CNC Pickup Winder

[Hugh Evans]

Hi everyone, I've been a long time lurker and this is my long overdue first post. However, I hope that many of you will find it of interest. I am the co-founder of a young upstart guitar company. For more than the past decade I have consulted for every major guitar manufacturer in the US, am a former Martin employee, and have been building, repairing, and modifying guitars since I was about 15. In college I spent my first year and a half majoring in Classical Guitar performance until overpractice lead to extensive nerve damage in my left arm. I moved on to study Mechanical Engineering and Biochemistry with full knowledge that my chops would likely never recover to their full former glory. No matter what I've done in life: It always ties back to guitars, so when the opportunity arose earlier this year I decided to finally pull the trigger and start a company of my own. (More details on that when our website finally launches, but you can always follow us on facebook by searching for Davis-Evans Industries.)

The reason for this post is to announce that we have begun work on designing a CNC pickup winder capable of stand alone operation (and capable of reading G-code from a SD/MMC card) or under computer control. We have already designed and built a CNC router that has proven precision of 0.0005" over a 2'x3' cutting area, and geared towards guitars. However, after looking at pickup/coil winding options in the marketplace we were less than pleased. What we plan to offer includes the aforementioned specs in a precision machined and constructed product. At the core we plan to base the design on Arduino and maintain an open code base to allow for updates and maximum flexibility. The big kicker is our target price range of between $200 and $300. It is extremely likely that a set of bobbin holders will be included and custom designs available for minimal cost. This price includes all of the necessary electronics (Arduino, LCD, Stepper Motors and Controllers, Power Supply, Enclosure with Base, Tensioner, Guides, etc.) Best of all, it's a project that could technically be built by hobbyists using off the shelf components at the cost of losing technical support and and form of warranty.

That's the gist of what we're up to. This is the biggest and most active community of pickup makers online that I'm aware of: so if you have any questions concerns, suggestions, or ideas... Please bring them up. We have most of the parts ready to go and I am just getting started drawing up everything in SolidWorks before fabrication and assembly of our first prototype begins. With any luck we will have a fantastic CNC winder ready for the market within the next 3 to 6 months.

I sincerely look forward to hearing your feedback. Updates will be posted frequently.

Thanks for reading!

[David Schwab]

Hi Hugh,

Welcome to the Pickup Makers Forum!

This sounds very interesting, and I look forward to seeing the finished product.

[big_teee]

Caught my attention Also!
What does the router sell for?
Welcome,
T

[kayakerca]

$200 - $300. . . I must be missing something here as a Schatten sells for what, ~ $375 as a manual traverse winder with a counter? While I don't always lean toward the "Show-me, I’m from Missouri" side of life, on this one, I think I'll have to wait and look forward to reading the reviews.

[Chris Turner]

Sounds very interesting.

IMHO, you may want to use something like a ChipKiT Uno32 though, rather than an Arduino if you'll be implementing more than one stepper motor and want to run more than one at the same time. The Arduino just can't handle running more than one simultaneously very well.

The ChipKIT Uno32 uses a 32 bit processor, gives you more pins and RAM, and can handle more than one stepper smoothly (I've seen people run up to 4 at once at high speed). Also, with the ChipKiT compiler, you can use Arduino code to control it.

[Hugh Evans]

Thanks for you interest everyone, and I appreciate your welcomes to this forum.

We haven't decided for sure if we want to market our CNC router or not, but the idea is being investigated. The final price will depend on available options. But off the top of my head, to assemble the complete gantry, including all of the extruded aluminum, linear motion systems, stepper motors, lead screws, anti-backlash nuts, (several mounts are available for a variety of routers.) I would venture to guess roughly $2000 as a ready to assemble kit. A full aluminum T-slot cutting bed would likely be about a $250 upgrade.

Kayakerca: The Schatten winder is exactly what inspired us. When I showed it to my business partner, his jaw was on the floor immediately. Why? Because he built his first tabletop CNC router (dremel actually) on a budget of $100. With the price of steppers dropping all the time and good control systems readily available we already know that this price is absolutely attainable. We are up in the air as to whether or not we will open source the entire design, although our current leaning seems to favor making the codebase open to to encourage tinkering. Even if you screw up a single component, none of the individual parts are more than about $30. The plan is to power bobbin rotation with a much more crude 4 step/revolution motor, because it saves money on a part that doesn't require as much precision. It may end up attached by a gear or drive belt to increase max RPM. The traverse system will use a 200 step/revolution motor for precision control of the winding pattern. Wind count will be handled with a hall effect sensor. If you're bored, just google a few of these parts and see what kind of number you end up with. I suspect you will be shocked by how affordable CNC parts have become. We will need to fabricate a decent amount in house and have the precision machining capabilities to do it.

Chris: Thanks for the input, we're already ahead of you. But again, if we are missing something stupid in the design process comments such as yours will be invaluable... So thanks again.

While the main goal of this product is CNC winding (BTW, we are absolutely open to making a modified version for anyone who is winding transformers as well) I suspect a similar high quality machine geared towards hand winding with foot pedal control will also be desired. We may even be able to set up the CNC system to accept a foot pedal control a bypass the traversing system, although that would come at added cost. For the most basic option (our answer to schatten's product) we could go so far as to retain the Arduino to drive the LCD and handle counting and speed measurements. The trickiest part will likely be finding a suitable pedal. I can get new empty Wah pedal shells for about $40. Since we do aluminum casting, it's possible we could roll our own as well. A cheap dirty yet perfectly acceptable drive system for steppers can be achieved using a 555 timer to send pulses. If we go that route, I would hope to reach a price point of $100-$150.

My suspicion regarding high market prices for winders is this: How many quality pickup winders do you know of on the market right now? Most CNC based systems cost a couple grand. Other simple hand winding designs actually anger me. Charging nearly $400 for a motor with two shafts in a plain box with a primitive LCD counter bolted down to a piece of melamine covered chipboard takes some real audacity if you ask me. First and foremost my company is a guitar company, and we love designing and building our own equipment. I figure I might as well put some of the stuff I had to study as a Mechanical Engineer to good use. When it's all said and done, communities such as this likely has quite a few people with electronics and programming backgrounds who may well find ways to improve what we have... and we welcome that.

Again, thank all of you for your interest and I look forward to documenting our progress. The next thing you see will likely be renderings from SolidWorks.

[RedHouse]

Thanks for you interest everyone, and I appreciate your welcomes to this forum.

We haven't decided for sure if we want to market our CNC router or not, but the idea is being investigated. The final price will depend on available options. But off the top of my head, to assemble the complete gantry, including all of the extruded aluminum, linear motion systems, stepper motors, lead screws, anti-backlash nuts, (several mounts are available for a variety of routers.) I would venture to guess roughly $2000 as a ready to assemble kit. A full aluminum T-slot cutting bed would likely be about a $250 upgrade.

Kayakerca: The Schatten winder is exactly what inspired us. When I showed it to my business partner, his jaw was on the floor immediately. Why? Because he built his first tabletop CNC router (dremel actually) on a budget of $100. With the price of steppers dropping all the time and good control systems readily available we already know that this price is absolutely attainable. We are up in the air as to whether or not we will open source the entire design, although our current leaning seems to favor making the codebase open to to encourage tinkering. Even if you screw up a single component, none of the individual parts are more than about $30. The plan is to power bobbin rotation with a much more crude 4 step/revolution motor, because it saves money on a part that doesn't require as much precision. It may end up attached by a gear or drive belt to increase max RPM. The traverse system will use a 200 step/revolution motor for precision control of the winding pattern. Wind count will be handled with a hall effect sensor. If you're bored, just google a few of these parts and see what kind of number you end up with. I suspect you will be shocked by how affordable CNC parts have become. We will need to fabricate a decent amount in house and have the precision machining capabilities to do it.

Chris: Thanks for the input, we're already ahead of you. But again, if we are missing something stupid in the design process comments such as yours will be invaluable... So thanks again.

While the main goal of this product is CNC winding (BTW, we are absolutely open to making a modified version for anyone who is winding transformers as well) I suspect a similar high quality machine geared towards hand winding with foot pedal control will also be desired. We may even be able to set up the CNC system to accept a foot pedal control a bypass the traversing system, although that would come at added cost. For the most basic option (our answer to schatten's product) we could go so far as to retain the Arduino to drive the LCD and handle counting and speed measurements. The trickiest part will likely be finding a suitable pedal. I can get new empty Wah pedal shells for about $40. Since we do aluminum casting, it's possible we could roll our own as well. A cheap dirty yet perfectly acceptable drive system for steppers can be achieved using a 555 timer to send pulses. If we go that route, I would hope to reach a price point of $100-$150.

My suspicion regarding high market prices for winders is this: How many quality pickup winders do you know of on the market right now? Most CNC based systems cost a couple grand. Other simple hand winding designs actually anger me. Charging nearly $400 for a motor with two shafts in a plain box with a primitive LCD counter bolted down to a piece of melamine covered chipboard takes some real audacity if you ask me. First and foremost my company is a guitar company, and we love designing and building our own equipment. I figure I might as well put some of the stuff I had to study as a Mechanical Engineer to good use. When it's all said and done, communities such as this likely has quite a few people with electronics and programming backgrounds who may well find ways to improve what we have... and we welcome that.

Again, thank all of you for your interest and I look forward to documenting our progress. The next thing you see will likely be renderings from SolidWorks.

Good luck with that Hugh.

You are litteraly stepping into a queue of 100's of others who are already there, with their (CNC) offerings, but hey who am I to say.

Most of the others who have been where you are comming to realized quickly the costs to market a machine are greater than first appear.

Specifically regarding CNC winders, if you browse the threads here you'll even find a recent one about the CNCDudes offering, which only went 2 rounds (batches of sales) before he bumped his price up into the range everyone else is in, and right out of where most people around here are willing to pay.

There have been many who came to this board with CNC related offerings looking to make a few bucks, I beleive Elepro was the first flogging CNC related hardware and since then (2009) while he has managed to sell a few dozen chips, maybe 3 or 4 buyers have actually built, an operate, their "CNC Winders".

It appears one of those "great idea's" that end up going nowhere. It's a good business practice to know your target market.

People here are trying to make money, building/selling/reparing pickups in a market that is oversaturated, and often stagnant in this economic time, you may not find this to be as fertile a sales opportunity as it may appear.

YMMV, good luck.

[Hugh Evans]

I personally don't think pickup winders are a big enough market to be worth basing a business around. That's why my main business focuses are on guitars, and continuing to serve as a consultant to the guitar industry. If pickup winders were our bread and butter, rather than a side project we need to design for our own purposes anyways, this would be a completely unsustainable plan and I would agree with you 100%. Also, since we already have a nice CNC router to work with, most fabrication steps will be automated and the assembly process will require minimal soldering and mostly involve mechanical fasteners. The electronics are all off-the-shelf parts and easy to assemble and program. I've been tied into manufacturing most of my life. This is nothing more than a fun side project (I mentioned above that the entire design may be released under an open source license) that I hope will serve to improve access to CNC pickup winders for anyone who wants to try it out without spending $500+. My ultimate dream is to see people openly trading g-code for specific coil designs. The SD/MMC reader and g-code interpreter will make this entirely possible.

Does my plan make a bit more sense to you now? I appreciate the sentiment of your post as I could easily see someone with less experience in manufacturing, business, and economics lose a lot of money believing they will get rich selling pickup winders at relatively low margin. When it comes to actual manufacturing, guitars are my business. Very few people have the knowledge, experience, and desire to manufacture guitars on a large scale. But it happens to be something I'm very good at. Thanks again, and I hope you might even stand to benefit from this project and the data it will generate.

[big_teee]

I personally don't think pickup winders are a big enough market to be worth basing a business around. That's why my main business focuses are on guitars, and continuing to serve as a consultant to the guitar industry. If pickup winders were our bread and butter, rather than a side project we need to design for our own purposes anyways, this would be a completely unsustainable plan and I would agree with you 100%. Also, since we already have a nice CNC router to work with, most fabrication steps will be automated and the assembly process will require minimal soldering and mostly involve mechanical fasteners. The electronics are all off-the-shelf parts and easy to assemble and program. I've been tied into manufacturing most of my life. This is nothing more than a fun side project (I mentioned above that the entire design may be released under an open source license) that I hope will serve to improve access to CNC pickup winders for anyone who wants to try it out without spending $500+. My ultimate dream is to see people openly trading g-code for specific coil designs. The SD/MMC reader and g-code interpreter will make this entirely possible.

Does my plan make a bit more sense to you now? I appreciate the sentiment of your post as I could easily see someone with less experience in manufacturing, business, and economics lose a lot of money believing they will get rich selling pickup winders at relatively low margin. When it comes to actual manufacturing, guitars are my business. Very few people have the knowledge, experience, and desire to manufacture guitars on a large scale. But it happens to be something I'm very good at. Thanks again, and I hope you might even stand to benefit from this project and the data it will generate.

Yes your plan makes Perfect sense, and we all anxious to see what you come up with.
So, Please keep us posted, I for one wish you the best, and hope you can market something very useable.
Also if the Unit can be used in a manual mode, that would be great, also!
Thanks for Keeping the Pickup Winding Community in mind.
Terry

[RedHouse]

...My ultimate dream is to see people openly trading g-code for specific coil designs. ...

Ok, nice idea, and again with all due respect good luck with that.

Around here you will see people using vaguarity's describing winding techniques, "loose" or "tight" or "heavy scatter", in fact in the (several years) I've been hangin' round here I've never seen anyone share/swap/expose their winding patterns to others.

Don't get me wrong, I'm not meaning to sound negative here, but it's just not an aspect of pickup winding that is (normally) shared for obvious reasons. I'm trying to visualize this forum in post-Hugh-gcode-world where people are sharing/swapping g-code for winding pickups, sure it would be cool, but not seeing it (realistically) but then again, who am I to say, what do I know anyway.

I'm not trying to be a buzz-kill, I'm just a reality check kind'a guy. I post these aspects of the notion as being new here you probably haven't read all the old threads where we've gone down the this 'ol CNC winder road before.

I'd certainly love to see what you have in mind. BTW do you actually have a winder unit (kit/whatever) right now or is this the opening volley "hey everybody, what would y'all want in a CNC winder"?

[kayakerca]

Ok, nice idea, and again with all due respect good luck with that.

Around here you will see people using vaguarity's describing winding techniques, "loose" or "tight" or "heavy scatter", in fact in the (several years) I've been hangin' round here I've never seen anyone share/swap/expose their winding patterns to others.

It may be that we are being a "little" unfair to our fellow forum members Brad. I have asked the tpl question and been happy with the feedback. I really wouldn't expect someone to detail out there million dollar magic formula, just take a little off the learning curve.

Jason Lollar:
"15 to 30 i would call scatter wound particularly if you have some slop in your traverser so it never actually repeats where the turns lay. 100 tpl is T top territory"

David King:
"My sense is that scatter-wound implies a much looser range with variability between say 15-30 tpl or 50 and 100 tpl."

captcoolaid:
"I do layers with a small scatter. Traverse forward at about every other turn and on the return it fills the gap. I do not use a cnc but not hand traversed either. I call it stack winding to a since. Roughly for 42 wire86 turns or so."

[David Schwab]

I personally don't think pickup winders are a big enough market to be worth basing a business around. That's why my main business focuses are on guitars, and continuing to serve as a consultant to the guitar industry.

Good idea, and having this on the side, like Schatten does.

I'd buy one of these as soon as they are ready. I think others would too. I have been accumulating parts for a CNC winder, but then putting the whole thing together is very DIY with very little in regards to an actual design! If I could get one ready to be put together, all the better.

Good luck with the project. I was checking out the Facebook page with the DND router. That looks like a fun toy to have!

[Hugh Evans]

I have a rough design worked out already. It's a project I wanted to take on at some point. We recently upgraded the steppers on our router, so with 3 perfectly good ones sitting around in the shop I decided it was at last time to pull the trigger. The main decisions still left involve whether or not I want to include variable tension during the wind and therefore use all three motors, if I want to drive the traverse with a rack and pinion or a lead screw (probably lead screw at this point), and how to implement the tensioning system.

Part of what I love about pickups is exactly what you've warned me will be stirred up. I have a weird enough background to have a lot of fun with this. I don't believe that there's anything magical about pickups simply because they are simple electromagnetic transducers. What's great about having a winder like this is that I can thoroughly test a wide variety of hypotheses. A pickup winder and my ears are not enough for my methods: I will also be constructing a test rig to empirically measure magnet strength, the impedance response curve, capacitance, and inductance. An additional trick remains up my sleeve that will be revealed at a later date.

Most likely, I will start off simple with Strat type single coils. Equal magnets, number of turns, tension, etc., varying only the winding pattern. One will be wound to machine perfection while the other will random and effectively non-repeating. It should fill the bobbin perfectly despite this (I have a specific algorithm in mind to accomplish this, and if you're familiar with sonar you might know what I'm thinking.) And then see if any differences can be detected between them and my mystery pickup. If there's enough interest, experimental design can be discussed in a thread followed by test results and possibly a set of recordings. G-code can be made available to seek replication of my results. Plus, if some really great winding patterns come out the other end I will make them freely available. I would *love* to get my hands on some original yet dead pickups from some of the most classic Fender and Gibson designs. Why? To put an end to the trade secrets. A few people who have done a ton of rewinds know what the patterns look like and can estimate other variables such as tension. I would like to map some complete winding patterns along with taking regular measurements of wire thickness to look for signs of necking due to tension. Best of all, I can take it a step further by popping some wire into an Instron and learning a bit more about it. My hope is that as the information starts flooding out into the world pickup winders can start perfecting the science behind the art.

Sound like fun?

As I said before, I'll post renderings of the design as soon as it comes together a bit more. There are a few enclosures I'm looking at and layouts I'm considering for the design.

[RedHouse]

It may be that we are being a "little" unfair to our fellow forum members Brad. I have asked the tpl question and been happy with the feedback. I really wouldn't expect someone to detail out there million dollar magic formula, just take a little off the learning curve.

Jason Lollar:
"15 to 30 i would call scatter wound particularly if you have some slop in your traverser so it never actually repeats where the turns lay. 100 tpl is T top territory"

David King:
"My sense is that scatter-wound implies a much looser range with variability between say 15-30 tpl or 50 and 100 tpl."

captcoolaid:
"I do layers with a small scatter. Traverse forward at about every other turn and on the return it fills the gap. I do not use a cnc but not hand traversed either. I call it stack winding to a since. Roughly for 42 wire86 turns or so."

I'v no intention of being unfair about our forum members, carry-on up the kyber.

[salvarsan]

Sounds very interesting.

IMHO, you may want to use something like a ChipKiT Uno32 though, rather than an Arduino if you'll be implementing more than one stepper motor and want to run more than one at the same time. The Arduino just can't handle running more than one simultaneously very well.

The ChipKIT Uno32 uses a 32 bit processor, gives you more pins and RAM, and can handle more than one stepper smoothly (I've seen people run up to 4 at once at high speed). Also, with the ChipKiT compiler, you can use Arduino code to control it.

The mind boggles.

The 32-bit ARM-based Arduino Due was released yesterday Monday 10/22/2012. It fits all the 3.3v compliant Arduino shields.
Downside: $49 vs. Uno32's $27, but the Arduino has ~4x more memory.

Question:
Why would you need more than a 1-D track for guiding wire onto a bobbin? A 3D gantry seems excessive.