Experience needed regarding FW Beater and Armor

[Runsler]

Hi there.

So... i am finally in a stage of planning where i got most things sorted out.

Still got one big and four minor questions where i just need some experience, and build diaries are either telling different stories or nothing about it at all:

1. (big one)
My Weapon will be in a somewhat beater-style, just two of them on one axis, with a 8cm "gap" of only axis between them and a bit asymmetric to give it a better bite, but yeah. At least roughly. but i am really struggling what material, thickness, and how wide the "spokes" should be.
Dimensions will be (roughly) 30cm long (along the axis) and about 25cm diameter when rotating (those would be really hard to change if i had to).
not sure about that yet (some parts still have to arrive), but i think i will have 4, maybe even 5kg to put into that. Somewhere in that range.
But would like to know the "minimum safe solution" for the thickness, so i can try to do the actual designing on my own and just know when i am trying to do something too likely to blow up.

2. Pulley/belt to drive the weapon. What diameter etc. do you guys use in your featherweights? When reading build diaries i get the feeling of "anything is possible", but long-term experiences are rarer in those. So... would like some advice on what actually worked for a featherweight vertical spinner for quiet a time. Else i'd just have to try, but since i can't make my pulleys myself and changing something on them would be some work... planning to get different diameters in case my spin-up-speed isn't what i imagined anyway, but to ask a friend for that, and a bit lter ask him to do it again just a little bit thinner/wider wouldn't be nice

3. Top Armor sheet metal (most likely 3 or 4mm titanium), side armor 20mm HDPE.
Now i want to screw both together.
Considering i might have to get the top off quiet often (for charging, repairs and whatever), hwat would be a reasonable solution here?
i've so far seen/read/heard of
"just take some wood screws und punch them in there",
"drill holes, cut threads in them, and use M5 (or M6, or M7, or... seems everybody got a favourite size) screws in those" and
"drill holes, put inserts in, and then use Mx screws."
Now... the last one seems to be the most professional one, but also the heaviest and most expensive (and possibly most time consuming) option. But also everybody seems to have different distances between those screw, and use different screws (regarding length and thickness), or even give such helpful remarks as "a lot" or "a bunch". I just can't see a pattern there, when you should use what and why and how much/long/wide the screws...
Guess i could somehow put that together, but i am feeling likely to either not make it strong enough or putting so much metal in my HDPE, i could have taken Hardox right from the start. So would be great to have some experience like "i take 5 screws m5 for every 10cm HDPE and (don't) use inserts" or stuff like that, and if you ever had any problems with it.

4.Now this was just an idea... do you know of any bushings coming as two halves? So i could put them on an axis without an "open" end? Would in this case be used to mount my beater to three bushings, two in the ends and one (almost) in the middle. So let's say take a bulkhead, cut it in half, screw it together again, drill a hole where the cut is, screws out again, put both halves of the bushing around the axis, screws in again, and got one mount in the middle of the axis.
Would think that would give it a bit extra stability, but should also work without... at least i hope so.


Now that's quiet a wall of text, hope someone takes the effort to read that stuff and possibly even answer. If you read this: thanks already

[Andy]

Pendulum uses a 20mm thick plate of toolox that is then watercut for the beater, with the ends turned to spin inside bearings. id say 20mm thickness its minimum, 250mm dia 300mm length is very big though, mines 145mm dia and 240mm side for comparison. Id avoid using multiple parts to make the beater weapon since itll just cause more weak points, id say 25mm thickness is safe if you do go with the size you want.

[lowndsy]

Regarding fixing things together I wasn't sure which approach either, so I'm doing them all!

Using bolts where they will fit and screws everywhere else (lots and lots of screws). The HDPE I have doesn't take a thread very well, and I was worried that if I kept inserting and removing screws for maintenance they would get loose so I have a hinged trapdoor for access, going to try the inserts route for sealing it.

[Runsler]

Pendulum uses a 20mm thick plate of toolox that is then watercut for the beater, with the ends turned to spin inside bearings. id say 20mm thickness its minimum, 250mm dia 300mm length is very big though, mines 145mm dia and 240mm side for comparison. Id avoid using multiple parts to make the beater weapon since itll just cause more weak points, id say 25mm thickness is safe if you do go with the size you want.

Yeah, the scetches already look... well, like an ice cone dropped on the side, to be honest. wish i could say "scary".
Problem is, i can't do the standard-beater-design. It will all be cut from one piece and stay as one piece, but in the middle will be an 8cm part where it is only axis, and not beater. can't avoid that... need that space for the weapon motor and to have a third point on the floor. the weapon got so big, and the only two wheels are so far on the other side of the bot, i think i'll need something to slide on somewhere in that region.
Could slide on the sides, too, but that would need way more weight, is more likely to get hit, and i'd have four points on the ground. somehow i think three points seem to be more stable, even more given the expected gyroscopic effects.
Also, trying to just move the weapon away from its motor would greatly increase the weight needed for armor, and even further increase the diameter of the weapon. And that one is already bigger than i wanted^^

But 25mm as minimum is a number i can work with, sounds good. And how wide would you make the spokes? just 25mm, too because that looks good? Or maybe there's a better way to do those than just rectangular?

Regarding fixing things together I wasn't sure which approach either, so I'm doing them all!

Using bolts where they will fit and screws everywhere else (lots and lots of screws). The HDPE I have doesn't take a thread very well, and I was worried that if I kept inserting and removing screws for maintenance they would get loose so I have a hinged trapdoor for access, going to try the inserts route for sealing it.

Trapdoor is a good idea, in my case i am going for an "inner shell" made from thin polycarbonate that can be taken out completely of the armor. "just" have to unscrew that and the weapon every time... not quiet sure if that is wise, but since i wanted to put that layer in there to save my esc etc. from at least some of the vibrations anyway, it seemed the easier way to do this. I'll see if it works.

(and next bot will be somewhat box-shaped. Really got enough of all these stupid angles and having to work around those.)

Anyway... so maybe just wood screws where i don't need to remove them often, and inserts where i do?
But what size of screws, and how many? every screw is some precious weight^^
Regarding that... are titanium screws a worthy idea, or doesn't the reduced weight count up for the (possibly?) reduced strenght?

[Andy]

my weapon spokes are more like 35-40mm just because i wanted to make the part as strong as possible but 30mm is probably ok, I do not recommend having the wheels at the back of the robot, it will drive badly as they dont have enough weight over them, try and have them in the middle or behind the beater, itll drive far better.

Do you mean youre trying to drive the beater directly from the motor?

[Runsler]

Yeah, i fear the driving might be a problem, but since it will be very slow anyway, i can't hunt anybody down even if i could drive it well. And with those gyroscopics... even turning will be interesting.
Sadly, battery and ESCs are way bigger than the gearboxes, so i had to put them further to the "opening" of the cone where the beater is, and the wheels further to the tip. i still got about 5cm wedge from the wheels to the end of the bot, but yeah, that's not much.

I know it might be stupid, but hey... it's my first bot, and since i don't expect much of it, i thought i could as well try to go for an interesting design not seen that often instead of building the typical box shape. There will most likely other bots follow, where i'll go more for function than form.

And no, no direct drive (i may be stupid and ignore some advices, but not this one^^)
That's why i asked about the pulleys (question 2). It's just... the Motor will basically sit "inside" the weapon diameter.
Imagine a beater. now double it up, with the axis joined together. now in this middle space the motor will sit, and the pulley will go there. (that's why i can't just order one if it doesn't work, i need one made from two pieces with a rectangular cross section i could then screw together again. Possibly another stupid idea, but if i put all of them in one bot, i have less left for the next one!)


and yeah, spokes as wide as possible to make them sturdy seems reasonable, only the weapon could store more energy with the weight further on the outside... still, "not under 30" again is something i could hopefully work with.


(maybe after the weekend i could post some rough scetches/pictures of the parts aligned as in the bot to clarify the design a bit, if you are interested. but not much to see yet, since many materials will only arrive the next few days. mostly just having the "guts" of the bot by now and playing tetris in a paper box with them)

[typhoon_driver]

I need a sketch as I'm not sure I understand the design idea

[Eventorizon]

From what you have said, this is what I am seeing...

Runsler Weapon.JPG

This is to scale, 250 OD, 300mm weapon length, from 20mm plate steel. The pulley in the middle is Ali. Rough weight is 5.6kg

Is this correct?

[Runsler]

That is looking quiet like what i have in mind/on paper. only the part with the pulley is 8cm wide. (just the "gap", not the pulley, that would stay close to one of the sides) so when the weapon rotates, the motor would fit into that gap and doesn't get hit. And the assymetrical part is done so one side really is a bit longer while the other is a bit thicker, effectively making it a one-tooth spinner. Well, if it works at all, didn't put in into CAD yet, and it seem to be a bit too thin or heavy... depending on how much pockets i could put into my side armor^^

Imagine a typical wedge that could drive on both sides, wedge angle about 45°.
now put a weapon motor in the middle of the back on it.
make the sides longer.
put the beater alex has pictured between those sides.
so the distance from axis to mid of the "back" of the original wedge is just a little bit more than the diameter of the beater. The motor would get hit, so make a gap in the beater for it, with the pulley on it.
now with the side walls continued at the 45°, the beater will also almost scraping the ground, no matter what side you drive on.
And if it stops working -> wedge.
Not a good wegde, given the position of the tires and about 60% of the total bot length is weapon ( o.0 ), but well...
Driving-wise it just might behave like Nightmare from BattleBots, but well... i'll be a terrible driver anyway, so better make a slow bot with the main purpose on driving "point the right end at the enemy" and nothing else.

For pulleys i have assumed a widht of 1cm, would that be good?

Also... while sleeping i came to think of:
since i hardly can change dimensions of the weapon without having to change everything (just making it smaller would lift it from the ground too fast, so i'd had to buld a completely different bot), would it be a legitimate idea to make the weapon from some different material (Ali? Titanium?), with a bit smaller diameter and thicker, and only put screws as teeth in there?
Seen screw-teeth some times, but i am not quiet sure about the details and how that worked out. specially how i would try to balance it with screws added, and if placed in a "softer" material the screws won't damage the acctual beater on impact.

[overkill]

I know we have a "be nice" policy around here, but that beater design has some major problems. If its done on the cheap, it probably won't survive even one match and it will cost a small fortune to make a part that will work reliably:

1) As described the beater is all one piece; how is the pulley going to be fitted to the central shaft? Cutting it in half and screwing it back together will add bulk and reduce its strength.

2) Placing the motor close to the central shaft means that the pulleys will have to be fairly small and the belt between them will be short - far too short to fit over the beaters on either side. The shortest belt that will slip over the beaters is about 400mm and that will place the motor too far out. The only solution to that problem is to use custom length urethane V belts that are joined in place. Having used those belts on smaller & lighter beaters, I know they will not transmit the torque needed to spin up this beater design. This problem is the real show stopper - I don't see any good solution without a major design change.

3) Fabricating the part is going to be difficult & expensive. Beaters are usually water-jetted from steel plate, then mounted between centres on a lathe so that the stub axles can be turned - no problem so far. Turning down the central portion from square to round will need a large lathe with at least a 250mm swing over the saddle. Getting a cutter in between the beaters will need over 100mm of tool stick-out which, together with the hardened steel, will create extreme tool chatter and a very poor surface finish. To do this operation successfully will take the sort of multi-axis CNC machinery normally used to make car crankshafts IE: https://youtu.be/81UjjSH2iFw?t=4m

Of course, if the central shaft isn't supported, then it can be left square and the two-part pulley can be made with a square hole .

4) having a central support bearing is a good idea - the beater will very likely bend or snap without it. Two-part bearings are used on the crankshafts of cars and work well when properly lubed and the shaft running in them is well polished. Getting the central shaft polished is also going to need the sort of machinery used on crankshafts. The bearing itself is easy; just cut a plain bronze bearing in half. The mount for the bearing is much harder; it needs to be precisely made, probably requiring a CNC mill.

5) The 80mm space between the beaters is really a bit tight to fit a motor powerful enough to spin up the beater. The closest beater design I can think of is Conker 3 and Alex needs a monster out-runner to spin it up. Once you add a motor mount and a pulley with a hub, the 80mm is more than used up. The pulley width for even small belts is 14mm plus around 8 to 10mm for a hub. The pulley on the beater can't be hard up against the beater arm as the motor pulley needs clearance from the beater as it swings past - add another 2 to 5mm. Then there is a motor mount; it has the support the motor rigidly and should be 10mm thick aluminium. That all adds up to at least 34mm, leaving just 46mm for the motor. Let's not forget some clearance between the back of the motor and the other beater, so around a 40mm long motor is the longest that will fit. Finding a motor with enough power in a 40mm length is basically impossible. I think this beater will need a motor at least 60mm long to spin it up to an effective speed - A scorpion 4530 might be a good starting point.

==============================

I hope this isn't too disheartening and that you can change the design a bit to make it easier and cheaper to build.

[Runsler]

Those are some thoughts i well... didn't hope for (since they are basically saying i can't do this) but still wanted to see since i didn't had (all) of them.

Did some thinking and calculating on the weekend, just didn't have internet to post much about that, so there are slight updates:

currently i think about using titanium, 3cm thick, and everything 3cm wide. Also making it (almost) symmetrical with 24cm diameter, and thinking about adding two bolts per side into it as teeth. Still thinking about some counterweight for those bolts, since not having them right away would make the lathe work easier and less dangerous. but already having the counterweight without them could pose another problem there. This gives me about 5kg for the beater (pulley not included) and hopefully enough strength (will it?^^) at 0,02625 kg*m² MOI.


Okay, gonna try to post more structured.

1. yes, the idea was to make a pulley from two half-circles, with holes to fit a square shaft, bolted together over it. Thought this way i won't need a hub for it, and the bolts could get right through where the belt will go (after all, those are just kinda small holes, hopefully not making the belt wiggle) and possibly bolting it sideways into one of the big parts of the beater, if needed. Wanted to get that part watercut along with the beater. But honestly, i have no idea on how much force there is on that part, so i don't know if that would be sturdy enough.

2. Now i actually don't think getting the belt over the beater would be a big problem, as you can get it over the top part down to the shaft, and only then over the bottom part, making the longest needed the diagonal distance from inner shaft/beater to outer edge of the beater. That's 145mm. Adding some more, since the belt got some thickness, but still it's way less than 400mm So i think you might have just wanted to put it over from one side, and hopefully those described problems won't happen. (Still the part about using the right belts is really interesting. i see every kind of belt used on bots, but seldom someone says which one was better... they all seem to work just good enough or simply get off too often, nothing about not transmitting enough force.

3. Originally wanted to leave the middle part square, avoiding those problems with getting it round.
If i could get it round for a third bushing... i have to ask a friend of mine, he got the lathe and knows how to use it, really can't answer that by now. But polishing won't be a problem, beside the material (it will take a long time) it shouldn't be harder than some stuff i already did.

4. Yeah, getting the mount precisely made will be a big problem. I am used to having to work really precise, but... i am a dental technician, i usually work with stuff the size of teeth. If i could work that precise on this scale... at least i don't have the equipment to do it the way most people would do it. Have to put some more thinking into it if i can find a way, or if not... well, 30mm titanium square are still likely to bend? damned. Even have access to a CNC mill, but that one only can mill small stuff, too -.-
But good thing to know this kind of bushing is used and where it is used, somehow thought if i cut a bearing it won't make one exact circle again afterwards, possibly causing some problems.

5. Regarding the motor, i got one where the description says "1410W". thats maximum output and it didn't have a description for continuous output (or torque, as it is almost always with brushless ones), and 660kv, running on 5s.
So i calculated a bit with what i have, assuming a 3:1 reduction (somehow found that often reappearing in the build diaries)
The run amok calculator gives me a bit over 2300J of energy stored at full speed.
So if Watt is J/s, and even if there is a lot of energy lost on the way, i hoped that would be enough to get to reasonable speed within two seconds.
Also assumed only 1cm for the pulley width, if that is too thin i have to redo that a bit.
Also was going to put the pulley directly onto the bell of it, again not needing a hub (there are threaded holes to do this, after all).
The usual front would be the side mounted, calculated with letting the 2cm shaft just stick through my mounting plate, so i could have 2cm for mounting without extra size.
With the motor itself being about 5,5cm, i could have all that stuff in 7,5cm and leave a 2,5mm clearance on both sides. Not much, but within the 2-5mm you spoke of and making the numbers nice.


I know there is a lot of guessing in it from my side... but well, that's why i am asking here
If i see it right now, my biggest problems (aside from "is there enough weight for the rest of the bot?") will be the supporting of the middle, if that would be possible.
Well.. got half a day without much work for my brain to think about it now.

Anyway... really thanks a lot for all this thinking and those insights. Always better to get told "you can't do that and this is why" instead of just wasting money and time for something that couldn't work.

[overkill]

Titanium isn't a particularly good choice for a beater, its quite soft and the edges of the teeth will round over very fast, making it ineffective. It is possible to add hardened steel teeth to the tips but that is complicated. The main reason not to use titanium is simply cost; a slab that size will be thousands. Titanium will not run with bronze bushings due to galling https://en.wikipedia.org/wiki/Galling it will definitely need ball bearings for support and that rules out a central support bearing.

The belt problem: I drew up a quick sketch based on Alex's CAD and your description - its probably a bit different to what you have but good for discussion. With a 240mm diameter and 10mm high teeth, the shortest distance to fit a belt over is 162mm. With a 20mm thick beater the total path length is 364mm (I originally used a larger diameter and no off-set teeth to get the 400mm belt length).

What belt length is needed? If you went with a V belt, an A sized belt would be plenty strong enough. V belts all have a minimum diameter pulley they will work on and for A sized belts its around 38mm for the motor pulley. with a 3:1 ratio, the beater pulley is 114mm. placing the motor as far out as possible, the distance between shafts is around 95mm. Using an on-line belt length calculator, I get 443mm so you are in the clear. The 14mm pulley width is based on a 10mm wide belt and pulley cheeks that are 2mm wide at the top; you could go thinner but the pulley will be weaker.

Many builders use timing belts, which will work with smaller pulley diameters but I think they will slip too much with this design.

With the motor, maximum power is usually quoted for a 5 to 10 second period. The continuous power for 3 minutes could be anything but a safe bet is half the maximum power. A motor with a KV of 660 and running on 18.5 volts has a no load speed of 12,210 rpm. with a 3:1 reduction that gives you 4,070 rpm but with bearing friction and air resistance you will get closer to 3,000 rpm which is low by today's standards. With the massive weight of the beater it should still pack a fair punch. Based on the motor size that other builders need to get smaller & lighter beaters spinning without their motors catching fire, the motor you have will not be powerful enough and will likely overheat.

[overkill]

Just noticed another issue: Mounting a pulley to the bell of a motor is possible but not recommended for several reasons. The internal motor bearings are quite small and not intended to take side loads that the belt tension will put on them; I have had bearings literally explode from this problem. The belt tension will also strain the mounting screws and the further away the pulley is from the mounts, the more leverage it has. The motor base is relatively soft aluminium and the screws a small, so the threads will easily rip out of the base - again, I have several dead motors with this problem. Finally, the end bell can flex a tiny amount, causing the magnet to work loose. I had that happen to an expensive German motor, not a cheapo Chinese one.

I still can't visualise the shape of your design, but it has room for a 300 x 240mm cylindrical volume to fit the weapon. If you changed to a conventional beater and put the motor on the end, it will fit in the same space and all the tricky engineering problems will go away and costs will come down.

Back to your first post, one of the best fasteners for HDPE is the barrel bolt.

[Runsler]

Damn... didn't realize Titanium would be that complicated (regarding the bearings and stuff) and that expensive...
Also good remarks about the stability of the motor, didn't think about that. Also, i believed maximum power for two secounds would be good enough (to spin up once) and to hold it on a good level much less would be needed, anyway. But if you've seen those things blow of with similar characteristics, i'll take that experience and rethink that...
(Oh, and i know where we were different when talking about the length of the belt, you took total length, i was just looking for one way, since it is symmetrical)

Putting the Motor to one side would be an option, i just like symmetry, and believed with the pulley in the middle, it would be less likely to get cut off if anything attacks from the side. But seems like i would have to let go of that idea, not going to build something THAT expensive as a first try robot when it will most likely not even be really good and the expensive part is so "special" i can't possibly reuse it in a better/later build. quiet sad now, design was planned out so far with 1.6kg left for pulleys, wires and (some of the) bolts, without any pocketing and other weight-reducing stuff. Hoped that would be a reachable thing.

Also, that thing would have stored about 2000J with the last design. Not that much, but if the weight is there i somewhere read a slower weapon with the same energy stored would be stronger than a faster one, since it has more bite.



Anyway, even though i am a bit sad now, that was some really helpful advice, and i am glad you wrote all that stuff.
Regarding the "be nice policy", being nice to me is pointing out the problems with my plans, after all that's why i am asking here.

Well... need something different now. Weapon about the same diameter and (with axis) same length, but motor on one side not the middle, weight limit for the weapon 5kg, this time hopefully not from Titanium. Without the motor on the inside, maybe even a drum would be possible, only i have no idea how drums are made. just milled out from one solid block? Was originally going for a beater since the bot was going to be wide and a beater seemed easier to make... well... easy... yeah... also with a beater i would have had a name for it.^^
Okay, time to pause the weapon stuff for a moment until i did some more research and come up with a new idea and possibly show some scetches this time right away, so you could say what i have done wrong that time.


But for fastening HDPE, i have actually no idea what you mean by "barrel bolt". Google only give me those bolts to lock doors when i search for it, and my dictionary seems to fail me on this one. Do you have a picture of one, or maybe a different name?

[maddox10]

Barrel nut in Dutch is "dwarsmoer."

Try that in google images.

€70 a kilo is a decent price for Titanium. You can expect to find cheaper, but also a lot more expensive.