153 replies to this topic

[cooperman435]

Dont have access anymore no but the design or a modified better one is esay to make, if theres anyone out there willing to help with the programming side Ill do the rest ?

[PyroSam]

When I go to insert an image to a message here it just gives me an option for a web address for the image, how do I upload an image direct from my computer?

Never mind - I see, you can't upload a picture directly, you must link to it.

Edited by PyroSam, 07 July 2012 - 03:25 AM.

[PyroSam]

I can do CAD reasonably well, I'd like to do a nice set of prints and a parts list so others could create the machine on their own.

I agree, the software is the stumbler.

[PyroSam]

Here are some pics of the various incarnations of the WASP:

As near as I can tell this is the original version and closely matches the patent diagram.


This is a WASPIII.


This is a WASP IV, this one used a 970 oz-in motor.


This is the Stinger - I think this is the first version to use the aluminum framing instead of wood / aluminum (at least I have ran across a WASP V yet so I think this is the one to follow the IV). I included this one as it has all the parts labeled and I would suggest we adopt this naming convention as we discuss the device so we're all on the same page.


This is the Super Stinger, I believe it does larger shells than the Stinger.


This is the latest version - the Mini, it does 1-1/2" through 6" shells and appears to be refined for easier/cheaper construction.

[PyroSam]

Just a couple notes from looking at the pictures of the various machines.

The WASP Stinger, Super Stinger and Mini are using what I think is 1" x 2" extruded aluminum t-slot framing (1 slot on the side, 2 on top I think equals 1" x 2"). A couple companies make that but the most popular in the US (where these are made) is a company called 80/20 Inc.
http://www.8020.net/

I think the center spine on the WASP Stinger and Super Stinger is made from 2-1/2" x 1/4" square tube (2-1/2" minus 1/4" walls leaves a 2" inside dimension which should be a nice fit on the 2" t-slot framing). This same material is used to make some of the brackets (they cut a side off) and the legs. This material isn't a common size so it's a little harder to find which means it's a bit expensive.

I think a lot of the parts - like the drive rollers, left & right bearings, etc. came from McMaster-Carr Industrial Supply (a large US supplier of industrial supplies). Probably not an option for us but we may be able to at least get some illustrations & dimensions, etc. so we know what to search for over here.
http://www.mcmaster.com

I'll look at the Mini, which seems to have several refinements, closer when I get a little more time.

[PyroSam]

It's raining so no grass mowing this morning, time to look over the Mini pictures. I decided to start at the bottom and just work my way up when analyzing this new offering.

The stabalizing feet are turned on their side and bolt directly to the motor mount so no legs need to be fabricated (they used cut off pieces of the 2-1/2 square tube in the Stingers).

The motor mounts are just straight pieces of flat, no bending or welding required. They cut the end of the crossmember the motor mounts bolt to at an angle to give the motors/rollers the proper angle. This simplifies things a bit but it doesn't allow for the motors/rollers to be slid closer or further apart to accomodate different sized shells like they do in the other machines. Whether that's a detriment or not I don't know. If one were trying to duplicate this model they may want to rethink the motor mounts if they are going to try to upscale it a bit to handle larger shells.

The spine is mounted below the crossmember. This was undoubetedly done to accommodate the smaller shells (it lets the drive wheels get closer together) and is probably a good idea to keep even if upscaling the machine to handle larger shells.

The center spine is made from straight pieces of flat instead of the hard to find and expensive 2-1/2" x 1/4" square tubing. Using square tubing it's necessary to trim away a portion of the top on each end to allow for the left and right uprights and that operation would best be done by using a mill, though it could probably be done by the D-I-Yer by some improvised means. The flats eliminate this issue though and simplify/cheapen construction.

The lower tape guide is now integrated with the spine by drilling a couple holes in the flats to mount it instead of making a separate bracket as was necessary when using square tubing for the spine.

No spacers on the bearing blocks - just not needed due to the angle on the uprights. If one wants to upscale to handle larger shells then spacers may be necessary.

The right side bearing block is no longer using the bent sheet metal spring thing. From reading Ned's Stinger Operator's Manual it seems the right bearing isn't supposed to make constant contact with the shell - it's just there to keep the shell from falling off. I've always found that bent sheet metal spring to be kind of odd, it just seems to me it could be done an easier way. You never know though, if they have a sheet metal roller it may be just as easy for them to make this type of spring as it would be to fabricate some other method.

The tape application arm is no longer one solid, welded together unit, it's now two pieces of flat.

The tape application arm roller is a lot smaller (smaller = cheaper?),

There's no burnish spring.

That's about all I can see - did I miss anything?

[barnsley-mark]

Rs would be very expensive, I've bought a few BIG stepper motors from china off eBay and they were excellent. Look under CNC mill or Cnc lathe ... They are sold for conversion and have the drivers etc

Edited by barnsley-mark, 07 July 2012 - 08:43 AM.

[Arthur Brown]

Why the hurry to copy an old design? The WASP is several years old and some things have been superceded so it should be possible to do a new design to do better and for cheaper.

[PyroSam]

barnesley-mark - Thanks for the tip! I'll give at look at ebay. I would like to find a reliable source that will have the motors long term though.

My thoughts are to develop a set of CAD drawings, a parts list with suggested sources and a solution for the motor control that we can package up into a nice little project so that others don't have to re-invent the wheel. I think Lloyd Sponenberg's ball mill book is a good example - I'd like to provide a similar resource for a shell paster (though electronically circulated and including the control software code for easy loading into whatever solution we find).

That doesn't mean tips like check eBay aren't a good idea though - I would think a standard set of plans and suggested parts would be a valuable resource for scroungers too - they would know the specs for parts that are proven to work and could look for similar more inexpensive substitutes that may require only minor modifications to the plans to accomodate mounting, etc..

Arthur Brown - Why the hurry? Because I loathe pasting shells! Seriously, I enjoy every part of pyro - except shell pasting, it just bores me to death!!! November isn't that far away!

The WASP has been an evolving project. The inventor wasn't an inventor to start with, he was an amateur shell builder who got tired of shell pasting too and decided to build a machine to automate it. There was a lot of interest in it so he started building it for others and I don't think we should forget these other people were also shell builders. Pyros are by nature mostly curious and inventive people and they've given lots of feedback and if you look at the pictures I think you'll see that the machine has underwent a fair amount of revision, no doubt as the result of the user feedback (though some of the changes I'm sure are to cheapen/ease fabrication).

I've done a lot of research on these and one thing stands out to me - no one is complaining! Whether they have an original model or any of the following models, they ALL seem to work! After so many years and so many in the hands of knowledgible and skilled shell builders I think that says a LOT about the design - it works!

Soooo, I'm not inclined to want to re-invent the wheel. There is a design out there that WORKS and it appears that it's not that hard of a job for a D-I-Yer to fabricate the mechanical portions themselves with little need for professional machine shop / welding shop help - that sounds like a good candidate for the foundation of a D-I-Y project to me!

That being said, that doesn't mean I feel we should duplicate one of the models exactly. Though this product started out as a personal project of the inventor it also became a commercial product so I'm sure marketing and replication discouragment plays into the designs being offered. It makes more business sense to offer different models targeted at different market segments - i.e. one for larger shells and one for smaller shells (and possibly a three tier offering - small, medium and large), so if one wants to make the full range they need to buy several instead of doing it all with just one. I'm sure we can design one that can do small through large though with just the need for some adjustments. Even if we couldn't, if we could design different frames and just interchange the motors it would be a huge cost savings. I'm sure we could probably cheapen up the construction cost too - things like is the t-slot framing really necessary? Could inexpensive wood framing lumber or plywood be used? Does it need to be a V setting on a crossmember, or could you make two crossmembers cross into a plus sign and then use 90 degree uprights to slide in and out and up and down for adjustment? Are stepper motors really necessary? Or were they chosen because not as many know CNC programming, pc based CNC control software is expensive and steppers aren't used as much in amateur robotics so adapting easily found programming, basic stamp based applications, etc. isn't possible? Perhaps the servo motors used in amateur robotics would be a better choice for our project?

My thoughts are to do a CAD drawing of the Mini as it appears to be the most inexpensive / easiest to replicate and it does a good size range - 1-1/2" for cakes, etc. up through 6" that are used in good sized displays. Once the drawing is in CAD we can start discussing changes - the nice thing about CAD is that it allows you to "build" the machine virtually so clearance issues, angle measurements, etc. can be taken without expending the time and trouble to physcially build the thing. If someone is in a hurry and wants to get a pasting machine running as soon as possible though then they'll be good to go with the basic plans, for those of us who like to tinker and aren't in as much of a hurry the project will continue to evolve and may become cheaper to build and able to handle a wider size of shells.

To be honest though, I'm not worried about the physical building at all - it's just not physically that complicated. The issue is the software to run the motors - that's where the magic is at. That's recognized in the WASP too, the original unit uses a Peter Norberg < http://www.stepperboard.com > stepper motor controller that has been customized for that application. I've never seen a WASP but my understanding is that the WASP software talks to the controller and interrogates it to determine if it's an original controller or not. If it's not it doesn't work. From what I gather if you try to talk to the controller with other types of software it doesn't work either as the controller also has protection built in and it also interrogates the control software and won't talk to any other software. The manufacturer obviously understands the software that controls the motors is the key to the system - and the biggest stumbling block to a D-I-Y machine.

That being said, I don't think the algorithm used to lay the paper tape down is that complicated and consequently I don't think the software needed to control the motors is that complicated - I'm sure a good CNC programmer could get this running pretty quickly and there are freebie pc based control programs out there such as the linux based one I listed in my first posting. Beyond that, I would imagine a PIC, Basic Stamp, etc. could also be programmed to run one of these, which would be nice as it would negate the need for pc to run the machine.

WE NEED TO FIND A CNC OR ROBOTICS PROGRAMMER! Anyone know one? If we could find one I would be willing to go through the expense of building a machine to test the software on.

[Arthur Brown]

Could you use a single DC controllable motor with two wheels say 50mm and 52mm DIA you'd get a different skew pattern but still cover the whole ball eventually.

Yes getting the software to calculate the number of steps of the motors per revolution of the shell and matching that to the width of the paper tape would be a challenge. Probably this is why the software is proprietry.

[PyroSam]

Hi Arthur,

I'd have to do a little investigation to see if a continual skew would give us the result we are looking for - I've simply never thought abou it.

My thoughts are that this is one area the WASP users have already thrashed out and their method works fine so I would be inclined to just follow suit with them.

If you read the patent you'll see there are three methods of laying the paper down that's discussed.

Method 1 is to lay the tape down in a figure 8 pattern and it's noted that this is the preferred method but apparently this has been re-thought as if you watch the videos of a machine running it does not move bi-directionaly, it only ever skews to the right.

Method 2 is to lay a complete revolution of tape down and then offset the motor speeds to cause the tape to skew in a new direction for the next revolution. This is listed as a less preferred method but it appears to be the one that's been adopted as that's exactly what the machines are doing.

Method 3 is as you describe - to always run the motors at slightly different speeds to cause the tape to continually skew as it's laid down.

I've looked at the way the tape is laid down a bit on the videos and the pictures in Ned's manual and I think I have a grasp of the concept of what the machine is doing.

It appears it starts rotation in a straight line, just short of making a complete revolution the motor speeds are offset to cause the tape to skew to the right. The amount it skews doesn't cause it to cover any part of the last revolution as is commonly done when hand pasting with paper tape. Once it skews the motors again are brought into synch and the tape is laid down in a straight line until it again comes up just short of a complete revolution when the motor speeds are again offset and the tape skewed. Eventually the entire shell is covered with tape and a slight "hole" is left at the poles. This hole is only one layer thick though. Once an entire layer is laid down the motors are not stopped short but rather allowed to over shoot before being skewed again and this causes the axis to change and a new set of poles to be formed while the "poles holes" of the previous layer are now covered.

I haven't taken the time to work out an actual algorithm but I don't think it would be that hard - once you know the diameter of the shell you could calculate how much tape to lay down before skewing - this would control the size of the pole holes. Varying how much the tape is skewed would control the width of tape variable.

I've read a very little on CNC programming but I believe that once the distance is calculated by the algorithm telling the motors to run for that distance is not that difficult programming wise. Telling a motor to run at a different speed to cause the skew wouldn't be that hard at all either.

I really don't think it's that complicated of a thing, it's just a matter of having the knowledge to program the CNC software which is a skill that's not overly common in the general population let alone the pyro population. Beyond that, remember the target market for this machine is the US where amateur pyrotechny is a much larger group than it is over here and the machine is covered by a US patent so there's no incentive for a US CNC programmer to develop anything for the machine as he can't legally share or sell it.

[PyroSam]

Still looking for a deal on eBay - lots look like good deals until you see shipping, I thought eBay was cracking down on sellers charging exorbitant shipping?

[PyroSam]

I haven't yet. I'm out of time for today and working double shifts the next couple days so it'll be a bit before I get back to the search. Thanks for the tip!

[cooperman435]

Ive no problem getting bits here as Id rather get generic and replaceable parts for those I buy, and make the rest.

Ali extrusions are cheap bits why do you need to replace them with wood? its also easilly replaceable and strong with pleanty of accesories avaialble.

Pyro Sam are you able to write an interface program to controll the stepper motors froma set of variables entered into a code? If so Im happy to make a machine if your ok to do the controll side?