Potthoff for $7 !!!
The hardest part to making the modified Potthoff (using the TFE slit tubing and metal side plates) is the TFE cog belt guide bottom. Here's a 7 inch by 1/2 inch TFE disk for $7 on Ebay. The vendor informs me that he gets in between 30 and 100 of these a month. So no need to get carried away bidding them up. More coming along. Another approach would be an aluminum 1/2 inch plate with TFE adhesive tape over it's edge, but that will probably cost you 10 times as much for a roll of the adhesive TFE tape.
(I have no affiliation with the vendor)
http://cgi.ebay.com/TEFLON-VIRGIN-W...ryZ61797QQssPageNameZWDVWQQrdZ1QQcmdZViewItem
Next up will be finding a 1/2 inch cog belt of appropriate length and a cog pulley for it.
Don
The hardest part to making the modified Potthoff (using the TFE slit tubing and metal side plates) is the TFE cog belt guide bottom. Here's a 7 inch by 1/2 inch TFE disk for $7 on Ebay. The vendor informs me that he gets in between 30 and 100 of these a month. So no need to get carried away bidding them up. More coming along. Another approach would be an aluminum 1/2 inch plate with TFE adhesive tape over it's edge, but that will probably cost you 10 times as much for a roll of the adhesive TFE tape.
(I have no affiliation with the vendor)
http://cgi.ebay.com/TEFLON-VIRGIN-W...ryZ61797QQssPageNameZWDVWQQrdZ1QQcmdZViewItem
Next up will be finding a 1/2 inch cog belt of appropriate length and a cog pulley for it.
Don
smoking-amp said:
It seems (to me at least) that the wire has to be bent by 90° when leaving the "storage loop" -what it could be: C belt or shuttle ring- and starting to runn the direction of the toroïd.
A kind of pulley at the point the wire is picked should help.
But its diameter must be lo enough to pass thru the donut !
That said, I'm trying to imagine a "combo" using Potthoff C belt storage AND an independent "wire guide/spill off" in form of brokable ring (so it can be insereted in the nut) and driven at a different speed than the belt.
The added ring is there to pick up the wire off the C belt and roll it on the nut.
This calls for two (electronically controlled/interlocked) stepping motors, but so, we should be able to maintain a constant wire tension, dispensing just the right lenght of wire from the belt for each turn of the guide.
Doing so, we just have to pass the spill off ring into the nut, not the -more cumbersome- shuttle.
Far from perfectly clear in my head yet
Yves.
"we should be able to maintain a constant wire tension, dispensing just the right lenght of wire from the belt for each turn of the guide."
According to Jovil, the dual control gear bobbins are only necessary for heavy wire. The simpler slider types (slack loop accumulator on the side) apparently work fine for up to (down to) size #22 AWG wire (#22 to #38 range). The gear bobbin type recommend size #24 AWG as the smallest wire (#12 to #24 range). Could be the friction dragger gear version breaks smaller wire with too much tension. Also, the belt covered type slider bobbins cover #16 to #24 range.
So primary OT windings should be fine with a slack accumulator/slider type. Secondaries for maybe up to a 100 Watt OT maybe OK. Can always put some spring loaded toothebrushes against the slider for heavier wire.
The problem with even a computer controlled dual drive to eliminate slack is that a winding on an irregular surface (second or third winding on the toroid say) will draw wire somewhat irregularly, possibly snapping the wire unless some tension measurement or compensation mechanism is included.
Don
According to Jovil, the dual control gear bobbins are only necessary for heavy wire. The simpler slider types (slack loop accumulator on the side) apparently work fine for up to (down to) size #22 AWG wire (#22 to #38 range). The gear bobbin type recommend size #24 AWG as the smallest wire (#12 to #24 range). Could be the friction dragger gear version breaks smaller wire with too much tension. Also, the belt covered type slider bobbins cover #16 to #24 range.
So primary OT windings should be fine with a slack accumulator/slider type. Secondaries for maybe up to a 100 Watt OT maybe OK. Can always put some spring loaded toothebrushes against the slider for heavier wire.
The problem with even a computer controlled dual drive to eliminate slack is that a winding on an irregular surface (second or third winding on the toroid say) will draw wire somewhat irregularly, possibly snapping the wire unless some tension measurement or compensation mechanism is included.
Don
I envisioned a fairly short tail for the takeoff shuttle, just enough
to keep the wire from bending hard 90deg. If that shuttle rides
on the rim and thru the donut hole, driven at constant velocity.
And the rim unwinds and/or partly rewinds at a fairly constant
low tension (torque whatever...)
As long as its reliable enough to run overnight unattended, it
don't need to go particularly fast. I am aware the rim reversing
direction to retake slack is probably incompatible with going fast.
to keep the wire from bending hard 90deg. If that shuttle rides
on the rim and thru the donut hole, driven at constant velocity.
And the rim unwinds and/or partly rewinds at a fairly constant
low tension (torque whatever...)
As long as its reliable enough to run overnight unattended, it
don't need to go particularly fast. I am aware the rim reversing
direction to retake slack is probably incompatible with going fast.
Hi,
I've been sorta following this. I like mechanical problems.
I would make the shuttle a 2-piece arrangement. The main outer part would be driven by the motor and wind the wire around the core. Inside this would be the wire spool or carrier (like another bike rim inside the outer one) that would free-wheel around the shuttle. There would be a fixed point on the shuttle where the wire spills and the carrier would be coupled to the shuttle with adjustable friction setting, to maintain tension on the wire. If the carrier was loaded in reverse direction to the winding operation, the slack in the wire would be taken up automatically.
Understand?
Thoughts?
I've been sorta following this. I like mechanical problems.
I would make the shuttle a 2-piece arrangement. The main outer part would be driven by the motor and wind the wire around the core. Inside this would be the wire spool or carrier (like another bike rim inside the outer one) that would free-wheel around the shuttle. There would be a fixed point on the shuttle where the wire spills and the carrier would be coupled to the shuttle with adjustable friction setting, to maintain tension on the wire. If the carrier was loaded in reverse direction to the winding operation, the slack in the wire would be taken up automatically.
Understand?
Thoughts?
"I would make the shuttle a 2-piece arrangement."
The Jovil mylar film shuttle, notice the leather belt around the outside for friction:
http://www.infantron.com.sg/manufact/jovil/7th-pic.htm
another:
http://www.infantron.com.sg/manufact/jovil/14th-pic.htm
Using friction on the Mylar stack (outside belt) and driving the rollered shuttle leads to the mylar making a 180 degree turn around one of the rollers onto the toroid I think. Slack mylar rewinds on the shuttle each turn.
If the rollered shuttle is instead friction dragged (becoming the spill-over feed or bottom feed effectively) and the outside of the mylar is driven by a moving belt, then the mylar could go on the toroid without the 180 twist around a roller. Slack mylar still rewinds on the shuttle each turn.
Thats effectively what the Jovil gear drive shuttles do. The wire sits in a normal shuttle that is driven, and the slider or spill-over is friction dragged. Slack wire re-winds on the shuttle each turn.
A Jovil gear shuttle, notice the two gear sprockets, inside and outside edge:
http://www.infantron.com.sg/manufact/jovil/16gh-pic.htm
This one has got a brake pad for friction in place of one sprocket:
http://www.infantron.com.sg/manufact/jovil/12gh-pic.htm
Don
The Jovil mylar film shuttle, notice the leather belt around the outside for friction:
http://www.infantron.com.sg/manufact/jovil/7th-pic.htm
another:
http://www.infantron.com.sg/manufact/jovil/14th-pic.htm
Using friction on the Mylar stack (outside belt) and driving the rollered shuttle leads to the mylar making a 180 degree turn around one of the rollers onto the toroid I think. Slack mylar rewinds on the shuttle each turn.
If the rollered shuttle is instead friction dragged (becoming the spill-over feed or bottom feed effectively) and the outside of the mylar is driven by a moving belt, then the mylar could go on the toroid without the 180 twist around a roller. Slack mylar still rewinds on the shuttle each turn.
Thats effectively what the Jovil gear drive shuttles do. The wire sits in a normal shuttle that is driven, and the slider or spill-over is friction dragged. Slack wire re-winds on the shuttle each turn.
A Jovil gear shuttle, notice the two gear sprockets, inside and outside edge:
http://www.infantron.com.sg/manufact/jovil/16gh-pic.htm
This one has got a brake pad for friction in place of one sprocket:
http://www.infantron.com.sg/manufact/jovil/12gh-pic.htm
Don
For the modified Potthoff design, I am going to use a thin (maybe 1/16 inch teflon layer on each side of the belt for the "shuttle" sides with a 1/16 inch steel sheet backing behind it. The spill-over side then gets a slit teflon tube placed over the edge of both as before. This way the wire always slides against Teflon everywhere (I might put another 1/16 thick Teflon sheet on the outside of the steel for the loop accumulator sliding surface.)
(Alternatively, the slider edge side of the shuttle could just be one Teflon sheet with the slit Teflon tube fitted to its edge. I would then machine the Teflon edge to have a beaded profile which the slit Teflon tube could just snap over.)
I may add the thin leaf spring (taped together ends) over the belt then to form a closed "shuttle" bottom surface, if necessary to avoid wire bending in the C gap. The Teflon tube edge slider tube surface gets routed thru the toroid and the tube ends connected together by a cylindrical plug, so as to form a continuous spill-over edge.
Don
(Alternatively, the slider edge side of the shuttle could just be one Teflon sheet with the slit Teflon tube fitted to its edge. I would then machine the Teflon edge to have a beaded profile which the slit Teflon tube could just snap over.)
I may add the thin leaf spring (taped together ends) over the belt then to form a closed "shuttle" bottom surface, if necessary to avoid wire bending in the C gap. The Teflon tube edge slider tube surface gets routed thru the toroid and the tube ends connected together by a cylindrical plug, so as to form a continuous spill-over edge.
Don
Yves:
"That said, I'm trying to imagine a "combo" using Potthoff C belt storage AND an independent "wire guide/spill off" in form of brokable ring (so it can be insereted in the nut) and driven at a different speed than the belt."
If the Teflon spillover edge plate on the Potthoff were made as two 180 degree halves, they could be fitted thru the toroid and fastened on to the C side as usual. Then an inverted U shape ring (a broken upside down U ring) fitted over the edge of the Teflon side piece (thru the toroid too). This inverted U ring would become the spill-over wire guide at its end joint. It would slide around on the Teflon edge. The inverted U ring could then be driven by a rubber roller(s) for independent control versus the C "shuttle" belt drive to make for slackless wire feed.
There are some pipe couplings around that are made to join two flanges on the outside that have about the right shape for the inverted U ring. Or one could machine it the hard way from a solid piece. Could even be another Teflon piece then, but hard to drive that with a rubber roller then unless its got a gear like surface. Maybe just put a thin Teflon shim or slit Teflon tube inside the U ring for a slider surface (sliding on the Teflon C side piece edge).
Hmmm...., the C side piece doesn't really need to be split (to fit through the toroid) if the inverted U ring is rigid. The inverted U ring could just jump the C gap and take the wire spill with it. This is beginning to look very nice. Have to work on this a bit yet.
Don
"That said, I'm trying to imagine a "combo" using Potthoff C belt storage AND an independent "wire guide/spill off" in form of brokable ring (so it can be insereted in the nut) and driven at a different speed than the belt."
If the Teflon spillover edge plate on the Potthoff were made as two 180 degree halves, they could be fitted thru the toroid and fastened on to the C side as usual. Then an inverted U shape ring (a broken upside down U ring) fitted over the edge of the Teflon side piece (thru the toroid too). This inverted U ring would become the spill-over wire guide at its end joint. It would slide around on the Teflon edge. The inverted U ring could then be driven by a rubber roller(s) for independent control versus the C "shuttle" belt drive to make for slackless wire feed.
There are some pipe couplings around that are made to join two flanges on the outside that have about the right shape for the inverted U ring. Or one could machine it the hard way from a solid piece. Could even be another Teflon piece then, but hard to drive that with a rubber roller then unless its got a gear like surface. Maybe just put a thin Teflon shim or slit Teflon tube inside the U ring for a slider surface (sliding on the Teflon C side piece edge).
Hmmm...., the C side piece doesn't really need to be split (to fit through the toroid) if the inverted U ring is rigid. The inverted U ring could just jump the C gap and take the wire spill with it. This is beginning to look very nice. Have to work on this a bit yet.
Don
"Any progress?"
I got my Delta bench drill press out of storage (150 lbs) and moved here, along with an Enco milling table add-on (65 lbs) and an Enco rotary indexing table (26 lbs). Add another few hundred pounds of bolts, motors, transformers, toroid cores, and assorted scrap metal stuff that might be useful. I'll be using the drill and add-ons to mill the teflon and some aluminum supports. I've got a real Enco Mill-drill in storage, but at 750 lbs, that was out of the question for hauling back in the car. The add-on milling table:
http://www.use-enco.com/CGI/INSRIT?PMAKA=201-2536&PMPXNO=953096&PARTPG=INLMK3
I recently ordered a tool stand from Harbor Freight to put it all on and it arrived made out of paper thin sheet metal (even though its rated for 500 lbs!). I'm ordering a second one to double up the parts for a safe stand to put the drill press and milling table, etc on. Next time I'll pay the higher price at Enco for a real stand, I shoulda known better.
Got most of my teflon and aluminum material in. I already have some nice big stepper motors but I just found a nice micro-stepping controller for them recently, which I bought two of. I want to test them out with the steppers first before I recommend them though, sometimes current limiting power resistors are needed too.
http://www.mpja.com/prodinfo.asp?number=17451+MS
The micro-stepping controller isn't really needed for the belt drive motor, but may be useful for the toroid positioning motor, where more angle resolution is required for progressive winding.
I've located sources for 5 mm pitch timing belts and sprockets, these can be cut to most any custom width and the sprockets are available in various widths including custom. Belts lengths are available in various standard lengths, but length cannot be custom varied. Some belt tightening mechanism is required in the design.
I still have to order these items, but I prefer to actually construct the teflon and aluminum assembly first to measure the best belt length more accurately. Some milled slots in the aluminum side plate for sliding stepper mounting bolts will suffice for belt tightening.
I may use a tensioned timing belt to rotate the core against rubber rollers too, one of several design options here. The belt can even have some teeth filed down to compensate for the change of toroid radius with the windings going on, to keep constant stepper to toroid angles for precision progressive winding.
OT:
Hauling really heavy loads of stuff in the car is getting more dangerous than ever, as I found out on the trip back. I was minding my own business when I saw a pickup truck and an SUV on RT 81 having a passing battle with each other, coming up from behind. I was thinking right away: here comes trouble. Then they pulled right in front of me, honking at each other, and I immediately thought "insurance scammers!". I was already half way into an evasive manuever into the next lane when they slammed on the brakes. I went sailing by them fortunately. They probably picked on me because the car looked like it was bottomed out from all the weight. Thank goodness I wasn't hauling the 750 lb milling machine in the back seat.
Don
I got my Delta bench drill press out of storage (150 lbs) and moved here, along with an Enco milling table add-on (65 lbs) and an Enco rotary indexing table (26 lbs). Add another few hundred pounds of bolts, motors, transformers, toroid cores, and assorted scrap metal stuff that might be useful. I'll be using the drill and add-ons to mill the teflon and some aluminum supports. I've got a real Enco Mill-drill in storage, but at 750 lbs, that was out of the question for hauling back in the car. The add-on milling table:
http://www.use-enco.com/CGI/INSRIT?PMAKA=201-2536&PMPXNO=953096&PARTPG=INLMK3
I recently ordered a tool stand from Harbor Freight to put it all on and it arrived made out of paper thin sheet metal (even though its rated for 500 lbs!). I'm ordering a second one to double up the parts for a safe stand to put the drill press and milling table, etc on. Next time I'll pay the higher price at Enco for a real stand, I shoulda known better.
Got most of my teflon and aluminum material in. I already have some nice big stepper motors but I just found a nice micro-stepping controller for them recently, which I bought two of. I want to test them out with the steppers first before I recommend them though, sometimes current limiting power resistors are needed too.
http://www.mpja.com/prodinfo.asp?number=17451+MS
The micro-stepping controller isn't really needed for the belt drive motor, but may be useful for the toroid positioning motor, where more angle resolution is required for progressive winding.
I've located sources for 5 mm pitch timing belts and sprockets, these can be cut to most any custom width and the sprockets are available in various widths including custom. Belts lengths are available in various standard lengths, but length cannot be custom varied. Some belt tightening mechanism is required in the design.
I still have to order these items, but I prefer to actually construct the teflon and aluminum assembly first to measure the best belt length more accurately. Some milled slots in the aluminum side plate for sliding stepper mounting bolts will suffice for belt tightening.
I may use a tensioned timing belt to rotate the core against rubber rollers too, one of several design options here. The belt can even have some teeth filed down to compensate for the change of toroid radius with the windings going on, to keep constant stepper to toroid angles for precision progressive winding.
OT:
Hauling really heavy loads of stuff in the car is getting more dangerous than ever, as I found out on the trip back. I was minding my own business when I saw a pickup truck and an SUV on RT 81 having a passing battle with each other, coming up from behind. I was thinking right away: here comes trouble. Then they pulled right in front of me, honking at each other, and I immediately thought "insurance scammers!". I was already half way into an evasive manuever into the next lane when they slammed on the brakes. I went sailing by them fortunately. They probably picked on me because the car looked like it was bottomed out from all the weight. Thank goodness I wasn't hauling the 750 lb milling machine in the back seat.
Don
I had an Enco Mill but it was just to heavy to deal with. I am so glad i didn't buy a real Bridgeport at 2,000lbs! I ended up selling my mill and now i am looking at the smaller table top sherline/Taig models. All i really want to do is machine faceplates and small parts. this way i can get a small lathe too and even the small ones will machine knobs and bearings and what not.
Only problem with the mini stuff is the tooling is MORE expensive!
Zc
Only problem with the mini stuff is the tooling is MORE expensive!
Zc
"Well done with the progress on this & a great 2009 to you!"
Thankyou jkeny! I'm just hoping that I can get going on the machining before too long. (ie, Spring.... kayaking, hiking.... )
Machining stuff is such a tedium though. And messy.
"I had an Enco Mill but it was just too heavy to deal with. I am so glad i didn't buy a real Bridgeport at 2,000lbs! I ended up selling my mill and now i am looking at the smaller table top sherline/Taig models. "
Yeah, I know, the Enco mill-drills looks so cute in the ads, but are incredibly heavy (750 lbs + 150 lbs for the stand, and maybe another 100 lbs for a rotary table) on close encounter. At least it uses the standard cutting bits and accessories. One thing I found when I put mine in storage was that it can be split up into 3 or 4 150 to 200 lb parts. I made some wood boxes with handles for each part. Not too bad to cart them around on two wheel dollies then.
Bridgeport.... that would fall thru my floor.
I've recently seen this interesting half size mill-drill at Harbor Freight (325 lbs):
http://www.harborfreight.com/cpi/ctaf/displayitem.taf?function=Search
(item # 93885)
Claims to have a standard R-8 spindle too.
But its still nearly the same price as the bigger Encos. $1000, and I am wary to say the least of anything from Harbor Freight. The quality there is totally all over the map. I would check one up close at a dealer first, definately.
They also have some smaller mini-mills at around 100 lbs to 150 lbs (and $300 or $400) that look similar (models 47158 and 44991), but on closer inspection of their manual, they don't have a sliding quill for the drill chuck. They move the entire head assembly, including motor, up and down to drill anything. This would be undesireable to me for making incremental milling depth adjustments.
It also appears that made in Taiwan machinery is generally better made, although more expensive than the Chinese stuff. I'm sure the Harbor Freight stuff is from China. But with the $1000 price on the half size model, maybe its a Taiwan model?
Some other interesting models at H-F:
model 97208 4 inch rotary table for $90
gotta see this one, a real OMG moment:
models 66051 and 66052 CNC hobby mills!!!! but $3000 to $5000 (but gee, if they can make a CNC mill for $3000, how come the toroid winders are $24,000)
Enco is mostly bigger stuff.
Don
Thankyou jkeny! I'm just hoping that I can get going on the machining before too long. (ie, Spring.... kayaking, hiking.... )
Machining stuff is such a tedium though. And messy.
"I had an Enco Mill but it was just too heavy to deal with. I am so glad i didn't buy a real Bridgeport at 2,000lbs! I ended up selling my mill and now i am looking at the smaller table top sherline/Taig models. "
Yeah, I know, the Enco mill-drills looks so cute in the ads, but are incredibly heavy (750 lbs + 150 lbs for the stand, and maybe another 100 lbs for a rotary table) on close encounter. At least it uses the standard cutting bits and accessories. One thing I found when I put mine in storage was that it can be split up into 3 or 4 150 to 200 lb parts. I made some wood boxes with handles for each part. Not too bad to cart them around on two wheel dollies then.
Bridgeport.... that would fall thru my floor.
I've recently seen this interesting half size mill-drill at Harbor Freight (325 lbs):
http://www.harborfreight.com/cpi/ctaf/displayitem.taf?function=Search
(item # 93885)
Claims to have a standard R-8 spindle too.
But its still nearly the same price as the bigger Encos. $1000, and I am wary to say the least of anything from Harbor Freight. The quality there is totally all over the map. I would check one up close at a dealer first, definately.
They also have some smaller mini-mills at around 100 lbs to 150 lbs (and $300 or $400) that look similar (models 47158 and 44991), but on closer inspection of their manual, they don't have a sliding quill for the drill chuck. They move the entire head assembly, including motor, up and down to drill anything. This would be undesireable to me for making incremental milling depth adjustments.
It also appears that made in Taiwan machinery is generally better made, although more expensive than the Chinese stuff. I'm sure the Harbor Freight stuff is from China. But with the $1000 price on the half size model, maybe its a Taiwan model?
Some other interesting models at H-F:
model 97208 4 inch rotary table for $90
gotta see this one, a real OMG moment:
models 66051 and 66052 CNC hobby mills!!!! but $3000 to $5000 (but gee, if they can make a CNC mill for $3000, how come the toroid winders are $24,000)
Enco is mostly bigger stuff.
Don
I'm still watching this thread for activity with great interest. I have access to a well equipped machine shop and am interested in the latest developments. I even have a little time to spare to make one. 
I'm also building a CNC machine that would be able to mill out aluminum and Teflon parts...
Thank you for the time you've invested so far!
I'm also building a CNC machine that would be able to mill out aluminum and Teflon parts...
Thank you for the time you've invested so far!
The microstepping motor I wanted for toroid positioning (it goes with the previously mentioned microstepping controller above, same manufacturer) has been out of stock for a while. Supposed to be back in stock in 3 months:
http://www.mpja.com/prodinfo.asp?number=17455+MS
But I recently figured out that this other one is also rated for microstepping. (has to have a sinusoidal waveform from its pole shapes):
http://www.mpja.com/prodinfo.asp?number=17748+MS
So I just ordered 4 of them to try out. Will likely arrive end of this week or early next.
I decided that my original toroid positioning scheme using a timing belt needed modification. I was going to use one stepper with a long belt pulling the toroid up against two rollers near the winder mechanism. But it became apparent that the belt has to be routed around a lot of stuff there, including the other stepper for the wire drive and the slack wire accumulator spring tensioner. This would require a bunch of belt idlers to steer it around them.
So now I am pursuing a design where the toroid positioning belt is double C shaped, just like the wire drive belt. The belt wraps around most of the toroid then does a 180 around two stepper drive pulleys near the winder. The belt then routes back around the toroid (but at larger radius, no rubbing). Two more steppers are positioned near the max width points of the toroid to route the outer belt sections safely and they also push against the inner belt and toroid, providing centering constraints. These two steppers are spring loaded against the toroid, and so accomodate various size toroid cores. Finally a 5th pulley is arranged at the far back of the toroid (not rubbing the inner belt) to tension the outer belt section and is mounted on an optical encoder shaft. It's spring tensioned, maybe 9 inches away from the back of the toriod, so the outer belt section has clearance from the inner belt section.
The four positioning steppers are run in parallel from the same microstepping controller.
I purchased all the belts and pulleys from here:
http://www.sdp-si.com/web/html/viewcat.htm
Mostly from the inch drive components catalog. They're pretty pricey though. About $12 per sprocket, $21 per belt. I priced out what it would take to make some idler shafts with ball bearing mounts and they came to around$100 each. That's when I decided that multiple $10 stepper motors were a much better way to go. Although, one could machine off the stepper rotor poles to make a ball bearing idler shaft I guess.
I suppose, if one used enough spring loaded steppers around the toroid, with rubber rollers, that the toroid belt could be dispensed with. That seems to be how the commercial design is configured. They must have some computer algorithm to figure out the toroid angle, since its radius changes (dynamically) as its being wound. Good accuracy of toroid angle is necessary for progressive winding.
My plan is to adhesively bind a thin rubber sheet (maybe serrated sections) to the back side (non toothed) of the toroid timing belt, of similar width to the wire layer being wound. This spacer layer ends at the C belt gap where wire is being wound on. This way, the belt runs at constant radius during the winding operation. (the wire layer taking the place of the rubber spacer during rotation) A bit of a pain to configure drive belts though. Probably have to have a collection of preconfigured drive belts handy.
One concern yet, is the phasing between the multiple steppers during a layer wind. Some inaccuracy is bound to creep in with the toroid radius (and spring loading) affecting the belt length between stepper sprockets. There are rubber flex couplings available for allowing a little flex of the belt sprockets. I'm hoping I can skip that (they're expensive as H), maybe just not tighten the belt so much. At this point, I will depend on the optical encoder for feedback of belt position. I will just have to experiment on this. (the optical encoder I picked up surplus some time ago, but probably are similar available on Epay. I'm using a 1000 count encoder, and the belt sprocketing is arranged for 2 micro steps per optical count)
Don
I also have another hot project going on, modifying a Tek 576 curve tracer for tube tracing. Slow progress on the winder.
http://www.mpja.com/prodinfo.asp?number=17455+MS
But I recently figured out that this other one is also rated for microstepping. (has to have a sinusoidal waveform from its pole shapes):
http://www.mpja.com/prodinfo.asp?number=17748+MS
So I just ordered 4 of them to try out. Will likely arrive end of this week or early next.
I decided that my original toroid positioning scheme using a timing belt needed modification. I was going to use one stepper with a long belt pulling the toroid up against two rollers near the winder mechanism. But it became apparent that the belt has to be routed around a lot of stuff there, including the other stepper for the wire drive and the slack wire accumulator spring tensioner. This would require a bunch of belt idlers to steer it around them.
So now I am pursuing a design where the toroid positioning belt is double C shaped, just like the wire drive belt. The belt wraps around most of the toroid then does a 180 around two stepper drive pulleys near the winder. The belt then routes back around the toroid (but at larger radius, no rubbing). Two more steppers are positioned near the max width points of the toroid to route the outer belt sections safely and they also push against the inner belt and toroid, providing centering constraints. These two steppers are spring loaded against the toroid, and so accomodate various size toroid cores. Finally a 5th pulley is arranged at the far back of the toroid (not rubbing the inner belt) to tension the outer belt section and is mounted on an optical encoder shaft. It's spring tensioned, maybe 9 inches away from the back of the toriod, so the outer belt section has clearance from the inner belt section.
The four positioning steppers are run in parallel from the same microstepping controller.
I purchased all the belts and pulleys from here:
http://www.sdp-si.com/web/html/viewcat.htm
Mostly from the inch drive components catalog. They're pretty pricey though. About $12 per sprocket, $21 per belt. I priced out what it would take to make some idler shafts with ball bearing mounts and they came to around$100 each. That's when I decided that multiple $10 stepper motors were a much better way to go. Although, one could machine off the stepper rotor poles to make a ball bearing idler shaft I guess.
I suppose, if one used enough spring loaded steppers around the toroid, with rubber rollers, that the toroid belt could be dispensed with. That seems to be how the commercial design is configured. They must have some computer algorithm to figure out the toroid angle, since its radius changes (dynamically) as its being wound. Good accuracy of toroid angle is necessary for progressive winding.
My plan is to adhesively bind a thin rubber sheet (maybe serrated sections) to the back side (non toothed) of the toroid timing belt, of similar width to the wire layer being wound. This spacer layer ends at the C belt gap where wire is being wound on. This way, the belt runs at constant radius during the winding operation. (the wire layer taking the place of the rubber spacer during rotation) A bit of a pain to configure drive belts though. Probably have to have a collection of preconfigured drive belts handy.
One concern yet, is the phasing between the multiple steppers during a layer wind. Some inaccuracy is bound to creep in with the toroid radius (and spring loading) affecting the belt length between stepper sprockets. There are rubber flex couplings available for allowing a little flex of the belt sprockets. I'm hoping I can skip that (they're expensive as H), maybe just not tighten the belt so much. At this point, I will depend on the optical encoder for feedback of belt position. I will just have to experiment on this. (the optical encoder I picked up surplus some time ago, but probably are similar available on Epay. I'm using a 1000 count encoder, and the belt sprocketing is arranged for 2 micro steps per optical count)
Don
I also have another hot project going on, modifying a Tek 576 curve tracer for tube tracing. Slow progress on the winder.
Hi all,
I've been studying the drawings and find the concept brilliantly simple!
Has anyone fabricated this device?
Would any DIYers be interested in investing in one? I'm giving serious thought to making a few.
My version would have a reduction/motor to drive the belt and the toroid's ring and pinion.
Just brilliant...
I've been studying the drawings and find the concept brilliantly simple!
Has anyone fabricated this device?
Would any DIYers be interested in investing in one? I'm giving serious thought to making a few.
My version would have a reduction/motor to drive the belt and the toroid's ring and pinion.
Just brilliant...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.