Help my penguin!

romshark

I recieved my penguin kit yesterday. Assembled it and ... it doesn't work correctly. When running the autononmous navigation program, it moves a bit, beeps, and kinda just shudders in place. Placing a debug phrase at the top of the program and viewing the output shows that the board is constantly resetting.
I have ran the calibration and centering program numerous times.
I suspect it might be a servo power problem, as the program actually runs normally if I disconnect one of the servos. Doesn't matter which one. And it's not a problem with a tight mechanical fitting either. I have the body sitting with the legs removed, just oscillating their servo horns without load, and it still happens. Sometimes it'll run without resetting for about 4 seconds, but that's as long as it lasts. I've checked the ribbon cable dozens of times.
Even running the servo centering program and trying to turn both servos by hand with a light amount of force causes a reset (verified with debug).

I even bought new batteries, in case mine were duds ($14!). No difference. At least I've ruled that out.

Anybody else have this problem, and/or a possible fix?

By the way, the Penguin manual says that CR123 batteries are sometimes called N. I bought some N batteries at Radio Shack, but they're smaller and are only 1.5 volts. Are there two different N type batteries?

Bruce Bates

romshark -

Regardless of the size designation, the batteries you want are Lithium and are 3 volts each. If you are using 1.5 volt batteries, the Penguin is certainly not getting enough voltage. Sorry I can't help with your other questions, as I don't own a Penguin.

Regards,

Bruce Bates

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
When all else fails, try inserting a new battery.

romshark

Already tried that. I did purchase actual CR123s after I bought the N batteries. Listed as 3V and everything.

Humanoido

If you have a meter, check the load. Check the battery voltage under load with penguin on. What current is it drawing? Check each battery individually too. Either the batteries are low and cause a reset and cannot power the servos, or there's a possible board/servo related anomaly. If you don't have access to the meter, look at the blue power light. It should be very bright. Is it?

romshark

5.74 volts across the batteries while running. Each battery is around the 2.75 to 2.9 voltage range.

Humanoido

What about the LED brightness on the motherboard with the servos
plugged in?

Can you test the servos with another device that uses servos?

Post Edited (humanoido) : 11/20/2008 2:14:34 AM GMT

romshark

The blue LED is pretty bright and constant, although it does flicker ever so slightly when it resets and beeps. It's almost too quick to notice.
I reattached the legs and used servo extension cables to attach the penguin to the spare outputs on my Lynxmotion 5DOF robot arm (an SSC-32 Servo control board) and ran Lynxmotion Visual Sequencer. Arm is powered by an AC adapter (6V .6A).
I've made walking routines before for a Lynxmotion BRAT, so I had a rough walk going in a few minutes. Great, but wierd to have a bird with an umbilical cord. In any event, no problems with the servos or mechanical systems.
Note that the Penguin circuit boards were not installed during this walking test.

Chris Savage

I would bet the servos are not mechanically centered so that when the go to move they're hitting their limit and causing a current surge. Did you perform the centering step?

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Engineering

romshark

Yup. Ran the centering program a bunch of times. I also still had them centered from the SSC-32, so I hooked up the penguin board. Still resetting. Without power, I can turn the servos by hand easily within the full penguin range, so they're not hitting the stops or binding.

Chris Savage

If they're not hitting the stops during the reset cycle then the only explanation I can think of is low batteries.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Engineering

allanlane5

Are these by any chance rechargeable batteries? And Parallax People might be able to answer if that matters?

romshark

They're lithium, but I don't think they're rechargeable. Not by any device I have, anyway.
Would it be possible to test the penguin with an external power supply, like running wires from my BOE-bot, or using the AC adapter from my Lynxmotion arm? I thought of this earlier, but I'm not sure if the boards could handle the possibly higher current. I don't see any 5 volt regulator in the circuit diagram, and I don't want fried chicken penguin.

Ken Gracey

Do you have the servos plugged in with the right polarity on the Penguin's backpack?

Using an external power supply would be a great start. It's possible you're getting a reset condition from the initial servo current draw.

Since you have a Boe-Bot, how about if you run the Boe-Bot's 4AA battery pack leads to the Penguin? Problem solved?

Ken Gracey

romshark

I tested the board with BOE-Power (4xAA, brand new fresh from the pack), and then using the 6V AC adapter. The BOE-BOT power does the same as the the CR123s (no other servos or components were plugged into the BOE-Bot). The AC adapter fared a little better, but it STILL reset every few movements. That's with the legs not attached.

I did have an idea, though. I plugged the penguin servos (still on the servo extension cables) into the BOE-Bot's servo ports. I wrote a quick BS2 program to make the penguin walk crudely forward. No problems so far. I then used some wires and connected the CR123s to the BOE-Bot. Lo and behold, it walks and there's no reset!

I'm gonna tweak the program and see if it'll work on the Penguins' 2px chip. If that fails, there might be some glitch on one of the boards.

If anyone wants to see the wierd BOE-Bot-Penguin-CR123 wiring mess, I uploaded the pic for fun.

romshark

OK, so here's what happened since my last post:
Tried code directly on penguin board. Board is still resetting.
One of the servos finally stripped a gear. Couldn't repair. Ordered new servos from Parallax.
In the meantime, tested the Penguin board with standard size servos (Hitec 422 type). Works without problem. Of course, those won't fit in my penguin.
Today, some stuff I ordered from Lynxmotion came, including two micro servos for another project. Tried them in the Penguin board (they won't fit in the actual body, so they have no load.) It resets.
Hooked up the CR123s to the BOE-Bot again. My test program drives the micro servos on the BOE-Bot without a problem.

Now here's the neat part: I turned on both the BOE-Bot and the Penguin as a test, so that I'm driving both the Boe-Bot and the Penguin board, as well as a pair of standard servos and a pair of micro servos, off a single pair of CR123s. Both units are running my test program. With micros on penguin and standard on BOE-Bot, the penguin resets and BOE-Bot does fine. If the micros are on BOE-Bot and standards are on Penguin, everything works flawlessly.

I'm going to try to see if there's some sort of weak solder joint or something on the penguin board, but I'm not sure how much I can do without voiding any sort of warrenty. Not even 100% sure that's the problem.

Any advice, thoughts, or suggestions? Anything is appreciated.

Ken Gracey

romshark,

Contact me offline kgracey@parallax.com and I'll get you a new Penguin board. You've got enough time into this one and it's possible that this board slipped through a test procedure. All I can imagine is that the power supply tolerances are loose enough to cause a reset. I'll send you an extra servo, too.

Ken Gracey
Parallax, Inc

GWJax

I've been following this thread from the back seat since I do not own a Penguin. This is one reason why I love the Parallax parts, mainly because of the support that goes into the products. Thanks Ken for taking the time to help out the members in need and standing behind the products..

Jax

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
If a robot has a screw then it must be romoved and hacked into..