my first build

[wires]

first off i guess i should say hello.
my name is Adrian and i am new here. recently i have found myself needing something like this and after some googling i found that it is doable.

this is going to be a tight budget build mainly because i was recently made redundant. i got some sound advice from a few guys here and what i realise is that i really needed to make some realistic decisions about what i want this machine to do. after some thought i feel that routing mdf or plywood is what i need, milling alloy would be nice but not a requirement.

the design will be a moving gantry with a workable area of roughly 2' by 4'. that is not set in stone and may be changed later as i work out the details.

the bulk of my budget will go on a controller/stepper package. i do have some electronics training but i wouldnt be confident in the reliability of anything i could build. since this is an important part of the machine i guess i should get this part right. i dont want to spend hours trying to get the machine working only to find the controller isnt working right. i am looking at a kit from http://www.diycnc.co.uk/html/cnc_systems.html
it is the three axis kit for £207 i am looking at.
if anyone has any feedback on these products, good or bad, i would be interested in hearing it.

the machine will be primarily constructed from mdf. i do have a table saw, router table, pillar drill and some other stuff so i shouldnt have much trouble making the chassis. i will keep an eye out for some second hand linear bearings and lead screws but for the moment the good stuff is out of my league so i will attempt to make my own. i will design the chassis with a view to maybe upgrade it if i get the chance.

my idea at the moment is to use threaded rod and make a delrin or nylon nut, use some ball bearings at either end and see how that works out. i understand that direct drive of a threaded rod will limit my feed rates but i am thinking i might try to gear it somehow, anyone any experience with this method? also i do understand that ball bearings will not handle the thrust loads and that is something i will look into.

on the linear end of things i am thinking of getting some nice straight angle stock, probably alu, and making a linear bearing like i have seen on some builds here. this idea basicly has some mdf chamfered on either end, angle fixed to the chamfered ends. on the gantry side it has some more angle with bearings fixed to it, one assembly fixed top and bottom of the rail. one problem i see with this design is possible loose or tight fit. i intend to make the bottom bearing on the gantry spring loaded with a bolt to adjust the tension. i can see it in my head but its a bit hard to put in words. i think it should work well, in theory at least.

i have gotten hold of solidworks and i am trying to model my ideas so i can see how the various parts go together. it is a steep learning curve though as i have only ever used CAD electronic design before, not mechanical.
anyway i guess thats it for now. i will continue to think through the design and try to model it, might post some pics of the model to get some feedbcak. it will be well into the new year before i put blade to wood because i want to have a clear plan.

i would like to thank all the guys for their advice so far and i hope to get some feedback on these ideas as i will probably keep second guessing myself.

[irving2008]

Adrian,

A comprehensive and thoughtful post. I have added my thoughts inline below...

Best of luck!

Irving...


... milling alloy would be nice but not a requirement. a realistic decision!

the design will be a moving gantry with a workable area of roughly 2' by 4'. that is not set in stone and may be changed later as i work out the details.

the bulk of my budget will go on a controller/stepper package. i do have some electronics training but i wouldnt be confident in the reliability of anything i could build. since this is an important part of the machine i guess i should get this part right. i dont want to spend hours trying to get the machine working only to find the controller isnt working right. i am looking at a kit from http://www.diycnc.co.uk/html/cnc_systems.html
it is the three axis kit for £207 i am looking at.
if anyone has any feedback on these products, good or bad, i would be interested in hearing it.
The DIYCNC package is used by several of our members and is well thought of, as is its proprietor. For your needs, if budget is an issue, you might consider his earlier System3 (http://www.diycnc.co.uk/html/system3.html) single board unit at £97. Dont forget you will need to add a power supply if you dont have something suitable to hand. Look for 30v at 8 - 12A, these can be picked up on eBay relatively cheaply (like this one (http://cgi.ebay.co.uk/30Volt-30V-8-11A-Power-Supply-PSU-ASI-Industrial-SMPSU_W0QQitemZ330386721833), get in quick) or DIYCNC do a 24v one at £50.

the machine will be primarily constructed from mdf. i do have a table saw, router table, pillar drill and some other stuff so i shouldnt have much trouble making the chassis. i will keep an eye out for some second hand linear bearings and lead screws but for the moment the good stuff is out of my league so i will attempt to make my own. i will design the chassis with a view to maybe upgrade it if i get the chance. With the savings on the controller you might consider Igus' Dryline-W bearings (http://www.igus.co.uk/wpck/default.aspx?PageNr=1994&CL=GB-en). I've not used these but the rail is £52/m and the bearing blocks under £7 so pretty price competitive to ball linear bearings.

my idea at the moment is to use threaded rod and make a delrin or nylon nut, use some ball bearings at either end and see how that works out. i understand that direct drive of a threaded rod will limit my feed rates but i am thinking i might try to gear it somehow, anyone any experience with this method? also i do understand that ball bearings will not handle the thrust loads and that is something i will look into. Threaded rod will work, but with a 1mm pitch its really very slow. I'd recommend you look at TR12x3 (which is around £12/m from Marchant Dice on about 4+ weeks delivery at the mo) if you plan to make your own delrin nuts (you need a 100mm bit of rod to make a tap). Igus also do TR10x3 which may be cheaper but no prices on their site, have to request them (let me know what they are if you do). For 10 or 12mm rod at the speeds you likely to get you can get away with bearing support at the motor end only with a single ballrace and a thrust bearing. Technobots (http://www.technobots.co.uk/acatalog/Online_Catalogue_Miniature_Bearings_382.html)do cheap thrust bearings at 8 and 10mm ID.

on the linear end of things i am thinking of getting some nice straight angle stock, probably alu, and making a linear bearing like i have seen on some builds here.... the idea has been done before and works well if you have the constructional skills

...it will be well into the new year before i put blade to wood because i want to have a clear plan. some people say, "dont think it, do it", but i agree a plan never hurts...even if you dont keep to itlol

[Swarfing]

Adrian

You have done it now you know there is no turning back :clap:. A few pieces of advice i can give you is that sometimes it is better to swallow hard and buy the right thing if you can. When you look at the cheap stuff and it does not work you then look at the next bit of cheap stuff and you end up paying twice. With that you could have just bought the more expensive item in the first place. A good place to save money is the mounts. Why use them if you can tap out the ends of the rod and put a decent beefy screw in? this also helps in that you are drilling smaller holes and better for accuracy. remember that bearings need to go onto turned down rod ends and not directly over the thread (my mistake).

I used the Chinese 5 axis all in one drivers you see on Ebay. these have worked out for me but did take a lot of working out. In hind sight i should have gone for a modularised unit as i am limited to no more that 5 amps.

Good luck!

PS set a date to first cut as it is a good kick up the jacksy when it looms i just made mine by 1 week

[wires]

the reason i am looking at this controller is that i have read that it is quite easy to blow a stepper driver, particularly when you dont know what you are doing. with this board that only means you need the driver board and not the whole thing as is the case with single board units.

power supply wont be a problem. like you said, will just keep an eye on the bay from now on for a deal.

with regards to the linear stuff i have come to the realisation that this will not be my last cnc machine. what seems to be the norm here is that you build your first machine and use that to build your second. i know my linear idea is tricky and will take some time to get right but i have lots of time now and i do have lots of patience to get it right. also i am pretty handy with the table saw and router. when i get it modeled i will post some pics for feedback.

i like that TR12x3. the price aint bad either, i might just go for that. it seems i will have to get the ends turned down to fit bearings and coupler. i will have to find someone local to get that done, might know a guy.

i had a good look around the dryline site but i cant decide which size rails and bearings i need. guess thats just my lack of mechanical training.

with reagards to 2e0poz points i do see the logic in buying the propper stuff and under normal circumstances (have a job) i would definatly just spend the £500 on linear bearings and be done with it but i really dont have the cash at the moment. that might change in the future but for now this idea of mine wont cost much to try. i already have mdf and i have a load of ball bearings which might do the job, all i need is some alu angle and a lot of time (being unemployed, time is not a problem). i am not sure i know what you mean about mounts, could you elaborate?

the reason for not doing any construction right now is there is no point in cutting a load of mdf only to find the design change and stuff going in the bin. in my mind its better to just get the design finalised and then as 2oepoz says, you get it made in less than a week.

i can see the advantages of propper linear bearings but i just cant afford them at the moment plus i am not convinced they would really make that much difference to my machine given i will just be cutting mdf or ply. assuming i can get my the home made linear tight engouh and pay close attention to minimising chassis flex i think it will work out fine.

like i said this is my first machine and it will have specific uses. no doubt i will make another more elaborate one soon after this is completed and if i have a job at that time i will put all the propper stuff into it.

thanks for all the advice guys even if i dont take all of it, its just a budget thing. i am getting pretty excited about all this, keep having ideas flying around my head. cant imagine what i will be like the day i fire her up:yahoo:

[irving2008]

i like that TR12x3. the price aint bad either, i might just go for that. it seems i will have to get the ends turned down to fit bearings and coupler. i will have to find someone local to get that done, might know a guy.


You can turn them down with the help of an electric drill, a file and some emery paper and a couple of 12mm ID ball races in some MDF pillow blocks. Support the TR12x3 in the pillow block bearings, suitably spaced, the one at the business end leaving say 50mm showing. Chuck the far end in the drill and whizz it at 1000rpm, then use the file to shape the end to a near fit and finish with the emery paper, doing trial fits on the appropriate bearings or measuring as you go...

You might find the use of a grinder to take the hardening off the surface helps, a dremel or similar is an alternative to a file and can be quicker...

At least one member has used this technique successfully...




Of course if you know a man with a lathe...

[wires]

i have been trying to think how i could do that myself, that sounds simple enough. i have a bench grinder and vernier calipers.

i do know a friend of a friend who works in a tool making workshop, might see if i can call in a favour when the time comes.

being a noob i am not sure what a pillow block is. do you mean a small square of mdf with a hole for the bearing? you then make a hole larger than needed in the machine chassis so that the pillow block can be moved to line it up perfectly?

also, if i am going to get some thrust bearings i need to understand them a little. my understanding of thrust bearings is that they handle the load laterally along the length of the screw but they have no ability to handle side to side loads. do i need a ball bearing and a thrust bearing in each pillow block or am i missing something?

[irving2008]

...being a noob i am not sure what a pillow block is. do you mean a small square of mdf with a hole for the bearing? you then make a hole larger than needed in the machine chassis so that the pillow block can be moved to line it up perfectly?
A pillow block is just a bearing mounted in a carrier, usuall cast or pressesd steel with ears to bolt down with. you can buy them ready-made or, for this purpose, two 12mm id races mounted at equal height in some blocks of MDF screwed to an MDF base will suffice... its only to provide some support for the leadscrew while its being worked on so moderate accuracy only needed...


also, if i am going to get some thrust bearings i need to understand them a little. my understanding of thrust bearings is that they handle the load laterally along the length of the screw but they have no ability to handle side to side loads. do i need a ball bearing and a thrust bearing in each pillow block or am i missing something?For the actual machine you need a bearing block which retains the leadscrew radially (ball race) and axially (2 x thrust bearing), alternately you'd use 2 x angular contact bearings but these are hard to find in small sizes. See the pic for a guideline

[wires]

ah i see. so if it is supported by thrust bearings either side of a ball bearing on the drive side of the screw there is no need for a thrust bearing on the other end, just another ball bearing. i think if i am going to go that complex on the machining of the ends of the screw i will just get someone with a lathe to do it right.

thanks for clearing that up for me, i was wondering how exactly i would do that. one more thing decided:smile:

[jonm]

hi wires
if you look at my build , i turned down 12x3 trapeziod screw
with a drill , grinder and file it turned out really well

i have the system 3+ board from diync also a power supply
support from roy is first class

i built my router from mdf to start with but i am now in the process of rebuilding in aluminium , should have went the alu route to start with

[Swarfing]

Sorry I totally missed the point of the cash flow issue and i can appreciate your position. I did explore the use of patio door rollers as slides. I buttered up a couple of window fitters to have a butchers at some old patio doors and i was pleasantly surprised at what i still believe to be a viable idea for a low budget machine? just a thought or you could just go for plain old Bronze bushes?

Paul

[wires]

that screw did turn out :heehee: good for you jonm, maybe i will tackle it myself.

as the design progresses i will be trying to source many of my parts localy. in doing that i will be visiting and looking for engineering supply shops. i have a few contacts who can advise me where to go. i think i will be beginning construction around end of january. that gives me plenty of time to go and see whats out there. i think its highly likely that i will find someone selling propper lead screws and some sort of propper linear bearings at a price right for me but at this stage i will not count on it.

i see your the point all of you make about getting the right stuff straight away but there are really two things stopping me from spending that sort of money right now...

*of course as mentioned there is the financial situation which is crap but there is no point crying about it. plenty of people in the same boat. the answer to that one is to just get on with it and dont let it get you down.

*more importantly is the fact that i dont yet fully realise the full potential of having a machine such as this and dont really know all the uses i will have for it and as such i dont want to buy expensive linear bearings only to find that i really need ones 200mm longer in the long run. i would rather build something cheap and then after using it decide if i need it more acurate or smaller or bigger...i do have the garage space to go for an 8' x 4' if i want.

i do have a quiet enough life, no kids yet, house is built and mostly finished so i can afford to spend a lot of time getting this finnished.

i promise there will be no more talk about budget or financial situation from now on. its depressing:rolleyes:
you all know the brief at this stage. keep it simple, get it working and then with the experience of doing that decide if the specs are right for a better one.

does anyone have any ideas for coupling the motor to the screw? i see a lot of guys using propper couplers and some guys making their own.
where do you get your couplers?

i have been busy learning solidworks and trying to visualise my ideas, pretty cool bit of software. i will see if i can get the model finnished over next few days and get some pictures up.

cheers for all the help guys, i would be lost in design decisions without this forum.

[Swarfing]

here is a quick hit and a good price....http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&item=220452134294&ssPageName=STRK:MEWAX:IT

[jonm]

v groove bearings on ebay 4 for £26

[irving2008]

here is a quick hit and a good price....http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&item=220452134294&ssPageName=STRK:MEWAX:IT
Thats a good price for L050 couplers, he also sells miniature bearings and thrust bearings and 10mm double row angular contact bearings at reasonable prices.

[wires]

i might just go ahead and buy some of the tr12x3, some bearings and couplers. does anyone know what internal diameter the bearings should be, ie. when i grind down the tr12x3 at what diameter does the thread disappear? i dont want to get bearings too big and end up with them supporting some of the thread. i assume the 12x3 relates to overall diameter and thread pitch so i dont know how deep the thread goes.

[wires]

those v groove bearings look interesting, never seen anything like that before. i would need quite a few of them between all the axis. seen another seller has them for 4.99

[Kn8]

Hi Wires

The root diameter is 7.69mm have a look at; http://www.nookindustries.com/acme/AcmeIsoInfo.cfm?id=15

Regards.

Ivan

[jonm]

hi wires
these ones http://cgi.ebay.co.uk/V-groove-Bearing-41mm-diameter-10mm-bore-15mm-wide-CNC_W0QQitemZ260522748294QQcmdZViewItemQQptZUK_Hom e_Garden_PowerTools_SM?hash=item3ca85da986

i looked at these be carefull .. could be wrong but i dont think they are accurate enough think they are butcher boy bearings

[jonm]

wires
there seems to be a few conflicting data sheets on 12x3 trapezioid screw mine came from marchant dice and with just the thread ground off a 8mm skate bearing will not go on ... needs a tad more taken of with a file to get a snug fit , so i think with just the thread off 8.5-9mm


http://www.roymech.co.uk/Useful_Tables/Screws/Trapezoidal.html

[irving2008]

Hi Wires

The root diameter is 7.69mm have a look at; http://www.nookindustries.com/acme/AcmeIsoInfo.cfm?id=15

Regards.

Ivan

As an alternative to turning the screw down you could look at these components (http://www.dumpstercnc.com/trapezoidal_12x3x1.html)from dumpterCNC

Note when specifying TR12X3 make sure you say whether its single or 2 start thread - the common one is single start. 2-start is faster (6mm lead) but the components etc are more expensive.

As to the Root Dia, DIN103 spec say max 8.5mm, min 7.68mm depending on fit and load spec. In my experience, generally expect it to be around 7.95mm making it a good fit on 8mm ID bearings.


jonm - I must type faster - Roy's site (http://www.roymech.co.uk)is a good place to bookmark...

[wires]

thanks for all the info guys, i think i now know enough about them to order some. going to go with merchant dice as you guys seem to reccomend them. cheers

[Swarfing]

Adrian

Something else you can look at is aluminum casting, a furnace is cheap to build and you can get some custom components to suite at the same time? look up charcoal furnaces for cheapness and you will be surprised what can be achieved in your own back garden for less than twenty quid and as much scrap ali you can get your hands on?

[wires]

i knew this would happen. i wanted to keep this cheap but i have been looking marchant dice online store and i see lots of stuff i like at reasonable prices:rolleyes:

they do hardened high carbon round rail
http://www.worldofcnc.com/products.asp?recnumber=14
from reading other threads i think i need 16mm for the bed. they have this...w16 (h6) at £1.90 per 100mm. now i assume that product is 16mm diameter, not sure what the h6 means though.

also have these http://www.worldofcnc.com/products.asp?recnumber=17
at £5.40 for the LME16UU i assume that is to suit a 16mm rail.

they do rail supports as well not too dear.

now, i am only guessing when i browse all this stuff. all the stuff looks really nice but i am not really certain what to get. i have guessed the above two products and eyeballed a few others, left to my own devices i would probably go with these. am i right in thinking that i can get away with thinner rails in the gantry?

i am now considering putting off the purchase of the controller/steppers untill maybe february so i can spend some money on these. does anyone know if these guys accept paypal.

this is all your fault by the way:heehee:
this is turning out the same as my main hobby, nitro rc racing. i keep getting tempted by nice shiny new stuff and spending a fortune on it, the rc car i currently drive is worth over €1000

for anyone interested

xray xb808
2 x hitec 7955tg
go tech gx7r
spektrum dx3s

any comments on the stuff i am looking at?

[irving2008]

they do hardened high carbon round rail
http://www.worldofcnc.com/products.asp?recnumber=14
from reading other threads i think i need 16mm for the bed. they have this...w16 (h6) at £1.90 per 100mm. now i assume that product is 16mm diameter, not sure what the h6 means though.

The h6 is the tolerance on the diameter I believe which is as you surmise 16mm... However these are not supported rail. They will deflect and bounce on long runs and are not really suitable for more than 300mm on a lightweight machine. Although they can be used where accuracy is not paramount, really you need to be looking at fully supported rail (http://www.worldofcnc.com/products.asp?recnumber=85). If you want to use unsupported rail you need to be considering 20mm or bigger...


also have these http://www.worldofcnc.com/products.asp?recnumber=17
at £5.40 for the LME16UU i assume that is to suit a 16mm rail.That is correct, however these round units are hard to use without the matching pillow block (http://www.worldofcnc.com/products.asp?recnumber=42)


am i right in thinking that i can get away with thinner rails in the gantry? Yes, but 16mm unsupported rail will deflect 0.2 - 0.3mm with a 6kg router/z-axis in the middle on a 600mm run, so this limits your depth accuracy and surface finish.


i am now considering putting off the purchase of the controller/steppers untill maybe february so i can spend some money on these. does anyone know if these guys accept paypal. MD accept PayPal, but dont expect fast turnaround... you could be looking at a few weeks based on some recent experience by members... they are a tad overloaded...

[wires]

hmm, there is always something. guess they would be fine for the gantry but not the bed. i will still definatly go for the tr12x3 but i will have to have a good think about the linear stuff. this is why i havent cut anything yet. thanks for the info

[wires]

i was also looking a the drylin stuff but honestly i am lost in the specs. i have no idea what i need to handle the loads i will have. every time i try to bluff my way through it i get a pretty high price. the bed length will be 1500mm. im sure you guys are sick of answering noobie questions at this stage but does anyone have any ideas on what might suit from drylin?
i just dont know what to go for, round rail or angular, single or double carriage. or maybe i will just go back to plan A:eek: