Watkykjy and Robert's 2DOF wiper motor simulators

Hi Guys, here we will keep the data and progress reports of our simulators. So far we have managed to build the circuits that TonicGr posted, and they are both working properly. We have also constructed the base of the frame, and today we will start bolting on the support structures for the motors, screen, keyboard, etc etc.

Here we have the frames built and the seats fitted to get a feeling for what they might look like when complete:



Then it was over to sanding, filling up and some more sanding as we prepared them to be spray painted. Finally that was done, and we managed to get the primer on.

After drying we started with the color. I went for a bright orange for the inner frame, while Robert (due to the paint running out) did his entire frame in black.




Next we will start with the motor mountings, screen mountings, and the rest of the stuff to finish them off, but the end is in sight

Next update soon!

Looking great! I really like the orange/black combo!

Good thing you make a thread about your sim, Hope for a lot of details about the setup with wipermotors. good luck! :thbup:

Hi Guys,

Thanks a lot, we have really been burning the midnight oil with them.. So, regarding the wiper motors.. Well, they are harder to work with than one can imagine... Firstly, we had to build the supports for them and the fan belts that will pull the frames up and down in their respective direction. This turned out to be a huge task, as the amount of travel one desires, as well as the placement of theses supports has to be thumb sucked. Luckily it all turned out well for us, and we were able to find a spot in the front and a spot on the back of the frame that will give us ample travel and also mount nicely.

Now, here is something BIG you need to watch out for when you are building a frame out of steel... The housing of the wiper motors are connected to ground, so when you mount your motors, you have to ensure that they are isolated, because when the H-bridge changes the polarity on any motor, guess what's gonna happen.. You are gonna go flying out of your sim, that's what's gonna happen!

So, tomorrow I will start playing with the electronics (if my wife lets me... Tomorrow is my baby girl's birthday ), and I will try to find ways to properly isolate the motors without taking away structural strength. I will upload a few pictures of the supports to far and go a little into the detail on how we built them, etc. We opted for lightweight aluminum, simply because it's easy to work with, and it's also very light, providing (hopefully) enough strength for the system too.

Anyway, that's about it for today, I will take some pics tomorrow morning and send post them...

Have a good one guys!

Johan

For insulation use a harder type of rubber or PVC 2-3mm thick. Great insulators, cheap at any hardware store, easy to cut to any shape and it will help reduce vibration from the motors.

And of course, ALWAYS check the electrical connection of your electronics! Especially the motors. You could take it apart and see if you can disconnect the housing from the (-) and ground it with the frame.

Be VERY VERY careful not to short anything while you are building! Grounding your frame will protect you, but if something goes wrong you will probably shoot your electronics. Although you are not working with high voltages, the motors will generate enough to get you out of the sim

And don't even think about playing with electronics on your doughters birthday. All this stuff will wait an other day, but her birthday is just once a year. Spend the time with the family because it is a lot more important than a stupid piece of electronics.

Watkykjy and Robert your project is coming along very nicely, good inspiration for me, I can't wait to see the new update photos :thbup:

Hi!...

Congratulations... seems to be a very clean project. :hi: and welcome to the joyrider's world

I've been looking your frames, and although they are made with metal instead of pvc, your designs looks different from the original joyrider.

I'm talking about the reinforcements placed at the base frame. Have you think about the vertical forces? Look at this img to show you what i mean (i've taken an old picture of my joyrider):


I think those reinforcements were designed for distribute the forces along the tubes... note that the base frame will sustain all the weigth of the entire joyrider (plus lcd, steering wheel) and your own body as well... that's would be around 90-100kg. Add the Rotational inertia and you would end with more than the force your vertical tubes will handle.
If you have calculated that, and you know it won't blend, perfect, i will keep my mouth shut :tape:

Regards :
knob2001

Although steal is much stronger than PVC pipes, the welds can easily break or bend. It will strengthen the frame for longitudinal forces, and I also recommend it.

Hi Guys,

Thank you very much for the comments and suggestions! We have tested the frames for the inertia and weight that we would potentially handle, and taking into account the fact that the tubes are 3mm thick, and we welded everything almost all the way, I am quite confident that extra strength is not going to be required. However, only time will tell. We twisted, jumped on and pulled them as hard as we could, and so far it all seems to hold up very nicely, but once the motors are doing the work, we will have to see what happens. As far as the motors go, I managed to disconnect the ground from the casing, and now I can turn each motor individually with it's own floating ground, so I am not stressed about that any more. The last part (and this seems to be a problem), is how to mount the 10 turn pots to make sure that they turn the full 10 turns when the pulley does only 1 turn. If I take a few close ups of our supporting frames, do you think you can throw some suggestions my way?

I had a look at how Tommy did his, extending the shaft from themotor to outside the casing, and then using lego gears to turn the pot, and that seems to be good, but I am just wondering if there isn't an easier way to do it?

Let me quickly run to take a few pics of all of this...

If you buy 10 turn pots, make sure they are high quality industrial pots that turn very easily. Cheap pots will wear out gradually and your simulator will act glitchy at certain spots, and it will get worse and worse. Better to buy an expensive one and change it every once in a while, than buy cheap ones and change them every 2 months.

About the mounting. I would personally find a place without gearing to give you approx 5-10 turns. Alternatively, you can measure rotating points such as motor shaft, motor output gearing, or the pulleys, and see how many times they turn for an endpoint-to-endpoint movement. Then you can go out and buy a pot with the required turns. If the motor gear output turns 23 turns, you can easily use a 30turn pot. If you don't use the full range it is not going to be a huge deal. If the AMC uses 10bit resolution, you will get 512 steps for half turns, and 1024 for all of the turns.

Hi...
Glad to see you have tested the weigth first... although i'm still thinking placing reinforcements is a healthy habit

About the pots... do you know we are using hall sensors instead?
Look for SS495A. They cost around 5€ each and with the magnet (from an old HD) you can do it ever better than the pots. In my topic you can see pictures of it, and of course, in tronic's original topic there's info too. Mounting the sensors is something easier than the pots, becouse each sensor is just 2cm! :clap:

Anyway, the 10k trimmer is there to fix the pot's turns. Put it to 0 on one movement limit and then turn the frame until reach the next limit of the movement. Then, looking the data on the lcd, turn the trimmer up to 1024 without moving the frame from this limit. With that you will have 0 on one limit and 1024 on the other, no matter the turns it has to do to complete the angle. At least is what i did when i began with the pots. :? In my opinion, the 10bit resolution is far enought to play games (even less than that).

There's a firmware to calibrate the pots that you can use prior to use the normal firmware... i'm using it to test my suspicious 8535.

Regards :
knob2001

Thank so much for all the info guys, it is EXTREMELY valuable to us, albeit a few hours late

We just finished the design for our pot mountings today .
Not to worry though, I will have a look at how the hall sensor was used, and start converting them to hall sensors immediately.. I have worked with hall sensors before, so that shouldn't be too much of a problem, and if you say that they provide better resolution (and no moving parts), then this is surely the next upgrade for us. Coming tomorrow we will probably finish the frames and start the testing (with our fingers close the power cords for safety

2 things I wanted to ask for help while we are on the subject... Firstly, what kind of safety do you guys use, in the event of a hall sensor or pot failing? Thanos originally suggested using micro switches with diodes to cut the power to the motors. Is that still the way to go?

Secondly, at the moment we are using 245W old AT PC power supplies to power each motor, being put through a 14000uf 50vdc cap (to handle the spikes), and it's working well, but when the motor is turning, I can, with a little difficulty stop the motor completely if I grip it by the pulley... Is this normal, and is this because I am using those small power supplies? Do you think the problem will go away when I use a car battery?

Thanks again for all the information, I really appreciate your comments, concerns and suggestions! The moment we have everything back together in one piece I will start uploading the next batch of pictures!

Keep 'em coming

Johan

For safety the limit switches will be your best choice for now. Ideally the controller should take care of it, but the AMC can't do it. The downside to this is having 4 huge switches with 4 large diodes to carry all that current for the motors. With a controller designed with this in mind, you would need 2 small switches that carries no current. Once I finish mine, you can have all this

Hall sensors don't have better resolution that pots, in fact, what you use have no change in resolution, only in the movement range. The controller have a fixed Analog-to-Digital converter resolution, which is 10bits. That means Vref (5V) or whatever you set it to with the adjusting pot will be the top 1023 value, 0V will be 0, 2.5V will be 511 and so on. You can however use this voltage swing 0-5V differently with pots and hall sensors is terms of mechanics. That's where one might have an adantage over the other depending on the application.

You will need to use car batteries IMO, no matter what. A 140000uF capacitor seems to be huge, but electrically it really is TINY. I don't want to get into the calculations of how long it will supply you a given amount of current, but it will be in the milliseconds range at 20+ amps. All it will do is filter out the spikes, but when you load the motor, it is not a spike. You need real current.

Hi Guys,

Update time again. We moved our simulators to our different homes, so that we can be a little closer to our families while finishing the work we need. I have started the testing on the electronics, and Robert has been finishing the mechanical work on the frames for the screen mountings, etc.

We also redid our motor mountings, because we needed to include the feedback pots (which will not be redone AGAIN for the third time). And we will also do it a fourth time when change from using pots to using hall sensors, but that's in the time to come...

Anyway, on to some pictures:



Here are the four completed supports for pulling the frames up and down. Notice the new and improved motor mountings, as well as the extra strength added with the enclosing of the angle iron inside the aluminum ducting.




This is what Roberts frame looks like now. He finished the screen mounting and also mounted his supports firmly.



And here is a pic of my sim so far. Yesterday I used some old fan belt lying around (got the right one this morning), and connected the frame to the motor quickly for testing. It turned out that moving up was WAAYY faster than moving down, and my AMC got EXTREMELY hot while doing it through there. So there is some work to be done

Some of next TO DO's include:

Finishing the pot mountings and making sure they read back to the AMC properly.
Adding safety switches for maximum movement prevention.
Determining minimum and maximum movement values and positions for the frames.
Adjusting the wiper motor speed to be the same in both directions.
Finishing the support mountings and making sure they can't bend or twist.
Finalizing the screen and steering wheel supports, etc.
Connecting everything up to a computer with X-sim and testing on that.

Thanks guys for all your help and suggestions along the way. If there are any pitfalls we need to look at on the current to do list, please feel free to let us know!

Cherio

Johan

Hi watkykjy!

You are building awesome simulator!! I like it a lot! I decided to make base for my simulator of a metal too!
I see you changed proportion... Your sim is wider, am I right?
May be you could share the plans of your sim with us? It looks pretty balanced!!

Best regards,
Constantine

Hello. I have a question about your the setup for Pulling support, and I hope I am able to explain it properly. As I see in the pictures the belt will go around two pulleys, and one of them connected to the motor. I tried something similar, but the problem is that since the frame move aroun an axis, there will be slack on the wire or belt while moving in each direction, because the distance between the Wire/belt and the frame is not the same all the way. Have you tested the system with the frame?

I also will have the problem with the vipermotor having two different speed. For pitc-axis, I will mount it so that forward have the fastest speed, as breaking gives more and faster g-forces than acceleration. For roll-axis I would like to have same speed, so have you found a way too overcome this?

fredspeed

As to wipermotor speed take a look here: http://www.youtube.com/watch?v=XZHyXPuMxFo

and about lack on the belt:
Try to freeze frame in the postition where the slack of the wire is biggest, than pull the wire so the slack is gone... the wire should be well streched

I did try that, but then the belt will be stretced in both direction and some torque is lost doing that. Ofcourse moving the pulley-system a greater distance from axis point might make this effect smaller. Hope Watkykjy and Robert post how this is working in their setup!

Fred

I had the same issues by using steel wire... Had to tension it again every a few hours of usage. Problem solved permanently by using BUS engine belts!!!! No slipping, no problem with losing torque near end limits anymore!

updates??? no video no nothing else!!!2 years!!