K8055 - Dual Wiper Motor Control with Variable Speed

simplified actuator design... fail

I have been working on different actutor designs that are simpler and lower profile than the bungee supported ones I tested earlier in this thread... those ones are great, but not really practical for a home build unless you are lucky enough to have a well equipped workshop!

I tried the following design in a number of different configurations with an old wiper motor that had a damaged nylon screw inside, so it was no good for normal use...



The results were interesting - it is cheap, very fast and very very simple to make compared to the other type of actuator...
Unfortunately, it is also weak under load - the motor does not have enough power to spin up from a standing start with a heavy load so it is no use for my intended application - With a more powerful motor perhaps... but not with a simple wiper motor!

There is one feature in this design that worked very well though, and that I would like to share (as I haven't seen it anywhere else):

The screwthread moves an actuator forwards and backwards... this is supported on two bearings from a skateboard to support large loads and remove friction from the forwards and backwards movement.
Two more skateboard bearings are used in the uprights that support the threaded bar, and are included to reduce friction from the rotational movement.

This worked very well



The motor itself required very little in the way of manipulation to adapt it to this kind of application - I just popped the top off, pulled the damaged nylon screw out and spot welded (very badly!) a piece of 8mm internal diameter pipe from the worm gear on the motor spindle to a piece of 8mm threaded bar.

Conveniently enough, the diameter of the spindle thread in the wiper motor was 8mm (but had a different pitch)
I am tempted to test this setup to destruction to see if an acceptable power can be achieved by overloading the motor - maybe with 15V like Deriy... As I mentioned before... the wiper motor is little use for a standard application.



Back to the drawing board with this concept for now

Tim

Interesting, Im test on paper at the moment to include the sarrus link into my boat winch design.
My winch can already std life about 180kg with minimal load on motor but travels in a rotational movement, so by adding this type of link setup, I hope to achieve a vertical movement of about 200-300mm with increased force and speed.



Now time to work the levers out,drawn 1:1 in autocad this shouldnt take long.
Possibly can be done with a wiper too.

Hi eaorobbie

You should be able to achieve increased range of travel and speed, but it will be by sacrificing power so you wont get the increase in force and speed that you are hoping for - fast transfer through 200-300mm of movement is completely possible with a powerful motor and a suitable arrangement of the arms of the actuator, but much as I hate to be the bearer of bad news, I'm afraid we are stuck with only ever losing energy in a transfer machine so the increase in speed and travel results in a decrease in power

By using bungees to support the actuator you will be able to reduce strain on the motor and support the arm of the actuator in the right starting position, but it will not boost the power of the machine

You have probably seen it already, but there is a post on page 6 of this thread that shows how my own actuators work including a video of my prototype in plastic corrugated board that may give you the kind of dimensions you need - it took a lot of experimenting to find the right ballance of range and power with the wiper motor... shouldnt be a problem for your winch, but the layout might still help!

Tim

@ eaorobbie

... for quick reference, here is the video explaining the principles - you definitely shouldnt have to worry about overloading your motor, but the linkage concept transferring rotary to linear motion should be useful to help you avoid destroying the setup if your motor spins out of control!

http://www.youtube.com/watch?v=LoN6CnaUxdc

Tim

Hi all

After the disappointing attempt to make a screw type actuator, I got to thinking about power - and how best to multiply the power of the wiper into useful movement that can easily be incorporated into a design...

I spent a morning playing with different voltages, but the motor was little better and I was getting frustrated so I went for a ride to chill out... well - I nearly went for a ride! As soon as I got the bike out, I got my idea!

I am now experimenting with gear reduction by cog and chain!

In keeping with everything I am trying to achieve with this project, the cost of this experiment so far is £0 8)
For anyone wanting to try this, all you have to do is go to your nearest mountain bike servicing shop, and ask for either a freewheel or a cassette. As they usually find their way into the bin during servicing because of a single broken tooth, or slight warp, or just for being slightly worn, you should be able to get them for nothing (except maybe a tip to ensure an ongoing supply )

From the bin in the bike servicing department in Halfords today, I rescued all these lovely greasy (and one brand new but slightly warped) gear sets and chain!



Continuing my love affair with cheap bearings from skateboards, I was overjoyed to discover that the standard bearing is a perfect fit into the centre of the freewheel/cassette! The gear set itself contains a ratchet that freewheels in one direction only, so it is necessary to add a bearing to the centre that can freewheel in either direction (not necessary for the cog attached to the wiper motor, but definitely needed for the other sets in a construction).

Just whack it in with a hammer



I then stuck a bolt through the middle, with a nut as a spacer and added the arm of the wiper motor - as the wiper motors have a special fitting onto the arm, you really REALLY need to make sure that it is the right way up before welding it in place!!





With this all held in place by a mole wrench, I tacked the wiper arm, spacer, and gearset together, and then disassembled it to do the proper weld (badly as usual!)



The finished gear/arm set:




When attached back on to the wiper motor, I use a long bolt onto the end of the wiper drive centre, which supports the centre of the gear set via the skateboard bearing - which means that more force can be applied without overly straining the welds (which is particularly important for me, as my welds are rubbish!)


Here is a video of the motor running - the gear set is very slightly warped, but at the speeds I need, I dont believe it will be a problem.

http://www.youtube.com/watch?v=qb1y58D_FEc

So how much power do I hope to be able to get?
The calculations are pretty simple - at the greatest reduction, the smallest cog has 14 teeth, and the largest has 28 so the power is doubled and the speed is halved with each small to big connection - with 7 variations in speed to play with (for each pair of gearsets) if needed to find the right ballance of speed with power.

So the small driving cog on the wiper set connects to the big cog on the driven set, and the small cog on the driven set connects to the... well... watch this space!!

The power of the wiper motor is doubled, and doubled again by this (although the speed is halved and halved again as the price you have to pay)

Scary Terry (http://www.scary-terry.com/wipmtr/wipmtr.htm) reckons on his wiper motors producing about 13.5 foot pounds of torque on high speed... assuming the same basic power, this setup should be able to produce (13.5 x 2) x 2 foot pounds, = 54 foot pounds of torque

I am already part of the way towards building this into a functional actuator... hopefully before next weekend there will be more!

Tim

Chain Drive Proof of Concept Test

First proof of concept test with chain drive:

http://www.youtube.com/watch?v=ahbCQsiIjnI

Not as clear as I would have liked in the video, but by looking at the brass link on the vertical chain, you can see that the speed of movement achieved with this design by attaching an actuator arm to the vertical chain will be approximately 10 inches (250mm) per second at a little more than 4 times the force of a wiper motor (a little different because the radius of the cog is not the same as the length of the wiper arm).

In theory, there is realistically no limit to the useful length of the vertical chain - if the available force proves to be in the range I am anticipating, I hope to have approximately 2 feet of movement available by extending the upright.

The attached video below shows the type of sim I am now aiming for - the sim pit concept of RolandVanroys is superb, but a complex build... this type of actuator is much simpler to build! It is very possible that i will end up with a hybrid concept - a chain driven setup like Rolands belt drive, but for now I am sticking to pure concepts to establish the boundaries of possibility with a simple wiper motor.

In fact, it was a friend showing me this video that got me interested in sim building in the first place!!

http://www.youtube.com/watch?v=qIe9O529MR0

Tim

The video shows a setup with really powerfull AC motors. I doubt that you will achieve comparable performance with wiper motors.
It may work with enough gear ratio reduction, though. But of course not with comparable speed. Using springs to counter the payload will be probably neccessary.

regards

I know that it wont be as good as this, but i think an actuator with 4 times the power of a wiper motor moving 250mm per second isnt too far away from being useful - and like I say... I'm really interested to know where the boundaries with a wiper motor are!

As you say, gearing down further and trading some more speed for power (aiming for approximately 150mm per second like an SCN5), I believe that the power would be comparable... time will tell!

Interestingly, the guys in buggiesbuiltforfun.co.uk have got a similar design which appears to use a motor from an electric window:
Heres a link to the actuator page: http://buggies.builtforfun.co.uk/Sim/linear-act.html

They have a really nice scissor type solution to offer bungee support on the frame too!

Heres a link to the platform page: http://buggies.builtforfun.co.uk/Sim/platform2.html

Here's a clip showing the power of a similar but smaller actuator - not in youtube unfortunately, but still definitely worth looking at!
Really dont try this at home!!
http://buggies.builtforfun.co.uk/Sim/Images3/DSCF0123-c.avi

Tim

Actuator test - massive success!!!

I had a really good day today with my prototyping - I think the actuator design is now in a good state to work towards completion of the build... then the fun begins with trying to get the profile right!!

below are a few clips showing progress, and also giving a few hints on the use of different electronic configurations with a switch that could be useful in your build

The last clip shows just how powerful this type of actuator can be - this is a cut-down version of the finished build I am aiming for - the total range of movement in this prototype is about 1/4 of my final plan.

I think this shows that the power and speed are more than good enough for a build - it really holds it's own considering it is only half built, and is under such a massive load!!

Enjoy!

How to wire a switch to change the direction of a motor:
http://www.youtube.com/watch?v=OEqk4v_J5Gw

How to use the same switch setup to protect a linear actuator
http://www.youtube.com/watch?v=aOjfuV1Qj-g

What this looks like built up:
http://www.youtube.com/watch?v=eYOIPfVahHQ

What wiper motors are REALLY capable of - this uses 12 volts only from a PC power supply!
http://www.youtube.com/watch?v=grfrH1DTS20

Tim 8)

I would use two push switches (with rollers preferably) and direction recovery diodes... This doesn't really stops the motor, just reverses the direction which would be useful only if you would like to make a rocking chair (or couch) lol!!!

Tim,
I think this is a great illustration of using gearing to increase the power at a reasonable speed. I get the point of the switch, just for your prototype, not intended to be at rocking chair Thanos! lol....im pretty certain that most of us on here weigh less when balanced then a couch. Way to step outside the box and inspire thinking. I very much enjoy your posts where you provide explanations as well as the prototype videos. This thread has become a part of my daily routine to check on your process. Watched this before I went out to my shop last night,.,,,unfortunately i caught my cloth chair on fire while welding, but still motivated! Good job and keep the great posts coming.

Great job Tim as always!

The wiper motor seems to be faster in on the down stroke without load. You could reverse it and make it faster against the load, which would somewhat compensate for the speed difference.

Your guides and video demonstrations are great for newcomers! Keep them coming!

Thankyou

The finished actuator design that I will hopefully have built in prototype next weekend has a different layout, but all I need to do in order to correct the imballance, is connect the driving arm to the other side of the chain... exactly why it's always worth prototping! :thbup:


OR... theres the old wiper motor speed fix trick - for anyone who hasn't seen this, here is a video posted by tronicgr:

http://www.youtube.com/watch?v=XZHyXPuMxFo

Tim

Hi Thanos

I am really interested to know more about the switch and diode setup you suggest - is there any threads in the forum or other web resources you could point me to please?

In fact, I started with 2 switches...

I tried a different setup to begin with that had 2 make to break switches that simply turned the motor off at the extremes.
The problem I had with simply stopping the motor at the extreme of posistion, was resetting it back to an acceptable position - the mechanical resistance in the cogs and chains system here is huge!!
(you can see this from the last video where the mechanical resistance alone supports my heavy sofa before I even switched the motor on!)

There is no way that I could turn it back by hand, which meant a complicated system of switches to allow you to manually fire a burst of power into the motor to reset the position to an acceptable state and release the switch... and always the possibility that you might send it the wrong way, and accidentally do the damage that the switch is in place to stop!

At least with this setup, I dont have to think about which way I need to turn the motor to fix the problem, and if for what ever reason the software crashes, I just have to reset it and play on rather than having to manually reset the actuators...

Like I say though, I am completely open to any suggestions - many thanks for any information you can provide on this (and any other!) subject!

Regards

Tim

PS. I dont think I'd get away with a rocking couch - I'm in enough trouble as it is if my wife finds the video of this load test! :rofl:

Here is how you can use two switches and diodes for automatic direction recovery!!!

Adding_Safety_Limit_Switches.pdf


Just be sure to use large diodes that can handle the current your motors draw. The load will be momentary until the switch is released so don't worry to be exactly on the current requirements!

Thanos

:clap: :clap: :clap:
Superb - It's so simple too... I cant believe I didn't think of this!! I guess I just wanted it to be more complicated :

I will build this into the prototype when I get home on Friday and post a comparison video alongside the other switch solution to show the difference in action.

Many many thanks Thanos!

:

Tim

A quick update:

I went to Maplin after work, and picked up a couple of switches for £1.39 each...
they can handle 12Amps and 250V so they should be fine in this build.

The diodes are from my lucky bag again - I'm getting some real value for money out of that!

They can handle enough current for a short time, but as you can see from the video, they dont allow the full current that the motor really wants... (you can tell because when the direction is reversed, and as the switch is released so that the current isn't just flowing through the diode, the speed picks up)
Thats ok though, because as soon as the motor has kicked the actuator arm away from the switch, the full power is allowed to flow through the switch once again.
(Under load this will probably be insufficient though and they may burn out, so I guess I will have to buy some after all! )

Here is a video explaining what is actually happening:

http://www.youtube.com/watch?v=fW0AI_wsC14

As soon as I have this rigged up into the actuator that I showed before with the other switch solution, I will post a video to allow direct comparison.

Tim

Wow, I am very interested in this thread, I have been reading it 6 or 7 pages at a time whenever I get a free moment. Very interesting work, amazing job guys. I have a bit of a question for Tim, or anybody for that matter,. I have a Velleman board that I had bought 2 years ago and not really gotten around to doing anything with it. Anyway, to my question, I have a pile of Combat Robotics parts sitting in my garage and was thinking of doing something similar. Essentially will my Victor 883 Speed controls http://content.vexrobotics.com/docs/ifi ... -25-06.pdf operate off of the 8055? I am going to try to use my robot's drive motors for motion if they do. 24 volts DC. I have these NPC-T64 drive motors with the 14:1 Gear set as well as the 20:1 and the billet machined gear cases. These have been sitting since the last season of Battlebots and unless I make a remote controlled beer cooler, I fear they are wasted. http://www.npcrobotics.com/products/vie ... 0&mode=gfx

Thanks again for your assistance...

Hi Supermodfan - glad you like the thread!

I'm afraid I have no idea about the victor 883 speed controllers you have, as I am working entirely with super cheap components and using no prebuilt boards with the exception of the K8055...

...but I can tell you a site where you will find the answer you need

I suggest that you post the question in the Velleman forum in exactly the same format, and with the same link to the pdf.
They should be able to give you a very good answer - particularly as you are able to provide a technical document from the controller manufacturer

If you find that the board is compatable with the K8055, I would be delighted to help you with profiles or other setup details, though you may find all the other answers you need earlier in this thread - the more hardware options we can come up with the better!

Good luck!

Tim

Here is a video of a basic test using switches to stop the motor at the extremes of position, and diodes to enable their restarting by simply sending a signal in the opposite decision

(I have had to slow the motor down to usefully demonstrate the action).

http://www.youtube.com/watch?v=fW3C4pglcXE

Here's the original switch based solution - which, I have to be honest is rather crude compared to Tronics solution, as it doesn't stop the actuator... instead it just changes the direction!

http://www.youtube.com/watch?v=eYOIPfVahHQ

Tim