Toddler Mini Servo Problems

Pezi_Pink!

Hello all,

I have recently received the Toddler robot kit and firstly I would like to say I think it's a great product and I'm sure I will have lots of fun with it.

However, I have a slight problem with one of the mini servos that are included with the package. When I attempt to centre the servo (using a 750 pulse at 20 intervals as suggested) one of the servos centres itself fine, but the other one just goes crazy, it moves to a certain point and then continually seems to be attempting to move back and forth a few degrees and it will not calm down no matter what I try. Sending it different pulses does make it move into different positions, although the behaviour is still very erratic. It would seem strange it's only one of the servos which would indicate I'm not doing anything incorrectly. I have switched the servo connections on the project board to eliminate that as a possible cause, and I have also smashed it reapteadly over its proverbial head with a great big hammer, although this doesnt seem to have helped a great deal either.

In all seriousness though, I'm pretty sure I'm not missing sometrhing - I know some servos take all sorts of different pulses even if they are the same model, but from what I have read the Parallax ones should be very stable and all operate identically.

Any help would be most welome. Obviously posting three lines of code would be pretty pointless so I'll leave it to your imaginations [noparse]:D[/noparse]

Thanks, and beware the pink squirrel!

LoopyByteloose

PLEASE!
Post your code that it is programmed with.
That will help determine if the hardware is faulty or you just have a software bug..

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)

······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan

Pezi_Pink!

It's identical to that in the manuals, I use the same code for both servos but only one funcitons correctly, hence my reluctance in showing the code. It's literally ;

Center:
Pulsout TiltServo, 750
Pulsout StrideServo, 750
Pause 25
goto Center


I'd say I certainly have a hardware problem, as swapping the servo connections round still results in the same servo doing strange things whils the other one works fine

Thanks for looking at it for me.

Tom Walker

IF the facts are as you have given them, then this sounds suspiciously like a hardware problem in the servo electronics. In a typical servo, the electronics try to match its "internal conditions" to the pulse being sent. With the swapping you have done, the code you claim to be using and the behavior you describe, it sounds like the internals are not quite in spec [noparse]:)[/noparse]

I recommend a nice cup of Earl Grey, or maybe swapping the servo <g>...Parallax is great about their product satisfaction policy....

Give 'em a call...

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Truly Understand the Fundamentals and the Path will be so much easier...

Pezi_Pink!

Thanks for you input Tom [noparse]:)[/noparse] I just wanted to clarify that this was indeed not standard behaviour for the Parallax servos. I'm actually in the UK and as such received the kit from a localish distributor, I gave them a call this morning to see if they have any more (as I don't) which they are currently investigating. It would seem most unlikely to have a servo with a fault like this, but the facts are as stated; I wouldn't come on this board and start telling everyone loads of rubbish as that would just be pointless.

Reading your post it would appear that you don't seem to be convinced with the content of my original post, especially from "the code you claim to be using" and "IF the facts are as you have given them.." This would lead me to believe that you must get a lot of posts here which are just a bunch of slightly bent truth? Hence being immediately sceptical about posts from a new user.

However I would just like to say that I'm far from clueless on the front of programming and electronics. I'm a lead software engineer specialising in fraud prevention and investigation techniques for a major global bank. Electronics has been a hobby of mine for many years, although I'm just getting into robotics, hence, I'm a little unsure with the servo specifics as I haven't really tinkered with this sort of mechanical device before now.

Once I have the Toddler up and running my first project will be to interface the Toddler with my XGS (XgameStation) using the Parallax RF modules. I intend to mount the Ping))) on a bracket allowing it to rotate 180 degrees or so. I would then allow it to sweep say 60 degrees in front and return an array of distances per degree to the XGS. The idea then is to get the XGS to translate the distances into "Raycaster" type of video kernel; This would generate the illusion of three dimensions onto the XGS which would also be scaled with colour, so further distances would generate smaller and darker lines. This should then be able to show me basically what the toddler is seeing Wolfenstein style. I was also considering using the infrared as well to detect close objects and display those as blocks of red on the TV based upon which IR zone they fall into. I am also going to attempt to develop an override mode on the XGS whereby you can press the fire button on a joystick to switch to manual mode, and then be able to walk the toddler around with the joystick remotely whilst viewing a basic three dimensional representation of what the toddler is seeing on the TV. This part could be quite challenging though I would imagine. What do you think?

Anyhow I can appreciate it must be annoying If you get lots of silly questions on this forum, however it might be nice if the community wasn't quite as fast to question the validity of a new user's question. I actually very rarely ask for help on a forum, however in this case it makes sense as you guys have had lots of experience with these particular Parallax products, and you might have said "Yeah, your problem is XYZ I have seen it a million times!" On the other hand, if I just got the wrong end of the stick, I apologise in advance [noparse]:)[/noparse]

Anyway, thanks for your help and I look forward to contributing more on these forums. I am really looking forward to getting my project underway and I have a load more project ideas stewing in my head as well, one of which will be attempting to make a servo coprocessor with ramping using a SX28, I imagine this should be possible.. ?

[noparse]:)[/noparse]

Bruce Bates

Pezi_Pink! -

A Pink Squirrel is an alcoholic cocktail on this side of the pond, what's its significance over on the rational side of the Atlantic?

Although I should let Tom speak for himself, as he's adequately capable of doing, I suspect he was was questioning the simpilcity of the program vs. the oddity of the hardware response, and not your abilities, or your veracity. Indeed there are some folks that are new to the hobby, and tend to tell you what they THINK you want to know or what THEY deem is significant and often overlook either the obvious, or offer up yesterday's program, rather than the one about which they're speaking <sigh>. So it is with Help Forums in general I guess, but please don't take it to heart. We really are a freindly group and try to be as helpful as we can.

One thing I'm not sure we've established here is whether this odd "hunting" behavior is occuring when the mini-servo is installed, or whether it's just sitting on the bench awaiting installation?

I would also have to put my money on a faulty servo, and if I had to place a bet, I'd certainly suspect a "dirty" connection or defective contact on the wiper mechanism of the internal feedback potentiometer. I suppose the other things that MIGHT cause something like this is/are one or more of the following, all of which are in the faulty servo category:

a loose or intermittant signal wire, internally

an out of tolerence compnent internally (resistor or capacitor)

a bad servo driver IC

IF it IS nothing but a dirty wiper contact, it's POSSIBLE that running the servo end-stop to end-stop a few dozen times MIGHT clean it up. No promises, but it might get you back in business for now, until you can get a new servo.

Regardless, you are certainly entitled to a NEW servo if any of the above conditions are suspected by the Parallax Technical Support folks. Parallax is VERY GOOD about replacing any of the products which come from them, which are deemed to be faulty, as there are generally so few occurances of defective products. That, in and of itself, is probably another reason for Tom's surprise at the symptoms.

Regards,

Bruce Bates

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
<!--StartFragment -->

Chris Savage

Before you do anything else, try changing your PAUSE value from 25 down to 20 to see what happens.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
csavage@parallax.com

Pezi_Pink!

Thanks Bruce, Tom's response probably just came across to me slightly different to how it was intended. I am grateful for his reply. In answer to your question, the servo performs this way both when mounted and not. The way it is performing it almost seems to be rotating in an attempt to centre, then its getting to its maximum angle and it seems to want to go further than it can, as if the centre is off its maximum possible alignment, so it sort of jerks back and forth at the end of its tether. Strange behaviour indeed. I'll give your advice a try and see if I can get it running end to end for a while, although this might prove interesting as it seems to think the centre is past the end of the maximum rotation, so getting it to run back and forth might be difficult; I'll try and plug some different values into it when I get home and see if I can analyse what it's actually doing a bit more, and I'll let you know. Thank you very much for the suggestion.

As for the pink squirrel, it's far more menacing than a cocktail.

Pezi is the pink squirrel from the Abyss. He's mean, violent, has foot long teeth and a passion for Jelly Babies. He also has a M16-A which is a version of the standard M16 rifle, modified for firing compressed acorns at optimal accuracy at a great range.

Generally he's meant to be on a quest to retrieve a holy relic, the Acorn of Plenty, for his family back in the Abyss. A lot of the time he gets side-tracked though and just jumps around in bushes and stuff.

So, next time you are walking your dog, beware!!!

Pezi_Pink!

Hi Chris,

Thanks for the suggestion, however I think I tried this and a few other values last night; I'll give it another go when I get in. However, wouldn't both the servos operate in the same manner ? What's the theory behind decreasing the pause get one to work, when they are identical?

Thanks!

Post Edited (Pezi_Pink!) : 4/5/2006 3:30:07 PM GMT

Chris Savage

Servos require refreshing about every 20 ms or so.· Many can handle up to 50 ms delay between pulses, but the erratic behavior you describe is similar to servos that are also refreshed too often, or not enough.· In the case of both servos doing this, every servo is slightly different and so I was just being complete.· It could in fact be damaged, but that is so rare I wanted to make sure we covered all other bases.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
csavage@parallax.com

Tom Walker

Pezi,
I truly apologize if my tone seemed anything less than pleasant. As Bruce stated (thanks Bruce), there are many newcomers posting on this forum, and it can result in a long "diagnostic" thread because a lack of experience leads the poster to either leave out a small detail (What do you mean, I have to connect the grounds together? I'm using separate power supplies...), or assume that their "small change" to the published demo program "couldn't possibly cause this, so I didn't think I needed to mention it" (I changed the I2C code to use pins 4 and 5, but that's all...). This was the only reason I stated things the way I did.

Please don't let my lack of communication skills prevent you from continuing to use and, I'm sure, contribute to this community.

As for your servo problem, Bruce pretty much covered what I intended to point towards...

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Truly Understand the Fundamentals and the Path will be so much easier...

Pezi_Pink!

Tom,

No problem, sometimes it's difficult to tell peoples "tone" over the internet as you know.



On the servo front, after a fair amount of playing around, the servo seems to settle down a lot when refreshed at a swift 10ms. However, it still isn't really stable like the other one (but at least it goes where I want it), and still shakes around a couple of notches on the spline back and forth. I can't see this behaviour doing the poor thing any good...

The toddler walks now, albeit a bit erratic with a madly vibrating stride servo [noparse]:)[/noparse] I think with some tewaking I can compensate for it, might be interesting.

I haven't heard back from the distributor yet though, but if they have any in stock they are going to send me a new one so it's all good.

Thanks !

Chris Savage

If you're still having problems with the servo, contact Tech Support directly and we can arrange to get that taken care of.· It seems we have gone through all the troubleshooting and it doesn't seem to be much better.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
csavage@parallax.com