We have a transversal climb powered by two 775pros through a versaplantary ( I don’t know the ratio off the top of my head ) but it’s a ratio we have used in numerous successful climbing games. We made it through the entire build season and our first competition without much issues.( we had one motor come unplugged in our first match and had the 2nd motor burn up as it was too much for the single motor) but we replaced this at the competition and continued on without error.
We have since improved the climb but know are running into a stalling issue. And have burnt out 3 consecutive 775 pro Originally we thought it was something we changed in the software but after reverting back to the competition program the problem persisted.
We have gone through replaced both talons speed controllers with new ones replaced the motors with known good ones. ( new) gone through all the wiring and double checked everything but still no luck.
Physically the climb starts out as expected With the correct speed and amperage but slowly as it approaches the bar struggles and eventually stalls out.
We peak amperage at 70 per motor before stalling out however when we initiate the climb we are at 20 and then steadily increase as we approach the bar until we stall which we wouldn’t expect I think it should be a similar amperage throughout the climb. our previous data logs show this being the case during our competition climbs.
both motors are consistently with 1-2 amps of each other at any given time, and the output voltage to the motor are also consistent within a few from motor to motor.
We can bandaid the climber, and have it successfully climb by removing the decel command within the program,( we had this commend in at our first competition) but this doesn’t fix the root cause and the symptoms remain the same just momentum at this point helping us. Climb still struggles at the very top and amperage pattern is the same. And we feel over extended time we are still gonna burn motors as is.
We are at a complete loss as to what could be the cause to this.
Any help would be greatly appreciated.
I’d check all the mechanical things first. Did you put all your thrust washer in your VP (I know that is a things for the Andymark Sport boxes. Maybe a thing for VP). If you are using a spring-rope system, is your role catching on something on the way down.
That’s where I would start. I’d set a software current limit so you aren’t as likely to burn up your motors.
We use a rack and pinion climber
The only contact the drives the system is the rack on the gear. And we check the versaplantaries and everything checked out. image|375x500
Honestly, if I were you I would consider trying a different motor. 775s are plenty powerful, but they Do Not Like being stalled, and will burn up very fast compared to other motors. A CIM, miniCIM, or NEO would be a good starting point. You’ll have to adjust your gear ratios, but those motors can be stalled for much longer without burning up.
Alternatively, if you are dead set on using 775s, I would put in some software controls to ensure that you don’t stall the motors. TalonSRXs have current limits, and you could implement limit switches to ensure you don’t continue driving the motor past the mechanisms stopping point.
You could also increase your gear ratio. This is kind of a brute force method that will slow down the climb, but increasing the ratio will give you more torque and therefore more overhead against stalls.
We have considered this but that would require a redesign of the mounting, gearing and our software for our encoder set values, which is a lot to change going into our next competition also this setup works based off our first competition just have to find the bad almond in the setup causing the new amperage spikes which we didn’t not have before hand.
We also do have limit switches on top and bottom of our climb as a safety in case our encoder skips or breaks etc. with our current issue we do not get to climb completion at all either through the encoder or the limit switch. And the current control we had but was being activated prior to climb completion so we had removed it to investigate the route cause.
Gearing is an option but like you said that’s a work around. The gearing we have has pulled the robot up 40 times effortless prior to this, so yes we could change ratios but, that not fixing the route cause of the problem .
This appears to point towards a mechanical issue. It may be helpful to have several people watching the climb from different viewpoints, to make sure all the parts are doing what they are intended to do and are not misbehaving or deforming in some way. Having someone call out the motor currents may also help.
Can you post a better image of the mechanism that holds the rack gear and the gear that drives it? What kind of bearing surface does the rack/arm slide against? It doesn’t look like there are ball bearings to keep it moving side/side, so I’m guessing it’s a low friction surface?
Also, are your two extending hooks connected to each other near the end (perhaps a standoff blocked by the hangar bar in the photo)? Are they both straight? If you are able to grab the bar with the hooks the wrong distance from each other, you will start to see a large side load on the arm where it slides through the box that drives the rack. It will get worse as you get closer to the top, the difference between the width where they’re driven and the width where they hook on the bar becomes more pronounced.
We have done this and it function as they always have, we have to reset the climber by hand post match to calibrate our zero and we manually push the arms back down every time and we have never had an increase of force needed to accomplish this and this remains the same as of last night diagnoses.
All I have access to right now is the CAD of the climber mechanism as I’m not at the shop but I can send a picture of that if it helps
Hooks are sync to be operated on
The same tooth when we install them
And the design makes it impossible for it to skip and tooth and become
Out of sync.
I see that there is some sheet of a slick kind of material bolted to the inside of the rack gearbox to let the arm slide against the side of the housing. Do you know what material it is, and if you are using any grease or graphite powder to keep it slick?
I’ve had some time to precision CAD the failure mechanism that I would be worried about with this kind of system as you have it built. Since the hooks are both independent of each other at the very ends, and they are relatively thin, they could become bent ever so slightly, or even just wiggle into a bad position when you line up to climb. If they grip in the wrong locations, this can happen when you climb:
You would expect to see the current draw from the motors increase as you go higher up; the angle gets worse and the friction increases. This would be a very easy thing to test: next time you line up your robot in your shop, make sure the spacing of the hooks when placed on the bar matches what it would be when they are stowed. If the robot can climb easily in that case then this likely the issue. Easy to fix by adding a standoff between the two hook ends to ensure they always have the same width.
Ayo! Good idea, to add to this we added friction tape on the hooks to help the swing between our first event and now, somthing I wouldn’t have thought would contribute to this but after you mentioned that it would make the situation worse as the hooks metal on metal could slide some as we pull up self adjusting the potential pinch point but with the addition of the friction tape that self adjustment freedom is gone. Causing it to lick up instead of slide on the bar, we will test this first thing when we get to the shop tonight!
My impression is that the climber is geared wrong. The amp reading of 70A is not an instantaneous reading but most likely a filtered average. Also are you operating the motors at less than 100% speed? Speed controllers operate on a non linear basis. So at 50% speed expect your peak current to be about 400% higher. Our two NEO climber is set to operate at 36A and 100% speed. To get more controllability, the programmer slowed the motors to ~50%. The SparkMax shows a 60 to 70 amp reading during the climb. In reality it actually has 130A instantaneous load. Try increasing the gear ratio by ~30% or using a faster controller motor speed. Past experience with 775s, robot lifts, and reduced speed controller setting has led to motor burnout when the gearing is too fast.
Thank you!!! This was the issue, when extended the spacing of the hooks can fluctuate between 11-16 inches apart.
without grip tape as we winched in the arms would slide on the bar and center itself as it went up.
The grip tape was preventing this from happening causing the rack to jam in the slide,
We removed the tap and attempted a climb and it was successful and immediately saw the current values back in the range they should be. We are back to having a reliable sub 20 transversal. We are going to consider putting a brace between the hooks to keep them straight to error proof the climb some more. Thanks everyone for your inputs
