|
|
|
 However I do wish to reassure people that the judges did not just fall off the back of a turnip truck... - dtengineering [more] |
 |
|
|||||||
|
||||||||
By: Kevin Leonard
New: 19-07-2013 23:31
Updated: 19-07-2013 23:31
Views: 1619 times
The aftermath of our first match- and the reason we put up so few points all day.
Note to everyone: NEVER get into a pushing match with The Hawaiian Kids.
This is the broken dog gear that rendered our drive train useless all day.
We have now reinstalled our single speed DT from championships, so we should be good to go for Saturday.
Discussion
view entire thread
20-07-2013 10:52
DampRobot
Re: pic: 20's IRI Carnage
Wow, that's amassing!
What company's shifter did that dog come from? What kind of ratios were you and Hawian Kids using? How did you end up fixing your DT?
20-07-2013 15:15
DonRotolo
Re: pic: 20's IRI Carnage
Gives a whole new meaning to the term "Bad Dog"...
20-07-2013 15:27
ErvinI
Re: pic: 20's IRI Carnage
That looks like it would make a great logo. Some team should change their name to "The Broken Dogs" and use that.
You could see the wheel rip off at around 1:08 in this video:
http://www.youtube.com/watch?v=0SO7zcf4T9w
20-07-2013 16:37
Garrett.d.w
Re: pic: 20's IRI Carnage
Wow, that's amazing. We had a similar problem when we broke the bolt that holds the dog to the shifter pin. We weren't going to cancel on the Girl's Gen scrimmage happening the next day, so we drilled another hole and pushed a roll pin in there as a bandaid. After the scrimmage I rebuilt the trannie and put a new dog in. The old dog was about to break because the "lightening hole" that had been put in it to fit the roll pin had severely weakened it.
21-07-2013 10:59
Michael Hill
Re: pic: 20's IRI Carnage
I guess you could say...it wasn't very clutch?
21-07-2013 21:23
Kevin Leonard
Re: pic: 20's IRI Carnage
|
Originally Posted by DampRobot
Wow, that's amassing!
What company's shifter did that dog come from? What kind of ratios were you and Hawian Kids using? How did you end up fixing your DT? |
After the break, we asked around for a replacement dog- the only team that had the right dog was 868, who loaned it to us. Due to some issues, that one ended up breaking as well (incorrectly drilled hole, I think. I was in the stands scouting the whole time.)
We ended up replacing the shifting gearbox with our old single-speed from during the season (the 2-speed was an improvement to allow us to cycle better at IRI). Great improvement, right?
We also added a back-of-the-pyramid shot, a full-court blocker, and a centerline autonomous routine that we didn't end up getting to use. Rather disappointing that we put in all that work for nothing.
|
Originally Posted by Michael Hill
I guess you could say...it wasn't very clutch?
|
Too Soon.
On the bright side, we gained valuable experience!
And the matches at IRI were incredible.
Those finals were unbelievable.
21-07-2013 23:23
KrazyCarl92
Re: pic: 20's IRI Carnage
Some more back story on this failure:
During build season we decided to use a single speed gearbox in our drive train because of our lack of experience as a group with shifting gearboxes along with our desire to keep on schedule. Given that our strategy was floor loading to pick up discs scattered around the field, we did not deem high speeds necessary anyway. The 11 fps gearing with direct drive from the kit of parts gearboxes to 6 in. wheels was a nice sweet spot for a single speed that would allow us to still traverse the field when needed, but give us decent acceleration and not trash our CIMs. To make this happen, we "West Coasted" the tough boxes from the kit and ran 1/2" hex 7075 aluminum shafts in our 6WD. We used the 6 in. wheels from the kit because they performed well in 2012 and we wanted enough clearance for the frisbees on the carpet. This robot was the most reliable one we have ever built. For example, in Connecticut we had 4 total minutes of repairs all weekend.
Fast forward to after champs. We were applying for IRI, and in hopes that we would get in we started planning improvements for our robot as a good summer project and a way to really push the envelope with our design. These added features included a hockey stick/window motor blocker, an 1114 inspired pneumatic shot deflector for shooting from the back of the pyramid, and a 2-speed transmission from West Coast Products. We realized that the strategy of Ultimate Ascent had evolved, and especially at IRI it would require all teams to be proficient at cycling if they hoped to be effective in eliminations. Our limiting factors as a team were our speed to traverse the field when cycling and the sub-optimal drop on the floor loading. We couldn't deal with the sub-optimal loading problem easily (4 sec. on average versus 1-2 sec. for most good cyclers at IRI), so we decided that speeding up our drive train while still protecting our motors and maintaining pushing ability would be important.
After getting all of these improvements finally installed with time before IRI, we started programming and drive practice. We got about 30 hours of solid drive practice in. By the end of drive practice, we were rather consistently hitting 5 cycles, sometimes 6, on an open field with our 3 disc auto from the back of the pyramid, along with cleaning up missed shots and scoring those with our end game hang. We also had a 4 disc center line auto that returned to our side of the field from the center line before anyone in the world got there in the first place...I really wanted to see that one in elims!
So then at IRI, we have our first match in Quals 6. We are with 2056 and 2826, and against 67, 359, and 3641. We knew we could not leave HOT entirely open for FCS, so we decided we would "puppy guard" the loading station. We would have our blocker up on them when they went to full court shoot, but in between pick up the deflected shots and alliance member misses and/or score our discs we had collected. As a result of this strategy we found ourselves in a pushing match with 359. Our robot was up on end at one point and then slammed down, presumably breaking one of our KoP wheels. The wheel ripped in half. Then our dog from our shifter shattered into 3 pieces, perhaps because the gearbox was subject to abnormal conditions in low gear due to the broken wheel.
After the match, we quickly realized the dog was missing and started asking around for assistance. Luckily Team 868 had a spare dog for this shifter and gave it to us, along with helping us find the necessary tools for the job. When installing this dog, the aluminum threads for the 4-40 bolt hole stripped out using the non-torquey end of an allen wrench. We tried an 1/8th" roll pin, but the hole for the pin was drilled too large for it to grab. This spare dog subsequently shattered because of the oversize hole. At this point we began trying to lock our gearboxes into one of our gears in hopes of working Friday afternoon. More replacement dogs were nowhere to be found because the WCP dogs go on a 5/8" hex shaft, unlike the 1/2" AM ones. None of our jobs worked, so Friday evening the decision was made to go back to our original single speed gearboxes that worked all season. We got back working Saturday morning and were finally able to put up 70, 67, and 49 points in our 3 matches.
Looking at the wheels after the competition, we can clearly see bubbles in the plastic in the places where they broke. I spoke with Andy Baker at the competition and he did say that AndyMark will be looking to improve the wheels next season. This was not the first time this season the wheels have cracked, but it is the first time they ripped in half like this.
The WCP gearbox was actually a success. After 30 hours of drive practice, the gears held up great and our drivers love the fast, smooth ride. The shifting did bring an added layer of complexity to the failure which meant a very serious failure, but the dog only failed when it was subject to conditions it is not designed for. The prevailing theory is that rather than being traction limited, the gearbox was now limited by the stall torque of the motors in low gear. At least until the dog broke. This is roughly 3 times the torque that the dog is intended or expected to handle.
Thank you to Team 868, especially Charles, for your assistance in trying to get us back up and working. Even though we didn't pull it off for the 2-speeds like we had hoped, it was really awesome to have competitors like yourselves helping us out even though we had a match against each other shortly after. THAT was gracious professionalism. Also thanks to 11, 71, 195, 1310, and 1625 as well as Chris Picone and Nick Lawrence for lending us tools and brainpower for the attempted repairs throughout the weekend. And thanks to 16 and 27 for being understanding pit neighbors when our stuff undoubtedly spilled into your space during the turmoil that was the Friday of IRI 2013 for Team 20. Additional thanks to all of the great teams and volunteers who provided the most inspirational product on that field. We were hoping to really help raise the level of competition, but we came away with a great experience nonetheless.
It was disappointing to finally earn our way to an event that my teammates and I have been dreaming of competing in for years, only to lose more than half of our competition within the first minute. It felt like we never got the chance to show the world what we had worked so hard on. Even more embarrassing was being the team to show up to IRI that didn't move. I really didn't get the chance to enjoy IRI this time around until the eliminations started (which were spectacular, by the way!) because of the stress surrounding the robot failure. It was encouraging to finally see our robot working well Saturday morning, and we got quite a few figh hives. You can bet we will be working hard in the next 9 months to earn our way back next year and hopefully actually compete in 2014.
21-07-2013 23:35
brennonbrimhall
Re: pic: 20's IRI Carnage
|
Originally Posted by DampRobot
Wow, that's amassing!
What company's shifter did that dog come from? |
West Coast Products was kind enough to manufacture a modified version of their gearbox for us. We collaborated to find a solution that could replace the single speed gearboxes we had on for WPI, Hartford, and Archimedes, since our robot was not originally designed for a two-speed.
Even with the footage and the accounts from eyewitnesses, we still don't quite know what the root cause of the failure is, though we have a few theories. We hope that once we fully understand what exactly happened that we can give them some feedback on this gearbox design that they graciously made for us, and we implemented as an offseason project/experiment. If anything, this was an unfortunate accident, but we hope it becomes a valuable learning experience for us and others.
Thank you to all the teams who helped us throughout the day – there were many teams who lent us drill bits, a spare dog gear, and gear pullers. Your gracious professionalism and coopertition was absolutely inspiring.
22-07-2013 01:05
waialua359
Re: pic: 20's IRI Carnage
Aloha everyone,
Sorry to hear about your transmission and wheel issues. We can only imagine the frustrations of trying to fix a robot during a competition, after spending so much time preparing for IRI.
Our robots right side transmission failed during semis 2-2 and we were quite disappointed to not be able to give it our best as well in that crucial match. We plan to fix it at TRR on Friday once we check in. Fortunately for us, we designed our robot for easy access to swap out the transmission with a spare. We always pack spares of everything given that we travel a few thousand miles away from home in order to compete.
Our driver did quite a bit of shifting during those situations, and its too bad it gave out as early as it did.
When our driver tried to push your robot out of the way from HOT's full court shooting, we noticed your robot was going to flip so we backed off a little then continued to free HOT's FC shots.
I guess people took notice since we had many teams visit our pit to ask us about our drivetrain and take pictures.
Our 2010 robot was arguably the best pushing robot in FRC that year. We won a lot of matches including 47 in a row (with ties) and IRI that year primarily due to that fact.
On a side note:
For our drivetrain, I posted a photo of our transmission that we were working on, on FB to FRC members to see which had the motor combination that was described. However, we elected not to use them and still utilize just a 4 CIM motor drive. We do however make custom/swap changes to our AM SS transmissions.
The drive we plan to use next year will effectively be much stronger (for pushing) and quicker. 
We already have done observational tests on a prototype. As mentioned once by a 254 mentor, increasing the speed of a robot should only be done if your driver can handle it. We think he can as he has expressed to us when I asked him. Our driver is a sophomore who has already been to IRI 3 times as a drive team member. He has 2 years left.
22-07-2013 09:15
apalrd
Re: pic: 20's IRI Carnage
WCP dog gear is 7075 Aluminum. Hardenening spec not listed on website. It's 5/8" hex as stated earlier.
AM dog gear is stainless steel. No alloy spec on website. It's 1/2" hex. Some use a 4/40 bolt, others use a roll pin to attach the dog to the shift linkage. We've always gone with the roll pin variant for the comp bot and never broken one.
Maybe 7075 isn't up to the task? I've never seen an AM dog break like that. Every one I've seen break is because of the bolt/pin breaking off inside.
Shock loads can be brutal on drivetrain components. Just sayin. The dog can certainly handle the torque loads of the motors, I would guess indefinitely. But shock loading (especially high-speed impacts) can subject drivetrain components to many times the load the motors are capable of. Just think about the time (acceleration) required to get up to speed under motor power, vs the time (acceleration) required to get back to 0.
22-07-2013 10:15
KrazyCarl92
Re: pic: 20's IRI Carnage
Andrew,
The bolt/pin sees none of the torsional loads because the hex transmits the torque. Therefore, the bolt failures are a result of shear stress.
You may be right that 7075 Aluminum is not up to the task. However, in this case, the task would have been 58.4 ft. lbs. of torque after the wheels broke. At a 40 Amp traction limit (normal conditions), the load is 17.6 ft. lbs. of torque. The fact that the dog broke indicates that the failure was a torsional failure. Normal conclusions about the comparative strength of the dogs are not applicable because of the wheel failure preceding the dog failure.
The video of the match shows these events happening in order:
1. Pushing match with 359, with 20 slamming on the ground on the end with the broken wheel
2. Wheel breaking off
3. Robot still performs multiple 0 turn radius pivots, indicating both sides of the drive train are moving
4. One side of the drive train stops moving
The transition between 3 and 4 indicates that the dog failure happened after the wheel failure. Therefore, the conditions which that dog saw were very different than most (if not all) conditions seen in previous failures. This, along with the different material and geometry of the part as compared to the standard AndyMark dog make it hard to determine if shock loads or stall torque could cause the torsional failure.
In what ways can the torque seen by the dog at the output shaft in shock loading exceed the traction limited torque of the gearbox by a factor of 3? The most I can physically make sense of is a factor of 2 when changing from full speed reverse to full speed forward or engaging in a pushing match at full speed to a stop.
22-07-2013 11:06
apalrd
Re: pic: 20's IRI Carnage
I agree this is a torsion failure. I was saying previously that I had never seen such a failure of the AM stainless dog.
You are only considering torque originating from the motor. Hitting another robot at high speed can cause very fast acceleration on the robot, which results in a huge force on drive line components. While all of the steady-state math would see the wheels spinning and such, there is still a shock on the drive line components momentarily (and it is HUGE).
I would additionally guess the dog had been failing for some time, and there were already stress cracks at two or all three of the fracture locations, from repeated shock loading.
I still believe the cause of failure is improper material selection or component design more than anything else, not improper use of the gearbox (it should be able to handle this). 7075 is a significantly weaker material, and without changes to the dog design (e.g. increasing thickness, additional hardening, etc.) the overall strength is significantly lower vs stainless steel.
22-07-2013 11:17
KrazyCarl92
Re: pic: 20's IRI Carnage
|
Originally Posted by apalrd
I still believe the cause of failure is improper material selection or component design more than anything else, not improper use of the gearbox (it should be able to handle this). 7075 is a significantly weaker material, and without changes to the dog design (e.g. increasing thickness, additional hardening, etc.) the overall strength is significantly lower vs stainless steel.
|
22-07-2013 12:15
apalrd
Re: pic: 20's IRI Carnage
I would guess that the gear would still fail before the stainless dog (or possibly at this point the 7075 hex shaft would become rounder), but the failure load would go up quite a bit relative to the 7075 dog, but that is just a guess. YMMV.
The dog is a tiny, high-load piece with many stress risers. It is subject to significantly more overall stress than either gear mating, as it sees loads when in both high and low (as opposed to only one of the two) in addition to shock loads during shifting (again, ~2x that of either high or low). It's small enough that, even in stainless, it only weighs 0.044 pounds (According to AM website).
I have never run aluminum shifting gears in a drive gearbox, so I can't comment on durability with good evidence. I have run aluminum gears elsewhere (in FIRST) and seen failures of many aluminum parts (not in FIRST).
22-07-2013 12:16
AdamHeard
Re: pic: 20's IRI Carnage
We ran 7075 Dogs w/ 6 motors (4 CIM, 2 775) and 2" wide roughtop wheels w/o issue in 2011.
Unsure how our dogs then compare in size to WCP's.
I'd have to check for exact ratios, but it was something like 7/18 fps.
22-07-2013 13:17
apalrd
Re: pic: 20's IRI Carnage
I was curious about this failure, so I pointed a more knowledgeable person to this thread. His suggestions:
|
-Without a small radii at the corners, there are huge stress risers at all of the corners. -The cross section at the bolt hole is extremely small, and the stress concentration here will be extremely high. -It would be interesting to try Ti for this part. |
22-07-2013 13:48
Cory
Re: pic: 20's IRI Carnage
|
Originally Posted by apalrd
I still believe the cause of failure is improper material selection or component design more than anything else, not improper use of the gearbox (it should be able to handle this). 7075 is a significantly weaker material, and without changes to the dog design (e.g. increasing thickness, additional hardening, etc.) the overall strength is significantly lower vs stainless steel.
|
Most common grades of stainless are actually substantially weaker than 7075, unless AM is using a prehard grade (13-8, 15-5, 17-4) or 410/416
22-07-2013 14:08
KrazyCarl92
Re: pic: 20's IRI Carnage
|
Originally Posted by Cory
7075 is not the problem. We have used it since 2007 with zero failures.
Most common grades of stainless are actually substantially weaker than 7075, unless AM is using a prehard grade (13-8, 15-5, 17-4) or 410/416 |
22-07-2013 14:12
AdamHeard
Re: pic: 20's IRI Carnage
|
Originally Posted by KrazyCarl92
Do you mean 7075 is stronger per unit weight, or stronger for a given cross-sectional area?
|
22-07-2013 14:16
Cory
Re: pic: 20's IRI Carnage
|
Originally Posted by KrazyCarl92
Do you mean 7075 is stronger per unit weight, or stronger for a given cross-sectional area?
|
22-07-2013 14:35
KrazyCarl92
Re: pic: 20's IRI Carnage
Just did a quick matweb search and found that the following types of stainless steel all have ranges of yield strength beyond the 73000 psi listed for 7075 Aluminum:
302, 305, 308, 309, 310, 329, 347, 348, 403, 405, 410, 414, 416, 420, 422, 446, 450, 455, 630, 651, 660 & 675
Some of these appear to be dependent on the heat treatment while others are not. I have no feel for which of these types of stainless steel AM may be using or which kinds are even common, so it is tough to have a good basis of comparison for the different materials used in the dogs.
22-07-2013 14:42
apalrd
Re: pic: 20's IRI Carnage
After looking at a bunch of material sheets, I agree with Cory. Stainless steel is actually quite weak compared to many alloys of non-stainless steel. I surprised myself here.
Why stainless was chosen instead of hardened steel for the AM part is then a question as well (unless it is a prehard stainless alloy).
Which brings us to component design failure. Either the design load was exceeded (possible, but then design load should be increased) or design was not properly analyzed (also possible, likely shock loading or fatigue not correctly estimated in that case).
Is this picture the first dog to fail or the second? If it was the second that might explain the failure more, as the hole was drilled larger, at the point of highest stress.
22-07-2013 15:01
KrazyCarl92
Re: pic: 20's IRI Carnage
The dog depicted is the second one that failed (I can tell by the lack of threads in the through hole).
However, the first one that failed also fractured at three points: 1 at the threaded bolt hole and 2 other places exactly where the dog goes from thick to thin, as shown in this picture. I can post a picture of this failure as well later tonight.
22-07-2013 15:51
R.C.
Re: pic: 20's IRI Carnage
|
Originally Posted by AdamHeard
We ran 7075 Dogs w/ 6 motors (4 CIM, 2 775) and 2" wide roughtop wheels w/o issue in 2011.
Unsure how our dogs then compare in size to WCP's. I'd have to check for exact ratios, but it was something like 7/18 fps. |
This dog model is pretty much the exact same since the one we've used on 973/1323 bots since 2011. We didn't randomly pick 7075, its been tested on 254/968/973/1323/1477 etc... for a pretty long period of time.
The gearing seems about right, 7/17 or 7/18 should be it.
We haven't had a dog fail except:
https://fbcdn-sphotos-c-a.akamaihd.n...22823429_n.jpg - 2012 MTTD Offseason event - Practice Bot.
This guy was before WCP started making them, our sponsor made these and this dog was drilled off center. There was more material on one side than the other. We treated this as a random occurrence and really aggressive driving.
But for us, since 2011. 6-8+ robots (pbots included) have used this same dog design with probably over 1,000-2,000+ hours of driving with no issues. So we feel pretty safe with the material choice and run time, more or less we try to test all our products out for an extended amount of time before putting them into the market.
For this occurrence we'll be looking into it and trying to:
-Replicate Breakage
-Find the exact problem
-And a fix for it. (Going to 4140 or some other steel alloy)
I'll post back when we find something out.
-RC
22-07-2013 17:01
Adrian Clark
Re: pic: 20's IRI Carnage
I've often wondered about the safety of al dogs, so I find this thread very interesting!
R.C./OP, if either of you are interested I would love to run a simulation of this situation and post the simulation report as a whitepaper. To do this I would need to know a little more about what happened, mainly how much approximate torque the dog was subjected to.
-Adrian
22-07-2013 17:27
R.C.
Re: pic: 20's IRI Carnage
|
Originally Posted by Adrian Clark
I've often wondered about the safety of al dogs, so I find this thread very interesting!
R.C./OP, if either of you are interested I would love to run a simulation of this situation and post the simulation report as a whitepaper. To do this I would need to know a little more about what happened, mainly how much approximate torque the dog was subjected to. -Adrian |
Email me when you can and I'll CC Carl in the email.
Thanks!
-RC
22-07-2013 19:31
JChang
Re: pic: 20's IRI Carnage
I would like to know how 254/973/1323 connect the dog to the moving shaft inside the output axle (roll pin, bolt or something else).
Our team used the WCP dual speed gearbox for this season, but a few days before ship date, the 4-40 bolt (threaded all the way) snapped. In the same week, the same thing happened to our two years old AM SuperShifter as well. We didn't want to have this problem on the field, so we drilled through the thread with an 1/8'' drill and punched a roll pin in. Our very old AM Shifter have the the roll pin setup and never failed on us. AndyMark now improved the dog design by custom making a bolt with threads only at the end (am-1272).
After 4 events (BMR, CRR, Champs, Indiana State) and weeks of testing for IRI, the modified dog gear finally failed, just 6 days before IRI. By that time the hole was enlarged by the shifting motion, causing the dog to break. We rush ordered the dog from WCP and thankfully, it came in time for us to complete the repair on IRI load in day. On the new dog, we drilled only halfway through the threaded part, leaving more materials on that side. I handed the partly drilled dog to Team 20, and they tried to repair it with the 4-40 bolt. There wasn't enough thread left for the bolt, so they went with the roll pin method. Unfortunately they used a bit that is to large, and when we went up against The Rocketeers, the second dog shattered. I hope this never happen to any team ever again.
Some suggestions for improvement:
- Make the OD of the dog larger, there is plenty of room before interfering with the 45t gear.
- Use roll pin or solid body 4-40, the current bolt will snap when inexperienced programmers attempt to write an auto-shift code.
- Thicken the thin part of the dog if there is room
-Josh
22-07-2013 20:05
Cory
Re: pic: 20's IRI Carnage
|
Originally Posted by JChang
I would like to know how 254/973/1323 connect the dog to the moving shaft inside the output axle (roll pin, bolt or something else).
Our team used the WCP dual speed gearbox for this season, but a few days before ship date, the 4-40 bolt (threaded all the way) snapped. In the same week, the same thing happened to our two years old AM SuperShifter as well. We didn't want to have this problem on the field, so we drilled through the thread with an 1/8'' drill and punched a roll pin in. Our very old AM Shifter have the the roll pin setup and never failed on us. AndyMark now improved the dog design by custom making a bolt with threads only at the end (am-1272). After 4 events (BMR, CRR, Champs, Indiana State) and weeks of testing for IRI, the modified dog gear finally failed, just 6 days before IRI. By that time the hole was enlarged by the shifting motion, causing the dog to break. We rush ordered the dog from WCP and thankfully, it came in time for us to complete the repair on IRI load in day. On the new dog, we drilled only halfway through the threaded part, leaving more materials on that side. I handed the partly drilled dog to Team 20, and they tried to repair it with the 4-40 bolt. There wasn't enough thread left for the bolt, so they went with the roll pin method. Unfortunately they used a bit that is to large, and when we went up against The Rocketeers, the second dog shattered. I hope this never happen to any team ever again.Some suggestions for improvement:
-Josh |
We used the new Andymark shoulder bolts for our dogs.
22-07-2013 20:06
R.C.
Re: pic: 20's IRI Carnage
|
Originally Posted by JChang
I would like to know how 254/973/1323 connect the dog to the moving shaft inside the output axle (roll pin, bolt or something else).
Some suggestions for improvement:
-Josh |
We use a bolt and shift at a lower PSI than most teams. We were able to shift at about 30 PSI with the 9/16" cylinder we use to use. We shift at about 20-30 with the current 3/4" Pancake Cylinder. A decent amount of teams shift at 60 PSI.
We are going to be using the solid body screw similar to AndyMark (Waiting for them to come in). We also used to just switch the bolt out at the beginning of each competition. As you can get a box of 100 screws for $4-5. Now we go pretty much go 5+ competitions and demo's without switching.
-RC
22-07-2013 20:33
DampRobot
Re: pic: 20's IRI Carnage
|
Originally Posted by Cory
There are good reasons to keep the dog as small as possible. It opens up the ability to use a lot of other custom reductions that would otherwise be interfered with. We had to actually turn ours down a touch to get enough clearance for the ratio we ran.
We used the new Andymark shoulder bolts for our dogs. |
I wouldn't worry too much about this type of failure. This dog saw far more torque than it ever usually would. This is the only failure of a dog I've ever seen, usually its just a broken roll pin or something.
After more fully understanding the issue, I'm surprised the debate has revolved around the dog, when it was the wheel that failed when it shouldn't have.
22-07-2013 20:34
AdamHeard
Re: pic: 20's IRI Carnage
|
Originally Posted by DampRobot
I wouldn't worry too much about this type of failure. This dog saw far more torque than it ever usually would. This is the only failure of a dog I've ever seen, usually its just a broken roll pin or something.
|
In high gear, a full speed direction change will result in 2x stall torque being applied.
22-07-2013 20:42
AdamHeard
Re: pic: 20's IRI Carnage
|
Originally Posted by JChang
I would like to know how 254/973/1323 connect the dog to the moving shaft inside the output axle (roll pin, bolt or something else).
[/list] -Josh |
We used 7/16" in 2012, but that was a single CIM per setup and had no initial reduction.
We use a standard #4-40, have never broken one.
We limit the stroked so that fully extended or fully retracted the dog is not loaded into the face. Of course when it's shifting if the dog gear hits the other gear tooth to tooth it will experience load while one rotates, but other than that we never actually load the gears.
I think the combination of shimming and lower force is what has caused us to never break a screw.
22-07-2013 21:15
KrazyCarl92
Re: pic: 20's IRI Carnage
|
Originally Posted by AdamHeard
I'm not understanding this point, how did it see more torque than it usually would?
In high gear, a full speed direction change will result in 2x stall torque being applied. |
We were in low gear when the wheel broke and stayed there for a while.
2 x 2 x 45 x 60 / 11 / 15 = 65.5 times stall torque of a CIM seen at the dog. More like 44 times CIM stall torque when you account for voltage drop. Normally the wheels would have slipped long before that point was reached, but one was broken, so the torque was unable to turn the wheels. Don't know to what extent this is true but there was certainly more than the normal traction limited amount of torque being demanded of the dog.
22-07-2013 21:27
AdamHeard
Re: pic: 20's IRI Carnage
|
Originally Posted by KrazyCarl92
That is in high gear. So 2 x 2 x 45 x 45 / 11 / 30 = 24.5 times stall torque. Also remember voltage drop and it's far more like 16 times CIM stall torque at the dog.
We were in low gear when the wheel broke and stayed there for a while. 2 x 2 x 45 x 60 / 11 / 15 = 65.5 times stall torque of a CIM seen at the dog. More like 44 times CIM stall torque when you account for voltage drop. Normally the wheels would have slipped long before that point was reached, but one was broken, so the torque was unable to turn the wheels. Don't know to what extent this is true but there was certainly more than the normal traction limited amount of torque being demanded of the dog. |
I did however forget the ratio difference between high and low, thanks for correcting me there.
22-07-2013 21:29
Aren Siekmeier
Re: pic: 20's IRI Carnage
|
Originally Posted by DampRobot
I wouldn't worry too much about this type of failure. This dog saw far more torque than it ever usually would. This is the only failure of a dog I've ever seen, usually its just a broken roll pin or something.
After more fully understanding the issue, I'm surprised the debate has revolved around the dog, when it was the wheel that failed when it shouldn't have. |
22-07-2013 21:38
apples000
Re: pic: 20's IRI Carnage
|
Originally Posted by apalrd
-It would be interesting to try Ti for this part.
|
Our Ti dog didn't break and we could make it even smaller, but it was expensive and a MASSIVE pain to manufacture, so we had a machine shop (that a mentor worked at) make our second set for us. Unless you have a real need, and a ton of machining resources, I would advise against this.
22-07-2013 22:10
Aren Siekmeier
Re: pic: 20's IRI Carnage
|
Originally Posted by apples000
... and a ton of machining resources ...
|
22-07-2013 22:41
apalrd
Re: pic: 20's IRI Carnage
|
Originally Posted by apples000
We've done this before with a PTO/brake setup where the wheels would lock when the PTO was engaged and the momentum of the whole robot could rip apart our hardened steel dog.
Our Ti dog didn't break and we could make it even smaller, but it was expensive and a MASSIVE pain to manufacture, so we had a machine shop (that a mentor worked at) make our second set for us. Unless you have a real need, and a ton of machining resources, I would advise against this. |
Titanium alloys have a low modulus elasticity but high yield strength/ultimate tensile strength, and fairly low density. So it's springy but won't break easily, and it's lighter than steel for the same strength, which is perfect for this application. But machining it is more difficult than steel or aluminum, and welding it is also hard (not that this piece requires welding, but it's hard).
I like doing this kind of failure/material analysis. It's fun and I always learn a lot.
23-07-2013 11:13
KrazyCarl92
Re: pic: 20's IRI Carnage
One odd thing that we did with our implementation of the shifters was putting one of our talons in coast mode and one in brake mode on each side of the drive train. Our drivers did not like that the robot was coasting so much when we lined up to shoot from the corner and it significantly increased our time to line up for shots. We tried putting all of our talons in brake mode on the drive train and the motion was far too jerky to be desirable. We then decided to try one motor in brake mode and one in coast mode and we found it to be a great sweet spot. The motion was not too jerky and we were capable of lining up for shots much more quickly.
Could this have contributed to the dog failure? Intuitively it would not seem to make that much of a difference, but I do not know how many teams have tried running a configuration similar to this.
23-07-2013 13:03
MichaelBick
Re: pic: 20's IRI Carnage
|
Originally Posted by KrazyCarl92
One odd thing that we did with our implementation of the shifters was putting one of our talons in coast mode and one in brake mode on each side of the drive train. Our drivers did not like that the robot was coasting so much when we lined up to shoot from the corner and it significantly increased our time to line up for shots. We tried putting all of our talons in brake mode on the drive train and the motion was far too jerky to be desirable. We then decided to try one motor in brake mode and one in coast mode and we found it to be a great sweet spot. The motion was not too jerky and we were capable of lining up for shots much more quickly.
Could this have contributed to the dog failure? Intuitively it would not seem to make that much of a difference, but I do not know how many teams have tried running a configuration similar to this. |
23-07-2013 13:19
AdamHeard
Re: pic: 20's IRI Carnage
|
Originally Posted by MICHAELABICK
This seems like it could have contributed to the failure. What you may want to try in the future is to have the drivers pull back on the joysticks a little when they want the robot to stop. This will create a situation electrically very similar to the break mode on the speed controllers, without the extra jerkyness
|
23-07-2013 13:45
apalrd
Re: pic: 20's IRI Carnage
In any case (Talons/Victors/etc. in Coast, in Brake, one in Brake, driver pulls back on stick) the drive line will transmit torque in the opposite direction (from motors>wheels to wheels>motors or to -motors>-wheels) and the gearing will transition through all of the lash, then hit the other load surface (a shock load on the dog and all of the gears/sprockets). The motor will spin more freely when in Coast, but the lash transition will still happen. The force of this is relatively low in any brake-mode case, as the motor provides zero brake force at zero speed (it acts as a generator to power itself, so the torque it reacts is proportional to the speed it is forced at, when in brake mode, not including electrical/efficiency losses). The torque reaction from the motor with reverse power braking will be related to the applied voltage and drive speed/
I've always liked driving full coast, and learning to coast/slow down properly into what I'm going for (or just hit it off-throttle if it's solid). The Cheesy and Culver drives were both fantastic improvements on the two-stick skid steer in that there is guaranteed to be no twist from asymetrical motions between the two sticks, since the throttle and steering are separated.
Edit: The more I think about cases like this, the dog is under a TON of repeated shock loading like this.
23-07-2013 16:02
KrazyCarl92
Re: pic: 20's IRI Carnage
|
Originally Posted by apalrd
Edit: The more I think about cases like this, the dog is under a TON of repeated shock loading like this.
|
23-07-2013 16:10
apalrd
Re: pic: 20's IRI Carnage
If anything, the code would have more effect, since it's actively reversing torque immediately, while the electrical brake requires the motor to be overdriven to begin producing reverse torque.
But a forward-reverse slam initiated by the driver would put way more shock load on the dog than this would.
I'm saying that this part likely sees extremely high repeated shock loading, so the analysis of it is more complex. This poor little part has such a huge job to do.
23-07-2013 16:16
KrazyCarl92
Re: pic: 20's IRI Carnage
|
Originally Posted by apalrd
If anything, the code would have more effect, since it's actively reversing torque immediately, while the electrical brake requires the motor to be overdriven to begin producing reverse torque.
But a forward-reverse slam initiated by the driver would put way more shock load on the dog than this would. I'm saying that this part likely sees extremely high repeated shock loading, so the analysis of it is more complex. This poor little part has such a huge job to do. |
23-07-2013 18:17
Travis Covington
Re: pic: 20's IRI Carnage
I think Cory forgot that we actually did break 2-4 of these on our practice bot this year in a similar manner. They were also from WCP but modified to be 1" OD. We had run the same size dog since 2007 without failures, so I am pretty confident that the changes implemented in the WCP ones, along with the additional CIM motor (we had 3 per gearbox) caused the failures for us.
In the past, we had either used a #3-48 screw (which failed a few times) or a 3/32" roll pin (which never failed). I believe the larger diameter hole of the #4-40, as well as the inherent stress risers that exist with there being a thread, combined with the additional torque of our 3 motor gearbox caused our failures. As some have mentioned, the cyclical/shock loads that this part sees are very high, and the relatively thin cross section at that threaded side of the dog does not provide much, if any, factor of safety.
We lucked out because we never had a failure on the competition bot, but did have drilled (non-threaded) dogs as backup. Our kids got plenty of practice and were able to swap a broken dog in under an hour if they had to. I am glad we never had to do that between matches though!
I think for the future we will use non-threaded dogs, as well as possibly increasing the thickness of that section of the dog at the expense of an ever so slightly wider gearbox.
Adam may be able to provide more insight in to their circumstances, but I think 973 may have just lucked out on not breaking their parts considering they had very similar loading characteristics to us. It may have had something to do with the CG of their robot versus ours, and the way it decelerates, but might also be because their dogs are the standard 1.125" OD (I believe?). We broke almost all of our dogs when stopping or changing directions, or when going fast over bumps/metal plates under the carpeting in our lab. Given the failure, those loading conditions make sense.
Hope this helps!
Edit: On second thought, 973's larger plastic wheels might have dampened some of the shock-loading as well. Who knows.
23-07-2013 18:32
AdamHeard
Re: pic: 20's IRI Carnage
|
Originally Posted by Travis Covington
I think Cory forgot that we actually did break 2-4 of these on our practice bot this year in a similar manner. They were also from WCP but modified to be 1" OD. We had run the same size dog since 2007 without failures, so I am pretty confident that the changes implemented in the WCP ones, along with the additional CIM motor (we had 3 per gearbox) caused the failures for us.
In the past, we had either used a #3-48 screw (which failed a few times) or a 3/32" roll pin (which never failed). I believe the larger diameter hole of the #4-40, as well as the inherent stress risers that exist with there being a thread, combined with the additional torque of our 3 motor gearbox caused our failures. As some have mentioned, the cyclical/shock loads that this part sees are very high, and the relatively thin cross section at that threaded side of the dog does not provide much, if any, factor of safety. We lucked out because we never had a failure on the competition bot, but did have drilled (non-threaded) dogs as backup. Our kids got plenty of practice and were able to swap a broken dog in under an hour if they had to. I am glad we never had to do that between matches though! I think for the future we will use non-threaded dogs, as well as possibly increasing the thickness of that section of the dog at the expense of an ever so slightly wider gearbox. Adam may be able to provide more insight in to their circumstances, but I think 973 may have just lucked out on not breaking their parts considering they had very similar loading characteristics to us. It may have had something to do with the CG of their robot versus ours, and the way it decelerates, but might also be because their dogs are the standard 1.125" OD (I believe?). We broke almost all of our dogs when stopping or changing directions, or when going fast over bumps/metal plates under the carpeting in our lab. Given the failure, those loading conditions make sense. Hope this helps! Edit: On second thought, 973's larger plastic wheels might have dampened some of the shock-loading as well. Who knows. |
23-07-2013 21:43
KrazyCarl92
Re: pic: 20's IRI Carnage
|
Originally Posted by Travis Covington
I think Cory forgot that we actually did break 2-4 of these on our practice bot this year in a similar manner. They were also from WCP but modified to be 1" OD. We had run the same size dog since 2007 without failures, so I am pretty confident that the changes implemented in the WCP ones, along with the additional CIM motor (we had 3 per gearbox) caused the failures for us.
|
23-07-2013 21:48
apples000
Re: pic: 20's IRI Carnage
|
Originally Posted by KrazyCarl92
My thought was that electrical braking would not even begin to approach the load demands of a full forward-full reverse transition, so the shock loading from that would be negligible in comparison. That's just a thought experiment though, not supported by any evidence.
|
This is just a thought, but we may have implemented a ramping effect on the drive motor outputs to limit acceleration and prevent tipping, which may have made the full forwards to full reverse a little bit less drastic. We may have also had a linear approximation of drive voltage vs speed, and would limit the drive output to being plus/minus 40% of the approximation. This made sure that if we were being pushed forwards at a high speed, the motors could only go down to 60% power, instead of 100% reverse. It also had the effect of maintaining traction in low gear.
24-07-2013 00:46
Travis Covington
Re: pic: 20's IRI Carnage
|
Originally Posted by KrazyCarl92
Did these dogs fail in the drive shifting application or the PTO application? And what would be the maximum powered torque that the dog would be transmitting in the given application?
|
view entire thread
Tags
loading ...