Hi Gang!

I received a startingly small but heavy skyscraper of a box yesterday from Patrick Hood-Daniel's CNC conglomerate world headquarters in Houston, TX. At first impression, I don't see how a whole CNC machine boasting a cutting area of 24" x 48" x ~6" could fit in a box just 12" x 12" x 3.5' Hoping to fill the surprising dearth of information on this particular machine here in the CNCzone, I'm going to detail my build of Patrick's kit.

I ordered one of his first 1.2 kits, sans electronics. I have already built a small CNC machine from scratch, but I viewed this machine as a logical upgrade, and will move my existing electronics to this one.

Patrick saves critical space in the shipping container by requiring the builder to source the largest parts: The actual 2x4 cutting surface, 4 long aluminum angle irons and a longer X axis linear motion screw. Everything else is included.

if you want to know more about this machine, visit Patrick's website (http://www.buildyourcnc.com).

Alllriiighty....

I have a long commute home, but it's an even longer commute when you've got a CNC machine kit in the back seat that you've been waiting months to assemble!

Upon opening the little cardboard skyscraper, I was impressed by the use of space. The box contained three shrinkwrapped bundles of MDF parts jumbled together like a 3-D puzzle, a shrinkwrapped stick of metal (the Z, Y screws and aluminum,) four boxes of hardware, and two crayons. I can only guess the crayons were a gift from the kids.

I think the inclusion of the hardware in this kit is a particularly significant advantage. Patrick includes everything- hundreds of screws, nuts, bolts, as well as all the bearings and couplers you'll need. Just the hardware weighs about ten pounds! I grossly underestimated how much the hardware contributed to the cost of my first CNC machine-- After the smoke cleared, I'd spent nearly $400 just on fasteners, leadscrews, couplers and bearings. Patrick also drills and taps the angle iron for the linear motion trucks.

In the photos, I've already sorted the hardware into bins for easy assembly, and I already assembled all the trucks. ( I couldn't help myself :) )

Also of pleasant surprise, not only are all of the edge holes for the cross dowel connectors already drilled, but all the I-beams for the main supports are pre-assembled. I think the kit will go together very quickly, save for one thing:

There are a lot of parts to this kit, but nothing is labeled. It is obvious how many of the parts go together, but some of them aren't so obvious. I'm confident it will become obvious as I get further into the build, but it probably helps tremendously that I have already built a CNC machine already.

Note: I'm not endorsing Patrick's machine specifically, and I don't want this blog to sound like a commercial. More to the point, the instructions for this machine exist solely in video form on his site, (and at this time I don't think the instructions show the final steps: completion of the main table, then again, it may be so mind-bogglingly obvious that it doesn't warrant another video.) While they are very entertaining, it is a little confusing to try to build this machine strictly from the videos. The best strategy is to watch a video, then go sort out your parts and assembly and figure out how everything fits together. Then you'll figure out what questions you'd ask (such as: specifically what kind of screw should be used where) I found they will be answered in the video.

Like I told Patrick: I fancy myself a reasonably smart guy, but I still routinely reverse at least one important piece when I assemble something from Ikea, and that's following well-written directions and clearly identified parts. :D

I'm hoping to get the Z and Y axis assembled tonight! I'll take more pictures and post my progress and thoughts here!

Cheers, Jim

On Patrick's website, the assembly instructions begin with the Z axis, so I set out to find all the pieces comprising the Z axis. Luckily, most of them are wrapped up in their own bundle of shrink wrap, but you're going to have to fish the smaller pieces out of the bigger bundles.

It's a really good idea to reference the assembly videos several times before starting this step. Because there are no printed instructions (yet- a manual is in the works) and the parts aren't labeled, it's easy to get it wrong.

For example, in the picture, I thought I'd identified and laid out all the Z parts, but that nut holder is wrong- that part is for the X axis.

Also, there are several different sizes of bolts included in the kit- save yourself from disassembling and reassembling everything by noting what size bolt goes where as you build.

The backbone of the Z axis is a small I-beam that came pre-assembled in my kit. There are tooling marks all over most of the parts, so for this reason and that I'm a super-precision junkie, I sanded the ends of the I beam flat and square.

Patrick came up with the idea of using angle-headed countersink screws to hold down the aluminum angle used as bearing surfaces. During this build I've found that if you tighten one side more than the other, your angle irons will be forced slightly off center in places.

Use a caliper and make sure all your angles are nice and straight, as well as offset by the same amount. I have advised Patrick to machine guidance grooves in all parts that will have angle irons mounted to them, so future kits will go together more easily and precisely.

Jim,
Great job so far on the build log, looking forward to watching your progress.

I do think you'll need a bigger can of dustoff than shown in the picture when this thing starts cutting!

Gordon

Hi Gord!

See the part with two giant holes in it (it looks distantly like a tennis racquet?) Well, that first hole is for the wet/dry vacuum hose. Patrick's machine is designed with a semi-built in dust collector. Just make your own dust skirt and you're good to go.

In other news, I stopped using canned air. I got a blowgun attachment for my scuba tank. Yeah, I could have gotten a compressor, but space is extremely limited in my workshop.

I should work on the machine tonight, but I'm pretty wiped out from a long day.

Cheers, Jim

Hi Gord!

In other news, I stopped using canned air. I got a blowgun attachment for my scuba tank. Yeah, I could have gotten a compressor, but space is extremely limited in my workshop.



So does that means you can use Nitrox to extend your dusting time? :)

LOL! A fellow diver! I just got back from Bonaire. Gotta love that nitrox kick after the dive. I could use a little stamina boost after my workday and commute. I think about working on the machine all day and when I get home I've got nothing left!

I did get started on the gantry, but I gotta snap some pics. Soon, I promise!

Jim,
Are you done yet? Hope all is OK and that you're just getting too much done to post here.

Gordon

I have some progress to show, but I hit a wall with one phase of construction and I have been emailing with Daniel back and forth about it. It's important to point out that my kit is one of the early R&D models, and all the kinks hadn't been worked out of it yet. It's essentially a "Beta" machine, and as a beta tester I've been trying to help Patrick streamline the design and construction process.

I probably should have waited with the build log until I had finished the machine, and then chronicled the adventure in a timely fashion, because this log has been a rather uninteresting read, eh?

I'll post some stuff soon! Cheers, Jim

PS: The other excuse: my camera is somewhere in the workshop...and I'm not sure where.

On the final version of the kit, many parts will have been significantly redesigned and go together easier

Hi Gang!

I found my camera! Actually, as the machine came together, I came to realize just how bad my "bass ackwards" approach to the CNC shed had become. Even though the machine is nearly finished, I was forced to reconfigure everything about the shed. During that process, I found my camera. More importantly, I found the memory stick full of pictures inside.

Here's a shot of the Z axis assembly fitted onto the Y gantry. It fit nice and snug. 10 nuts and bolts later, the sides were on.

On the last photo, note that small orange drillgun. That's a Ridgid Micros from home depot, the smallest most powerful drillgun I know of. It's 12v lithium ion batteries recharge in just 15 minutes, and they last longer than my full sized DeWalt 12v. One battery pack lasted this whole build, and that includes disassembling everything at least once. It's also the only model I know of that has a chuck instead of a screwdriver hex receptacle. Best $120 bucks I ever spent on a tool.

Anyhoo, it's time to finish the gantry. More to come....

The gantry is bolted to two boxes clamped around the X axis rails. The floor and ceiling of these boxes house the angle-aluminum rollerblade bearing trucks that will contact the rails of the X.

You'll notice that in these photos, the "inside" wall of the boxes has been cut away. Patrick suggested that the structure was plenty strong enough, and trimming the inside pieces would enable me to attach the actual cutting table to the X rails later, enhancing ridgidity.

Now it's time to tackle assembling the X rails.....

Patrick's machine is purported to be the first modular design for a CNC machine- this kit can be scaled from a 2x4 design to a 4x8. He accomplishes this by building the main structures with an I-beam design. If you want to make the machine longer, you can add additional I-beams to make it 8' long. Likewise with the width- adding another I-beam to the width widens the workspace to 4'. After adding I beams, you replace the leadscrew and the angle iron tracks with a longer ones and reassemble. This lends itself to some flexibility- you can configure your machine as a 2x2, 2x4, 2x6, 2x8, 4x4, 4x6 and 4x8. (and possibly even longer if you can find the acme screws for it!)

Space is a serious commodity in my shop, so I stuck with the 2x4' machine.

In my kit, all the I-beams came pre-assembled, which was handy. All you need to do is line up two I beams, grab 2 connector plates and 8 nuts and bolts. After bolting them together, attach the legs with bolts and dowel nuts.

Mine didn't go together quite so smoothly. None of the connector plate bolt holes were countersunk enough for nuts to get a bite on bolts, and the X parts are not symmetrical on the horizontal axis. This means it is very easy to mate one I beam "up" to one oriented "down," resulting in a joint that has a nearly .125" stairstep between them, which would prevent or at least mess up movement of the gantry back and forth. I evened them up and drilled out the countersinks and got them all bolted together.

At this point I also made the executive decision to stop using nuts on the angle-headed bolts for the aluminum angle. The bolts are stressed in at a slight angle and bite hard into the MDF- they aren't going anywhere.

This time, I also didn't crank the bolts down hard onto the alum angle- I'll align it precisely with a caliper later when I fine-tune it all up.

Once the X beams are constructed and have aluminum track mounted on them, it's time to mate them to the gantry. This is a little tricky, but it can be done by one person. In the assembly videos, Patrick puts the gantry on top of the rails and then attaches the bottom connector plate from the bottom. In my shop configuration, I don't have easy access to the far side of the gantry, much less the underside of it, so I assembled it from the top. I removed the main top structure of the gantry and put the lower assembly on blocks so it slightly supported the machine, x rails and all. Then I replaced the top structure, lined up the bolts and cranked them down. Viola! The thing is really starting to look like a CNC machine!

As you can see in the photos, the machine wound up being MUCH larger than the table it's on. I thought that the two outrigger boards I put underneath would support it, but once the whole structure was complete and I wired up the motors, I decided this machine really really needs a level, flat and stable surface. More on that in a bit.

This is the point where I misplaced the camera, so I'm afraid there aren't any photos between this stage and where I am now. (which is: almost complete! I hope to make the first cuts this weekend!)

OK, by now I'd bolted the gantry together and even wired up the motors. Working around the machine on this workbench could be very trying: there's practically no access to the far side of the machine, and once the cutting surface was finished and installed, there would be zero access.

With the machine supported on these outrigger boards, I gave the machine a few tests, just moving the gantry back and forth. Those tests were extremely disappointing, but access was so limited that I was powerless to do much about it. Heck, half the problem might have been just an uneven base.

I decided I had to do something drastic about this bass ackwards table approach. I attached clothesline pullies to the machine and hoisted it up to the rafters of the shed. I then gutted the shed and it's workbench out from under the dangling machine.

Once the shed was clear, I built a new, very very level and flat table for the machine. The new table is 6x3, and only 30 inches tall. Once it was finished, I lowered the machine into place. It went very smoothly.

BOY I wish I had some pictures of this process, but it was about 3/4 of the way through it that I found the camera.

Here's the machine as it stands in my shed right now.

The new table is great. I can get to all sides of the machine, and it's work surface will be at an ideal height.

In that last photos, you can see my stepper controller board on the pegboard on the side wall. That's a "mechatronics" board, and for a time was sold on eBay by a dude called autohec. I don't think he sells them anymore. It served me well enough with my first CNC machine. I used Cat-5 cable to wire the motors, which are Keling technologies 425 oz/in motors. Take a good look at how those cables are haphazardly draped, because they're pretty important in a minute.

So, with my nearly finished machine now standing proudly on it's dedicated table, I set out to troubleshoot my machine and get it going smoothly.

It's frustrating: the X axis moves well only in the middle 2 feet of the travel, and then only up to about 80 ipm. Within about a foot of either end I have to decrease the ipms down to about 20 ipm to reach the limits of travel, and even then it's struggling. The Y axis travels smoothly throughout it's range, but it tops out at about 40 ipm reliably. Frustrating! Patrick has claimed CUTTING speeds @ 80, and rapids much higher.

I had really hoped the new table would clear up the X movement problems, but it didn't help at all. So I set out to map all of the measurements and parallelled the guidance rails and nothing I did made any difference. As might be expected, as I ran out of options I began to get extremely frustrated.

With frustration comes carelessness and, right on cue, I blundered through those white cat 5 cables I told you about, ripping one of them right out of the board, which fried it.

Coincidentally, a thread popped up over in Keling Electronics about a similar situation involving a mechatronics board. (http://www.cnczone.com/forums/showthread.php?t=58195) More importantly, the main topic is just how improved his machine performance is after he replaced his mechatronics drivers with Kelings.

So that's where I am right now. Last night I bolted 3 Keling drivers and a breakout board into a unused computer chassis and wired them up with a beefy 36v/9a power supply.

Tonight I hope to get the motor cables (and an Igus cable carrier- I do not care to repeat that dumb mistake!) installed and wired up and maybe I'll have some great news to report soon!

Cross your fingers! -Jim

How about some pics. I need to mount my drivers but haven't got around to it yet.

Brian

Hi Sonex,

I mounted the Keling drivers in an otherwise abandoned minitower PC case. The breakout board and the drivers are bolted where the motherboard would have gone (I drilled new holes for them) and the power supply lives in the top chamber, where the PC power supply and drives would have been.

Don't try to wire your rig with these pictures- use a diagram from keling or watch patricks videos. I was cheap and didn't color code my interconnecting wires. You'll get lost real quick. The gray wire hanging out goes to the adjacent PC (running Mach) and supplies 5v from a USB connector.

Amazingly, I can't test it yet. The DB25 connector on this breakout board is female, the one on my old board is male. I'll have to splice another cable or get a gender changer, or buy a new one.

The Igus cable carrier is sweet.

Cheers, Jim

Amazingly, I can't test it yet. The DB25 connector on this breakout board is female, the one on my old board is male. I'll have to splice another cable or get a gender changer, or buy a new one.
.


Jim,
Please don't waste your money on a cable. I have some extras & can drop one in the mail to you on Monday. They are DB25 male both ends like you need, the same cable I'm using to connect my Keling BOB to a PC.

Gordon

Thanks Gordon!

I may take you up on that offer! Let me check my office first- I think I have a M/M db25 at work, but I know I've got a gender changer, it's just a matter of finding it. It's just my work is 40 miles away, so I'll check tomorrow, when I'm there for a slightly higher purpose.

Sigh, so close, and yet so far.

Cheers! -Jim

I don't see any connections on the breakout board terminals for limit and e-stop. When I was testing mine, I jumped those to the 5V connectors next to each one to be sure there was a signal. Be sure to check the NO/NC or active high active low settings in Mach for both of these.

On mine I decided that limit switches and e-stop should be normally closed so a disconnected wire harness connection would stop the machine.

Is enable connected as well? I can't make out if there are two power leads or three.

I'll bet you get this running even without the cable. If not, you'll have the cable before the weekend.

Hey, I'll give that a whirl. I didn't hook up an Estop, but I don't have one configured in Mach, either. I also like configuring all my functioning CNC machines with limit switches that are NC configured. My approach is simple first, then accessorize. Maybe that E-stop has to be used in hardware.

The EN is jumped to 5v+ with a bare wire that is completely invisible in the photos.

Thanks for the cable! Really appreciate it. I found my gender changer and stuck it in my pocket and walked around all day masquarading as a woman. Either I don't understand how it works or I'm just not doing it right because every time I checked, I was still a dude. When I got home, it didn't work on the electronics, either. Maybe it's faulty. :)

Anyhoo, happy hump day! -Jim

You're a braver man than me if you take something called a gender changer and put it in your pocket!

Here are some camera phone pix of my controller, you can barely see the two jumpers, both on 2-3 but can see the jumpers for e-stop and limit switch inputs

http://farm4.static.flickr.com/3069/2718161589_d4cb20c1a7_o.jpg

http://farm4.static.flickr.com/3289/2718137079_bc6639887a_o.jpg

What cable is that you are useing on the controler's? I like the nice neat connections. What connectors are you useing on the motors?

Thanks

Jim

What cable is that you are useing on the controler's? I like the nice neat connections. What connectors are you useing on the motors?

Thanks

Jim

On my controller, I used wire stripped out of CAT5 network cable (22 or 24 AWG solid wire in twisted pairs) for the low-current connections from the break out board to the step/direction inputs on the drivers.

All power connections are made with 18AWG stranded hookup wire from Radio Shack.

The cables to each motor are shielded 18AWG stranded with a high strand count for added flexibility. The shields are connected to ground in the controller and float at the motor end.

I didn't use connectors at the motor end of these cables, I just soldered them to the pigtails coming off each motor. The controller end of the motor wires just use a barrier strip for now. I may replace the barrier strip with 4 conductor locking microphone connectors but I haven't had to disconnect the cables much. I have the connectors so might as well use them.

On my next build, I'll add connectors at the motor end as well just to ease routing cable through the wire management e-chain.

Connectors get expensive quick. I wanted to use DB9 connectors since they're failrly cheap ($1 each) but had to use microphone connectors ($5 each) since I used 18 gauge wire. The DB connectors generally available are for 20-24 gauge wire. If you use 12 connectors on three motors (2 on each end of each of the three cables) the costs add up quickly. I've seen some nice builds that used barrier strips on the motors instead of connectors. I'd do that if I could come up with some neater terminal strips, 4 conductor with a cover and built in strain relief.

On my next build I'm not using 18 gauge shielded wire to connect the motors. I thought 18 AWG was needed because the motors are rated at 3 amps. After doing some research I realized that the wires never see a continuous 3 amp current so the 18 AWG conductors are overkill. Next time, I'll use 22 gauge hi flex shielded cable. This will expand my options for lower-cost connectors.

Color coding the wires is critical. This will make troubleshooting far easier. Another tip is to assemble and wire everything on a test board before building your controller box. It's natural to want to get things working quickly and hard to understand the best compnent layout when you're just starting your build. I started with a piece of pegboard and wire ties to hold things down. Once it was working I played around with the layout and then mounted the components to 1/8" aluminum sheets and fixed these to the box. It was important to me to separate AC, DC, and signal wires and to keep things neat to make troubleshooting easier.

I think he did do the pre-cut guides on his hand-built prototype, but because this is machined I am sure he thought he might be able to get away without yet another step.
The current version I think is milled all at once (i.e. not changing tools) but doing the pre-cut guides might necessitate a different bit size. That could be problematic. It will be intersting to see what he says! ... Great notes on your build. Thanks for documenting it for us all!

Back in the '70's, a computer company SWTP from San Antonio TX, (Southwest Technical Products) used Molex connectors that I think are still easily available at Radio Shack (I hate to recommend anything from there since their part selection is SO limited anymore, but that is where I have seen them recently).

They would be reasonable for motor and signal connectors. I would use different ones for signal and power, with different pin configurations just to keep things from being accidentally messed up. In these connectors, the pins were soldered and crimped, then inserted in the connector. Each pin had a male and female version, and most of the connectors are polarized to keep from inserting them 'incorrectly'.

http://www.radioshack.com/family/index.jsp?categoryId=2032286&cp=2032058.2032231

But they are available at http://www.Mouser.com , http://www.Digikey.com , and http://www.Newark.com also. http://www.Molex.com is the manufacturer and has details on their web site if you can wade through it.

As far as connectors go, these are fairly inexpensive, but are not totally enclosed with hoods, etc. But many can be mounted in boxes easily because of flanges on their side to either snap or bolt in.

gfc62,

nice to see someone on here that is local to me. I live in Springfield off the fairfax county parkway. I plan on getting some of the HI-Flex wire so i can wire up my Solsylva machine.

Jim

I work in Springfield, just inside the beltway. Jim here (gibbonsrock) is sort of local as well.

Are you also in the CAMS CNC group? Are you planning to attend the meeting tonight?

(http://www.cams-club.org)

I'm up here in Point of Rocks in MD, just over the bridge at rt 15. If you call that local to you guys, you're being awfully generous!

I didn't make it tonight's meeting, unfortunately, (couldn't book a flight :) but I'm looking forward to meeting some other CNCers sooner than later. I would also feel a lot better if I had an actual, bona fide, operational CNC machine before I delved into the playground of the seasoned.

Cheeers! Jim (the other Jim)

gfc62,

Where in springfield do you work? I work on industrial road at Washington Gas.

GibbsonsRock,

I didn't make the meeting last night either. (didn't know about it) I have a 7 month old that needed my attention last night.

gfc62,

Where in springfield do you work? I work on industrial road at Washington Gas.



That's funny, I work a block away from you on Commercial Dr, the Gannett Offset building.

Cool. I think i know where that is. Good to know.

Hi Guys!

I bet a few of you are wondering what the heck is going on in my shed these days. Sorry to leave you hanging there.

Long story short, I managed to get my Keling electronics fired up (big hat tip goes to Gordon, without whose help my motors would probably still be motionless, and I would be limited to running my electronics from the same zip code as the computer. You should see this db25 cable he sent me! :cheers:)

Anyhoo, the results remained firmly grounded in the territory of disappointment, which was the final straw for me: I dismantled the machine and began construction of something completely different.

I sent an email to Patrick about the situation, and after a few very friendly conversations we have come to a mutually beneficial and amicable solution. (thanks Patrick!)

As I pointed out earlier in the build, the kit I bought was one of the first ones he produced, and it was largely experimental. Patrick and I communicated extensively over the last few months about the continuous evolution of the machine, and he continuously improved his machine over time. I didn't keep up.

Patrick assures me that a lot has changed with his machine, and just as much has changed with his enterprise- I believe he's in the process of moving into a new house (and bona-fide workshop) potentially even as I write this. In other words: he's going pro.

Therefore, I strongly urge readers of this build log to reserve judgment of Patrick Hood-Daniels new 1.3 kit based on my experience- the new design is a very different beast.

It's important to note that the PHD kit is great for the casual user who wants to do basic woodworking and signmaking, but I've been trying to turn it into a Shopbot. I need a very precise machine that will carve guitar bodies, necks, and even do intricate inlays.

So anyway, I'm going to close down this build log with a positive note: For all the frustration involved, I really enjoyed working with Patrick on the modular kit, so we are going to collaborate on another CNC project. The new machine will be much smaller, but much more capable. (and that's all I'm going to say about that for the moment, or at least until we get a working prototype going.)

Feel free to PM me if you have questions about my experience, but I think most questions are best directed at Patrick himself.

Cheers, Jim

gfc62,

Is this the cat5 cable? I don't recognise the color code.

Jim

gfc62,

Is this the cat5 cable? I don't recognise the color code.

Jim

No, that's the 18 AWG shielded high-flex cable. That's the output to each motor

The twisted pair is only used on the step/direction signals from the break out board to each Keling driver. Those are the standard colors (blue, orange, green, brown, slate). These are the thinner wires connecting to the drivers below your circles. The CAT5 wires have their outer sheaths removed and I'm using only individual pairs.

Note that I took a shortcut here. I'm using one pair of wires for step and direction on each axis. Better would be to use one pair for step and a different twisted pair for direction on each axis, connectiong the unused wire in each pair to a common voltage level. As I've wired it there's potential for the step and direction signals to crosstalk on the twisted pair. I haven't seen any problems but that doesn't mean my approach should be followed by others.

When I started wiring this I was concerned about common grounds and wasn't sure what would be the best approach so I wired it as closely to the Keling diagram as possible (considering the somewhat confusing nature of that document)

In my personal opinion the Keling diagram should be re-done. First it should show three drives connected. Second, I'd suggest connecting the drives in a way that makes use of the common terminals on the break out board to power each driver and eliminates the daisy chain connection of +5V from driver to driver that Patrick shows in his video. This would involve connecting each driver with 4 wires from the break out board. This could be done with two twisted pairs for each axis. For example, the blue/white wire could connect to the X axis step signal and the white blue wire to the common terminal.

ok. i can't find the 18awg shielded high-flex cable. where did you get it from? i have looked at home depot, radi shack and fishers hardware with no luck.

Jim

I ordered mine online. The part number was Electro-Matic EMBZPUR185DC24 and it was expensive, I'll have to look up the price but I probably shouldn't remind myself. I recall it being a bit under $100 with shipping.

I think 18 gauge is overkill and I'm pretty sure high-flex is overkill as well unless you plan to run your machine 24/7. Just make sure to use stranded cable conductors not solid conductors and look for cable with the highest number of stands per conductor.

If you want to buy local you can try that electronic supply place Arcade Electronics off Gen Washington Dr (on the east side of 395) They didn't stock high-flex cable but had some high-strand count cable that would have been just as good for something like .40 or .50 per foot.

Does Home Depot have any shielded cable at all? I thought I saw some 20 or 22 AWG 4 conductor with shield there once. I didn't buy it for my machine because I thought the conductors were too small. If I had to do it over I'd go with 20 or 22 gauge instead of 18.

I just picked up 35 feet of 22 awg sheilded at Archade for $17. it looks like it will do the trick. thanks for the tip.

Jim