Just be careful with Chinese Amps, they are more powerful than the metric version ;)
Printable View
Just be careful with Chinese Amps, they are more powerful than the metric version ;)
Yeah, the single channel FeatherTwos can actually handle more juice than a TZ, as 85 China amps translates to about 40 metric amps over 3 minutes.
Thanks for all the information Guys-We are actually running 24 volt CIM wheelchair
motors, the power is via 2 lipo 4s which gives us 32 volts.
I dont really want to change that set up because we are really happy with the
performance so far.
The motor for the jaws I think is also a CIM but a slightly longer one than the
ones used for the drive-it's rated at 30 amp also.
The idea is that one member of the team will have a two button box coming
off the main radio controller for open and close of jaws, I'm thinking the buttons will
be on springs so a quick press on either closes the jaws ( in tests the jaws closed
in less than a second on 16 volts) Im not sure how much amps will be drawn when
the jaws meet thier stop point.-theres no provision to cut the power supply at present.
Also is there a preffered method to get 32 volts down to 24 as most ESC I have
seen have a top voltage of 24 volts.
Another Sunday, another batch of snow and ice!
For the first time in a few week, we both managed to make it to the days work.
Today was spent setting up the link, and mocking up the all the intended layouts for wiring locations and routes.
It took us longer than anticipated to do this, as we struggled to fit things into varying sockets and othersuch. However, we managed to complete it, and after we traced out the intended location of the batteries, ESC, and motors, to work out how much cable we needed for it all.
There's a few things we established from today - the motor that runs the weapon doesn't run on its own fuse, unlike the the traction motors, so this might be modified in later on. We also need to actually build the area where the ESC will sit, and we'll need to finalise a solution for controlling the weapon motor, and whether to stick with the homemade solution, or one off the wonderful internet. We're aware of another upcoming purchase, the hydraulic jack trolley for getting it around will be quite a bit of money. So we may see how the homemade solution performs in a few real world scenarios before we make a final push on that front.
Over the week, Colin is going to try to make a start on building up the area needed to mount the ESC firmly in place, and I'll be paying another visit to Maplin, if its still open, as we need a bit of extra cable and sockets.
Sounds like good progress!
I can't believe we live in an age where the future of Maplin is in doubt. Such terrible times :p
Realistically speaking. I am 100% part of their problem. I know full well everything they sell can be bought from Chinese resellers for about £0.25 to £5. I only ever walk into their shop if I need it in a time I can't guarantee delivery from China :pQuote:
Originally Posted by Theo
I just go in because I can't be bothered with online shopping half the time. 3 days delivery vs A trip to Maplin. Plus they sell Stylophones, bought 6 from them over the years
Another Sunday, this time with 1 hour less sleep than usual. But not to worry, we both still put in a good days work despite that.
Today was another focus on the electrics. However, let's go back in time slightly - over the course of the week. Colin has removed the original container for the much smaller ESC we first planned to use, and has built a much larger one in a different position to contain the rather large Cytron ESC. During the week, I had also sourced some cables which would come in use today...
Dragging back to the slightly less back in time of today, we had a few issues getting some plugs onto some cables.
Unfortunately for some reason, we didn't manage to get the soldering iron (two of them), to solder in any meaningful manner. After a while of faffing around and achieving not much, Colin ended up finding his blowtorch and using that. Over the top, but certainly got the job done in record time. Next week, I'll take the soldering iron I have at home and see if that does any better - as using a blowtorch. Colin had also acquired some junction boxes, this allows us to split the battery power between the main Cytron ESC and the two traction motors, to another as yet to be decided ESC and the weapon motor. We found a nice little spot to put the junction out of harms way, but still relatively easy to access without having to remove it from its new home for maintenance. We also found a spot for the main fuse, which is now lovingly zip tied in place. The circuit starts at the batteries, then immediately into the link, after it goes into the fuse, and then as evidenced on image 1, straight into the junction box between the ESCs and motors - each motor still has its own fuse.
We finished off today by drilling some holes into the new ESC container for all the cables to go through. Because the Cytron ESC doesn't use plugs, and you simply screw the bare wires into it, this was an easy task as we didn't need to account for cable plugs in the hole size. Colin has fashioned a lid for this new ESC container, as well as extra side armor to protect the fairly thin box of the container, which we'll hopefully have done by next week along with the rest of the wiring (image 2)
We are thinking of using a FeatherTwo ESC for the weapon motor. However we're unsure of the best way to reduce the LiPo batteries of 24V (the 'flat' voltage') to 33.6V (fully charged) (2 x 14.8v nominal) to the near 24V required by a FeatherTwo, and carrying an extra battery exclusive for that ESC is not the ideal situation for us. So we may need to scratch that and find a different ESC for the task.
the images
Attachment 7887Attachment 7886
One of the highlights of my week is reading this.
Do you think it'll be ready by Manchester? You may be able to have it on static display if not, I'm sure the audience will love it (beats another boring wedge flipper in the looks department any day). I've not seen Niall mentioned recently, is he still working on it?
Happy to read, thanks.Quote:
Originally Posted by Theo
I'm not sure when that is. However I think we're still a few Sundays away from having something that properly functions. Especially in the weapon side of things. I think we're at a point where the driving will work reliably as soon as it is wired up now. I'm also unsure of Colins availability for such events at the current point in time. I'd be available without issue though, that much I can say. I'm not sure RE static displays, if that's beneficial overall for all involved or not?Quote:
Originally Posted by Theo
Niall is unfortunately no longer part of the project. It's back to Colin & Myself, who are certainly in it for the long haul!Quote:
Originally Posted by Theo
It's sunday!
Myself & Colin got together this Easter Sunday, and focused on electrics again.
During the week, Colin had worked on the lid for the ESC box, it's now got quite a strong lid to it that should easily do nothing more than be a little dented from something like an axe attack.
We started off by creating the cable that'll link the two LiPos together in serial, seeing as that needed two connectors on both sides, and last week we had a few issues getting the soldering going accurately. This week was greatly improved in that regard, using a soldering iron I brought up that's a little more powerful, only by 10W, but it made all the difference. Again, we used a gas torch to help melt the solder deep into the bullet connectors base instead of just around the edges to get a much better and hard wearing connection. We then moved on to all of the other places that'll later need wiring together, adding connectors so we don't have to do these as we come along to them.
We also setup the status LED to wrap up with, it's currently being fed through a seperate set of 3 AA batteries, which will be reduced to two with a small modification to the battery holder next week, as 3 slightly overly stresses the LED. The LED circuit is looped through the main link, to ensure despite its seperate circuit, it'll always accurately reflect the status of the bot being armed or disarmed.
The LED is shone through a hole in the lid of the ESC box, which gives a very vibrant and immediately obvious effect.
Next week, we're hopeful we'll be able to wire it up and have a proper test run. We've still to determine which of us will later go on to drive it, so we may setup a quick assault course and see who can drive it more accurately. We also need to work out the best way of firing the weapon, using a simple extra remote that is standalone with its own receiver, or an extra panel wired into the main transmitter on a long cable.
Todays not many images, next week, we'll make sure we get a proper shot of the lid of the ESC cover. It's shown here in the second image, with the new armed status light brightly shining through it.
Attachment 7920Attachment 7919
A day and half session on the Robot build this week....
Yesterday I decided to change the springs on the castor suspension hinges-
during testing they were too weak and were allowing the Robot to drop too
low to the ground, thier main purpose is to reduce the shock to the bodywork
when 110kg Robot is flipped 4 to 6 ft in the air.
Earlier in the year I changed the clutch on a 4x4 pickup truck and after looking
at the old clutch on Friday, I decided the springs inside were a perfect substitute
for the castor hinge springs.
Sunday-Unfortunatley Adam was unavailable this week due to a Tummy upset
and generally feeling bad so I set too on my own.
Fortunatley Adam had left some items with me last week so I could finish the wiring
off.
I started by re wiring the LED to the removable link, adding a suitable resistor to
the circuit and a new LED, I also reduced the battery to 3 volts and cut down the
battery cradle to 2 cells ( it fits much better into a reserved space ) .
After this I lengthend the motor wires from the fuses and tidied up the layout
with the aid of bolts and zip clips.
I sent a message to Adam saying I might take the Robot for a test run if there
was time but finishing the wiring took while 4 pm so I didn't.
I will try to find time to test the drive through the link for the first time-
next mid week (wheels off the ground) and if this is all good we can take it
to our cicuit first thing next Sunday.
Here's some pics due to internet busy more pics later
Attachment 7964Attachment 7963Attachment 7965Attachment 7966Attachment 7967
Another Sunday has past us by, which means another update!
It's been a day of ups and downs, however we learnt from the experience - and that is what counts!
Before I arrived, Colin had setup the wiring ready for a test. So before we went anywhere specific to test, we plugged everything up, at this point only the batteries were needed, and ran an off-ground test just to ensure nothing silly would happen. It wouldn't try running away when the link was plugged in, unfortunately this did lead to one problem being spotted immediately. The motors had been wired in correctly, leaving one going forward whilst the other reversed. This is due to a simple forgetful mistake, in that in the past the motors were mislabelled red/black, however that was accounted for at some point in the past and fixed - but had forgot we did that so the motor was purposefully wired wrong under the assumption that was to do still.
Not to worry, the Cytron ESC is very quick to do a rewire and it was fixed within a few minutes.
Another issue shot up when we got it on the floor, now we were happy it'd maneouvre as anticipated motor wise.
The castors springs needed a little adjustment, as it ran one side of the bot tended to be on the ground a little firmer than the other, causing some one sided wheelspin. So a little tinkering was made there, to balance out the weight distribution a bit more.
After this, it was time to do a proper test for the first time, we packed up the bot and some tools, and headed for a secluded section of tarmaced area near the workshop. After noting the batteries at this point where about 60-70% charged, and we managed to get a good few minutes of usage out of it. Certainly more than the typical battle would last. We found out quite quickly that control was a little sketchy - unless you had perfectly positioned the controllers stick, there was a noticeable difference in output in the two motors, and driving was a little harder than we'd like. There was also a few further issues with the weight distribution, we'll try to see if we can correct this with more tweaking to the springs as we were still wheel spinning a bit on one side more than the other, although this may be related to the control issue. After a while, unfortunately but obviously, the batteries became a little drained to keep things moving along at a pace of value. We went to check up the bot - the batteries were both very warm, overly so. thinking nothing of it, we tried to have one last run straight line run to gather some speed up, however it just sat and didn't move. So we packed up and headed back to the shop.
Upon arrival, we decided to try to get around some control issues, by setting up a 'tank' style driving. Lifting the bots wheels off the floor, the batteries still had enough power to run the wheels, so we could test this new control method, and it seems to work. However, that is indeed all she wrote. With the batteries at this point being unable to spin the wheels, even off the ground.
We got out the charger, and set to work charging the now drained batteries for a later date. However, it's at this point we unfortunately found out that one of the batteries had been killed, my best guess is by the heat. It was so broken that the charger wasn't even able to detect it was a four cell battery, and only gave a total voltage (of 5 volts, on a 14.8v nominal LiPo!) and was unable to display individual cell data. We double checked and connected up to the other battery, and sure enough this one was seemingly in working order, with the charger correctly giving data for all four of the individual cells, and happily charging upwards. Whereas on the other, the charger just flat out refused to charge it due to a voltage discrepency.
I am thinking that the current battery holders are simply not getting any airflow, allowing the heat generated during discharge of the batteries to build up to damaging levels. I've included a picture of our battery holders below, so if anyone has an idea about this, and can confirm my suspicions, i'd be grateful. As if it wasn't damaged due to the heat, it may be a more severe problem than originally thought with the wiring or some other electrical part of the bot. We have a few ideas for alternative battery holders with much more airflow potential up our sleeves, if the heat and no airflow is the problem.
So all in all, today was certainly interesting. But we learnt from the experience, made some fixes already, the rest we will tackle on further sundays, here we don't hide the stumbling blocks. We share the entire experience for sure!
Some images, and some video taken during the test run. We managed to get up some good speed in parts, and we also know it spins well, so that's something.
I've cut down the video so the first minute is the most relevant to capabilities, and after that its mostly farcical stuff and spinning. Whether out of control, or intended - i don't know! It does however look like it spins with enough force to be somewhat damaging just from that!
https://www.youtube.com/watch?v=Bt3tS7rwIMk
All wired up and ready to go
Attachment 7986
All packed up for its first time leaving the workshop - note that we decided to put the lid on top of the ESC box for this drive out, incase things got a little bumpy as the ESC is loose in the box currently although the sides and bottom of the case inside have been isolated with non conductive tape, just incase. We took a brush and a shovel so we could clear debris from the area, to help traction and make sure we didn't run into anything too nasty.
Attachment 7987
These are the battery compartments. I suspect these are the cause of the overheating and battery failures, as there's very little airflow, especially to cool down the cells in the centre of the pack. Especially as the battery is 45C at 4500mAh so should be capable of over 200A sustained flow, and we shouldn't be touching even 100 of that.
Attachment 7985
The first battle damage! - It was also running on the bar on top, based upon the scratch marks. So we're not sure if it was just bouncing upon it occasionally, or actually dragging it across the floor all the time.
Attachment 7984
A few problems here to iron out :cry:
What looks like bad driving was due to the transmitter not been calibrated
on the trim for one-when I pressed forward it was wanting to go left or right.
secondly-there is still a lack of traction for the wheels, resulting in wheel spin
and wasted energy-the castors may need to be just under the shovel or jaws
when fitted.
The batteries were not fully charged and so combined with the other problems
was not showing the true potential of the Robot.
We need to buy a new lipo battery due to it overheating-any heavyweight
Roboteers with advice on this is welcome ie what run time you should get
and what is the lowest you would let your cell voltage/overhal voltage go to
and what methods you might use to prevent the voltage from going to
dangerously low levels when robot is in use.
Most ESCs these days have a LiPo function that automatically cut power when the voltage level is too low. The Cytron definitely has the facility to detect undervoltage (< 3.0v per cell) (https://www.robotshop.com/media/file...sersManual.pdf - Section 7b) but it doesn't cut it out, only on overvoltage apparently.Quote:
Roboteers with advice on this is welcome ie what run time you should get and what is the lowest you would let your cell voltage/overhal voltage go toand what methods you might use to prevent the voltage from going to dangerously low levels when robot is in use.
This had me panicked a little as section 7.8.2 of the rules (specifically pertaining to LiPo batteries) state the following:
You'll note that it is an advisory at the moment (as of the 2017 ruleset) so with this ESC and battery chemistry, you'd be OK but the only option then is to fit a secondary meter/alarm into the balance plug (https://hobbyking.com/en_us/hobbykin...arm-2s-6s.html) for each battery pack.Quote:
7.8.2
Voltage Cut-out (Advisory)
The robot maybe fitted with an under voltage cut-out or alarm set at or higher than the battery
manufacturer’s recommendation to prevent the batteries from becoming damaged by over-discharge
That's a neat little thing. I've just picked one up, although it gets shipped from the Hong Kong depot so takes a while to arrive :(Quote:
Originally Posted by Ocracoke
I wonder if itd' be possible to modify the Cytron to cut power in the event of a battery approaching its absolute minimum., if its already low voltage aware, it might not be too difficult.
You know what day it is, because I do!
So today, another interesting day.
We started by charging the old battery, which had been tested to ensure it wasn't damaged from last weeks test, and the new battery, to an equal amount that should give us plenty of test time later on.
Whilst these charged, we set off on rectifying a few issues we noted last week;
We started out by applying some additional castors further up the underneath of the shovel, in the hopes that it'd give a little extra weight to the rear to fix some of the driving issues, notable lack of grip and lots of wheel spin.
I spent a while going through the manual for the transmitter - trying to find a way to correct some control issues we had last week, and improve upon the new 'tank' style driving. However, to no avail. We had a bit of better luck fiddling with the cables, and plugging the wires into different ports on the receiver. However, in the end we settled for going back to using what we had before. This is due to us finding evidence of one of the bolts used to hold some polycarbonate armor in place had been scraping on the floor of the last test - it was scuffed up and worn. We later figured out this may have contributed to some control issues last test, as the bot did not entirely lay flat on the floor - one wheel was always slightly above an ideal level for drive and was easily spinnable alone due to an extremely poor grip on the floor. So the excess thread on the bolt was cut, and the bot laid back on the floor - and this time both wheels made sufficient contact with the floor that neither could be spun alone without the entire bot moving.
With that, we were once again on our way to the designated test area.
However, this test lasted not long at all - due to an unfortunate collision with a pile of... varying items of waste.
After the bot was pulled out, we couldn't get it working again - our initial thought was that the wheel was spinning on its entral pole - due to it not being worn down to a flat enough surface for the wheel to grip on after coming slightly lose. We didn't have the correct tools to fix this out in the field, so we put it back on the truck and got back to the shop. Here it became quite quickly obvious the failure was something a little more sinister. In fact the weld joint on the sprocket to the pole had entirely sheered off - not sure if it was a poor weld joint, or if there was much more force than intended occuring here. However the sprocket is off an ancient Austin A40 car, and spent nearly all its life covered in grease and oil, so this may have caused the welding to not take as well as could be hoped. Either way, Colin is going to aim to correct it in time for next Sunday. When hopefully we'll be able to test it without any issues cropping up!
An image, showing the weld joint and its fairly easy to see cracking around the joint
Attachment 8004
A Saturday morning build session this week...
Half way through the build I decided that the rear of the Robot looked empty
and decided that there was room to fit a fail and housing as described previousley.
After consideration the flail idea was dropped and so the housing was not required
so today I removed the over hanging housing.
The new idea is to have the wheel exposed at the back so we can literally climb up
walls with a 280 degrees of wheel able to touch the surface of the arena overall.
After cutting the housing off and some refabricating I also re cut the tread on the
tyres to hopefully give more grip-this is a very messy job and everything gets covered
in rubber dust.
Attachment 8027Attachment 8028
when do think it going to be combat ready ? as it damn near close
I think we are nearly there but we don't want to rush things to make any
specific event.
Originally we built the Robot for Robot Wars tv show which was of course cancelled
so I think Adam would agree that we would rather have a tried and tested machine
and get as much driving experience as possible rather than take it to the first event we see
after fitting the last nuts and bolts.
It's that day again! It's sunday, which means when I got home, there was a load of takeaway menus waiting for me in the post box.
Oh right, the bot, yes!
Today was a day of success, after a few recent tests being let downs. It wasn't without a few issues, which will be revealed later on. However, it was by far the most successful test we've had for a while.
Today, we started by charging both of the batteries up to 16.5v. Whilst that was happening, Colin did a little show & tell and demonstration of the strength of some metals, and how large but compact a Gazebo can be - this is actually not for anything to do with this new machine, but it was a bit of an idea fest for the Northern / Scotland events I'm trying to get up and running. This only took 15-20 minutes, and as we have figured out the safe limits our batteries can charge up to, it's all it took to get the batteries, both of them, up to the charge we were looking for.
Soon enough, the batteries were ready to go, and we put everything in the truck and to the usual test area. Quickly putting up a few boundaries to stop us from going in some very unpleasant pools of disgusting looking stuff.
Battery wise, there were still some concerns however. To begin with, the batteries depleted at even rates. However towards the end, one battery started tailing off a lot quicker than the other. Going from 16.5v each, to one being 13.2v, and the other 14.5v - up until the 15v mark, they were near enough depleting at an equal rate - with us frequently pausing to test the voltage and heat. When we got home, it was discovered that one of the cells in the 13.2v had become quite unbalanced with the rest, whereas the 14.5v battery was still all within 0.02vs of all other cells internally.
However, and - as pictured below, the lower 13.2v cell was quite a mixture. With the cells being 3.18, 3.30, 3.25 & 3.35. However, once charging had started, it only took a few minutes for the charger to balance everything back out before it charges them all up again, still in sync with each other.
This gives me and Colin some concern, as we are unsure as to why the cell are discharging at differing rates, and why the one discharging faster, is also becoming out of sync voltage wise with the independent cells. I therefore hypothesize this is the cause of the previous battery fault, where one discharged beyond its minimum without us noticing, and the other not doing so, resulting in a dead battery, and a bit of an unhappy hot one. It's worth noting here, that the new one is the one that ended up being the most discharged with uneven cell voltages, so it isn't the old one that is damaged here. So if you have any idea why this may be happening, i'd very greatly appreciate any feedback or suggestions. As the batteries were all charged to the same total voltage, and all cells were balance charged - so this all happened during discharge?
Colin notes that the drive is still a little off. Driving in a straight line is still quite hard to achieve. As hes been doing all the driving during the testing phase, I'll leave him to note down the specifics in a post soon after this, and hopefully someone can aid us in that area too.
Here is the video from todays test. Here we are able to demonstrate the speed it can attain when going straight(ish) - and how it manages to always get itself free from any debris its stuck over, including working its way out from awkward wedged positions. Furthermore, Team Death finally manages to beat a wall. Outdoing the infamous Piece de Resistance, now two decades ago.
Overall, the bot is certainly starting to behave in a manner we are aiming for, with a few little niggles - but we are definitely getting there, and the to do list after each test is getting fewer and fewer.
Note: you may wish to turn audio down - the wind was being unhelpful, and my phone has quite a high gain mic on it, so the wind is rather irritatingly loud.
https://youtu.be/weuArh5PyAw
and images
Note: on the cell voltage - the two last cells are non existant, as its a 4S battery, so we haven't lost two cells here! - here demonstrates that faster discharging batteries odd balance after the test (the other was higher overall voltage, and still balanced)
Attachment 8031
Not content with knocking the wall down, we also managed to push the bricks some distance too. At some point, we may line up some bricks and see how much push we have on them - we also may try to pull or push something like a pallet.
Attachment 8030
There is however, one downside to Colins modified wheels. The new treads seem to be ripping off slowly over time, this may or may not cause an issue later down the line.
Attachment 8032
err what batteries are you using ?
They're Zippy LiPo, 14.8V (nominal) 4500mAh, 4S @ 45C
(or specifically; https://hobbyking.com/en_us/zippy-fl...___store=en_us)
did you say you are running 2 in series or parallel ?
Side note: I would check them again as the cells can recover some voltage, and depending how far out they are, run a balance charge on them as that should bring it back in line
They are in series, as in parrelel they wouldn't amount to the higher voltage, just double the mAh
So in the configuration they are at, they are 14.8 + 14.8v (29.6v nominal), and 4500mAh (mAh doesn't get doubled in a series setup)
i am aware, 4.5ah does seam a tad low as i know coyote ran a 8ah 6s battery
I wouldn't have thought that would cause the batteries to discharge at differing rates, with one of them getting very poor balancing between cells over the other?
It could be as when i have completely drained my batties that when they have been the most out of sync.
It could also be the wonderful chinese hit and miss quality
I would have said 4.2aH was too low. Tango used a set of 2.1aH 6S Zippy (2x 3S in series) LiFe batteries from The Honey Badger and they were thoroughly dead by the end (all puffed up and such with the cells well out of sync). I would have mooted at least 6aH in that case and I'll possibly be looking for 8.4aH for Kaizen.
That is possibly down to where it is being tested. I would have thought laying down some plywood sheet on the ground may help.Quote:
There is however, one downside to Colins modified wheels. The new treads seem to be ripping off slowly over time, this may or may not cause an issue later down the line.
Colin has suggested using larger capacity batteries already, so we may have to go down that route, which is a little unfortunate as we are already a little over the expected cost of this build.
Yeah, a proper arena will be a lot more smooth and not that rough on the wheels, one would hope. We do our testing a short drive from the workshop, as its possibly the furthest away from things of concern getting damaged. So i'm not sure how viable it is to take something up there to be softer on the wheels.Quote:
Originally Posted by Ocracoke
if you got a halfords near you see if you can acquire a load of old bike tyre as that work great, really saved my butt at manchester
Apart from the batteries running down rather quickly and the wheel spin...
I noticed when driving the Robot I had to keep adding a little more power
to the left hand side of the Bot (as viewed from the rear) via the controller
ie :slightly turning right to keep the Robot on a straight track.
The gearing for the motors is identical and so are the motors and wheels,
I'm thinking the right hand side of the Robot could be getting a little bit more
traction than the left side- retro fitting bike tyres or similar could resolve this, so
it might be better to remedy this before altering any trim on the controller.
We may turn our attention to the weapon this weekend while we ponder
the best way to improve the traction and battery longivity
The other thing you could do is adjust the trims a little, I found I had to do this on my featherweight as it had a tendency to veer off when going straight (again, same motors, same gearboxes etc).
I've just remembered, i have a spare transmitter and receiver pair here at my flat. I could bring that along on Sunday, and quickerr replace the receiver in the bot and see if it has the same drive issues. If it does, it would point to either something in the ESC, a motor behaving not quite identically as the other, or a few other things. But it'd certainly eliminate the transmitter/receiver if this works fine - and we'll have a pinpoint to help wittle down the issues in those two if we need to stick to the transmitter/receiver pair we originally used.
The one I have is a fairly basic one, buts it has six channels - it was only ever intended for featherweight and below stuff i was doing at home at the time - https://www.ebay.co.uk/itm/381958044606
Thats great Adam-bring it along and we might be able to determine if the steering
problem is in the radio signal or like I suspect one wheel getting more traction than
the other.
The main issues I see are the castors and the wheels. The wheels are very large, close together and right at the rear of the robot. The robot is also quite long and thin. When large wheels are as close together as they are here, it's always going to be difficult to get the robot to go in a straight line; it's much more likely to have a tendency to veer off to one side (which side will come down to the dragging issue Colin mentioned) compared to a robot that has a wider wheelbase. Having a lot of the robot's weight towards the front of the machine and less over the wheels also means that it can possibly pull to one side depending on weight distribution etc.Quote:
Colin notes that the drive is still a little off. Driving in a straight line is still quite hard to achieve. As hes been doing all the driving during the testing phase, I'll leave him to note down the specifics in a post soon after this, and hopefully someone can aid us in that area too.
The issue with the castors is that they reduce the rolling friction of the robot. In principle this seems good as the motors will technically have an easier job, but it does mean that when the powered wheels stop turning, especially when turning/spinning, the momentum built up will continue to rotate the robot, giving an oversteer effect. Most robots these days do away with castors and run with their leading edge just scraping along the floor. This dampens the driving and associated effects such as veering and oversteering. It means the motors will be drawing more current to move the robot, but I doubt the difference will be significant.
Obviously there's nothing to be done about the wheel width, it's pretty much set in place. You could try experimenting without castors if the robot's leading edge wouldn't hinder this (or try replacing castors with skid plates made of, say, Nylon) and see if that improves matters. The other option to consider would be the use of a gyro, which monitors the robot's movement along a certain axis and corrects any deviations. Never used a gyro before personally, but they can be a bit finicky to set up and use correctly, so you may not want to consider that option yet.
Either way, good to see the end is in sight. It's looking very grungy and agricultural, look forward to seeing it in an arena.
Thanks for your input. We may try that next week. The castors aren't too hard to get on and off for quick testing without, so we should be able to do a before and after on that relatively quickly. 'Grungy and agricultural' has ended up looking quite good, and its definitely going to be painted in a manner that keeps it looking that way, when it comes to the final steps and paint jobs!Quote:
Originally Posted by RogueTwoRobots
Today we had another build day, although it was mostly testing and talking.
We got to work charging the batteries, although we knew the testing today was going to be brief so we didn't go for a full charge.
After chatting and having a bit of a laugh for an hour whilst they charged, we went up to the testing place, this time with the receiver and transmitter I had brought up from Wakey. The drive seemed a little better, but not perfect still. It was my first drive, so I couldn't really determine the difference, however when Colin tried he said it was running a little better. So perhaps the trims are off on the other transmitter - it's also possible it may be a technical fault, as its a second hand controller, whereas mine is a new one.
As noted, the drive is still a little off - however it is now at a point where it can be adjusted on the fly with quick fingers to bring it into alignment. Next week, we're going to try to see if there is a reason why one wheel is getting more friction than the other, perhaps the build is slightly off angle, we'll also try it without castors - as we'd already considered that. Although unfortunately we don't have a nice surface to run on, so we'll be all scratched up if we test it without castors and it runs on the floor too much.
At some point, we'll need to decide what transmitter/receiver to use.
Today was my first go at driving it, and I have to say. I'm surprised at how nippy it can be, due to the placement of the two wheels, it can turn rather quickly, to the point where you'd certainly not want to get in the way when its turning at max throttle, it really throws the front end around in a spin - may end up being a useful tool in a fight!
As a sidenote, today we discussed a few potential things to do after the HW was completed - and I think if we do get to do them, a lot of people may be pleasantly surprised by what we may have up our sleeves! We've also discussed what
During the week, Colin is going to try re-assembling the jaws ready to test that next Sunday. So hopefully by the end of next week, we'll be a step closer to battle ready - all we need is an event not too far away to run it!
The bot in the sun, just before testing. Was fabulous weather for it!
Attachment 8058
I think one wheel's slightly higher off the ground, you may want to check the shaft it's on. I'd test it first without the castors and work from there. Apologies if this is in a massive font, first time posting from a Blackberry.
Ups and downs on the Robot build this weekend-Yesterday I charged the batteries
to full capacity for a good test run today, I also re-fitted the jaws to see how
the Robot handled with them in place.
Adam arrived at 12 today as usual and we put them speed controller back in
the Robot and set off for the test drive area.
Adam had brought with him some battery monitoring devices so we fitted
them before we began to drive.
Almost straight away the Robot felt wrong, not helped by the fact we had wired
the speed controller motor connections the wrong way round.
It turned out the weld had broke on the reducer sprocket so we had to make a
quick retreat back to the workshop to weld it.
Upon our return we had a little more luck and got the Robot to drive with the
jaws in place for the first time, the skewing of the steering was slightly worse
than last week and I thought it would be because of the extra weight at the
front and we are still using the castors for the moment which we may change
to a skid.
Unfortunately after a couple of minutes a bungee strap we had been using to
hold down the the e E S C got wrapped round the wheel and one chain came off.
Apart from the niggling problems we are both pleased with the Robot and we
can finally reveal the Robots name...........
THE
MEAN
MACHINE
I've got the images from today;
Today is the first time we've had the jaws in, with the additional ESC box on top. So this is the current overall look of the machine, there may be a few changes, mostly to the wheel areas, but I think this is near to the final look of it all, minus the final paintjob to do at the end - which will maintain the red/black theme.
Attachment 8087
This is from the first thing to go wrong. Shortly after arriving to start with, the welding joint on one of the smaller sprockets went. It's worth noting, that this is probably due to us forgetting to redo some of the original temporary welds. We've now gone above and beyond what they originallly designed to do, and we're getting these issues. It's much better for them to show up now, than at an event.
Attachment 8086
One of the first signs of going wrong on the second trip. This is the bungee cord to secure the ESC in place in its temporary testing place - as it isn't secured in there by anything other than the wires. Unfortunately we forgot to secure the securing bungee, and it quickly ended up tangled and shredded, probably due to it coming into contact with some of the sprockets and chain? Whatever precisely occured inside, a chain was found to have come off when we got the bungee out.
Attachment 8085
Next week, we've already got a to-do list from todays events. We're going to have a go at the wheels next week to resolve a few further issues, although exactly what that'll entail and end up being, is unknown until we come to do it.