackets, bearings, etc. 
Here is my bot V1.0
Motors are in place, brackets are all connected.
The only problem i am having right now is the wires for some of the lower servos are too short to allow my bot to fully stand upright.
No program written yet.
I’m trying out Kroony’s recent ankle idea.
http://img484.imageshack.us/img484/3774/sh101702sj6.jpg
That looks mega cool! Is that entirely custome built or did you build upon the LM scout? Thats fantastic 8)
Looks like the 209 chasis, with modified legs?
Kroony posted a thread about lowering the ankle up/down hinge. I borrowed that idea so the dual servo hinge is on the upper leg while the short and long C-brackets are on the lower leg. Other than that, it’s similar to the picture on the “what can I build” page with that chassis.
The only thing I need now is some way to hold it up. I was sitting here last night holding it up with one hand, setting the servo limits. My arm is still kinda sore this morning from holding hat much weight for so long. lol
~Mike
You can go to Home Depot or Lowes and buy PVC pipe and fittings and make a suspension rig to “hang” the bot via a teather.
OMG!!! 22 5990TGs? I bet you are having mega fun with those. Those are probably the best servos you can get right now in the standard size!
Your scout is beautiful and with gobs of torque, it should be no problem for you to get that thing around and balance on one leg… How are you handling power? I would probably invest in some 5-cell sub-c packs, perhaps build your own… NiMH has hit 4800+mAH sub c so probably would want to invest in some of those.
Many RC cars use LiPos if you really want to shave off some weight and still have high capacity. Careful with those though, most are assuming an application for RC cars and will sell you “7.2V” packs but when charged, it will definitely go beyond, even with a high load (RC 540-sized motors draws a crapload of current when throttled fully).
I used to give my hs645 servos 6-cell sub-c (which is equivalent to giving steroids to humans)
Keep us updated with your project… I have an idea for a humanoid biped made with SES parts that is comparable, if not stronger and more reliable, then the Kondo’s out there… I just need to get my solid works installed and model it…
I will post it once I’m finished, it’ll probably take me a couple of weeks since I’m running 2003 and not sure if the user-submitted models, which are in sw 2005, will work in it. If not, I’d have to model it…
Again, awesome scout! Keep us posted with pics and videos (via youtube?)…
Oh yeah, what happened with your HS475HBs? Did you return them or do you still have them? Why not buy some brackets and build a mini-hexapod, or better yet, the phoenix hexapod when it comes out? Those two hexapods are quite light weights. If you slap an IR transmitter and receiver on each one with a bot board+ssc-32 combo on each one, you can have a bunch of walkers, bipeds and multi-legs, “network.” Possibly updating each other with sensor data? Just a thought…
