The NRZ N2 Waterblock Beta Thread

[thenrz]

Hey everyone,

As a person that works around CNC machines quite a bit in both a hands on and CAM oriented capacity, and as one who as accrued a high level of comfort in doing so, I have finally set the cogs in motion to fab up my own line of waterblocks. The first non alpha prototype is drawn, materials gathered, tools purchased, and machine time secured (for the most part, it certainly has me moving that much faster during the days/weeks to make sure I have a free space for a vice on the table.) I will be using a machined copper base with 0.018" w/0.015" h (0.020" if I can get there) channels, a stainless steel direction/pressure plate which can be swapped out for different CPUs over time and different relief heights for different types of setups and pump pressures, acrylic cap, and a stainless steel mounting system. The acrylic top has symmetrical G1/4 barb interfaces and features enhanced flow lines to shape the inlet liquid across the entire surface of the copper conduction zone to maximize efficient use of available coolant volume. Deck height of the pressure plate above the conduction zone is tight; this block is made for use with high flow pumps as there is no dead zones above the copper. Water is routed to direct contact with the surface area of the channels and then quickly out to dump that heat energy into the air.

I want to tap into this collective treasure trove of experience with the finer aspects of high performance computing and make some choices here based off something real and not what looks good on paper.

1. I will be targeting AM4 first. An Intel bracket won't be much to do after, but I want to focus on one thing at a time. Do you guys trust the stock backplates and the hook screws or is it better to just go with my original drawing and machine the full mounting system? My current mount is a hybrid option... it has the 54x90mm on center holes and a crossbar, tab, and hole for a thumbscrew with hook bolt. A custom stainless backplate is also drawn. The hybrid mount won't be in the attached screen shots. Took those a few days ago, and I'm too lazy to go boot up my dev rig at the moment... 13 hour day at work...

2. What is the ideal amount of contact pressure between block and IHS? I am using a machined stainless steel mount, I know that pressure will be directed towards the center of mass since there will be no flex and the resultant misdirection of force. Will including springs and stops on the mounting screws be beneficial enough to outweigh the additional cost?

3. How important are aesthetic features... really? If you were to buy a block today, would you look at performance first, aesthetics first, or a blend of the two? RGB sells, it's as simple as that. Obviously that would come later, but how important is it to your purchase?

4. My design can come apart completely for cleaning, pressure plate change out, gasket swaps, etc. Is this a feature that is desirable? Or does it add complexity that a majority of people won't want to deal with? Should some come sealed?


Attached are a few lower res perspective shots. I don't have a Visualize license right now because the new guy at work is using it... Too bad, it looks nice that way. The block is a modular sandwich design. The gaskets and threads are not in this mock-up. I will use a threading cycle on the machine/CAM software so there's not need to suffer though Solidworks's threading feature. Gaskets will be assembled out of 0.090" cordstock. There will be three of them total within the block. Two on the inlet and outlet interface between acrylic and pressure plate, and one between the pressure plate and the cold plate. (The acrylic is all one piece, and is not split as in the photos. I forgot to merge at one point and forgot to change it.) The 0.701" thick acrylic comes this week. I was able to score a 12"x24" sample from a vendor contact. Using stainless and copper cutoffs around the shop. A total material cornucopia where I work.


Total overall height: 1.045"
Total overall width: 2.200" (block) 3.885" (mount)
Total overall depth: 2.200" (block) 2.500" (mount)


I'd like to iterate this into something that performs with the best. Eventually I'll expand my designs to things other than CPU blocks, but for now I need to get my feet wet. I work in the mold and model industry, so I am so used to working with IPS (inches, pounds, seconds) that I don't deal well with the much more logical metric system when I'm drafting.

Thanks for looking everyone. Criticism, constructive preferred, welcome and desired. I will update as I get everything going and have some parts in my hands. Small details, radius finishes on some of the remaining edges that are drawn sharp, will come in time....

 

[Blaylock]

Welcome to Overclockers.com NRZ.

Here are my thoughts...

1. I will be targeting AM4 first. An Intel bracket won't be much to do after, but I want to focus on one thing at a time. Do you guys trust the stock backplates and the hook screws or is it better to just go with my original drawing and machine the full mounting system? My current mount is a hybrid option... it has the 54x90mm on center holes and a crossbar, tab, and hole for a thumbscrew with hook bolt. A custom stainless backplate is also drawn. The hybrid mount won't be in the attached screen shots. Took those a few days ago, and I'm too lazy to go boot up my dev rig at the moment... 13 hour day at work... I thing using the stock back plate is fairly standard in the industry. Keep in mind Intel uses a different thread than AMD when you get to that crossroad. If you are referring to the plastic hooks for mounting, I don't know of any other manufacturer that utilizes those, even for air coolers. I think each socket changes slightly and its easier/cheaper to just match the bolt through pattern.

2. What is the ideal amount of contact pressure between block and IHS? I am using a machined stainless steel mount, I know that pressure will be directed towards the center of mass since there will be no flex and the resultant misdirection of force. Will including springs and stops on the mounting screws be beneficial enough to outweigh the additional cost? I'm not sure what the pressure is supposed to be but I'm sure AMD and Intel have posted these specs somewhere. You might even be able to contact them as a future manufacturer and get this info direct. All the water blocks I've worked with use springs to apply even pressure. I did have an older air cooler that had a spring steel mounting plate that did the same function and lasted way longer than any spring will. Just a thought for you.

3. How important are aesthetic features... really? If you were to buy a block today, would you look at performance first, aesthetics first, or a blend of the two? RGB sells, it's as simple as that. Obviously that would come later, but how important is it to your purchase? In todays world aesthetics are everything. Manufacturers are adding RGB to their mouse pads for crying out loud. It's gotten out of hand IMHO but that stuff sells. For my money performance is boss, but I'm afraid I'm in the minority here. In truth there's no reason you can't provide both with proper engineering.

4. My design can come apart completely for cleaning, pressure plate change out, gasket swaps, etc. Is this a feature that is desirable? Or does it add complexity that a majority of people won't want to deal with? Should some come sealed? The ability to clean and replace worn gaskets/o-rings is critical for a custom waterblock, not for an AIO system. You mention your block will use three different gaskets if I remember correctly. That seems high. My older gen XSPC block uses two o-rings and my current EKWB only has one. As you know every seal is a potential leak point. I'm not suggesting that I wouldn't buy your product because of it, I am saying that if you sell a lot of them you have the potential for a higher fail rate, which will affect the future of your business. That should be a concern.

These are just my initial opinions so please take them with a grain of salt. I truly hope you are successful with this future venture. If and when you begin production runs, feel free to PM me and I would gladly test and review your waterblock here.

Best of luck to you.

 

[thenrz]

Blaylock, thank you so much for your input. I don't have the time to construct the proper response, but I wanted to say thanks. You bring a lot of good points to the table, and I take from your response that you have a ton of experience in these matters. Once I have a chance tonight, I will continue the dialog.

First toolpaths created. Quick photo attached. I'm not so familiar with machining copper alloys, so I'm taking really small stepdown/stepovers for now. That will change in the future once I become more well versed in a new material.

 

[trents]

A couple of suggestions for consideration:

1. I'm not sure if they still do this with their most recent CPUs, but for years the IHS on Intel processors was not flat but cupped. It was lower in the center than it was at the corners and sides. When you went to lap them this became evident. And many air cooler cold plates were manufactured with a convex shape to accommodate. A good example was the Thermalright 120 Ultra Extreme. Not sure if AMD does that too. But my question is does you cold plate design compensate for this or should it?
2. Does your design compensate for the asymmetrical placement of the Ryzen 2 chiplets under the CPU IHS? They are not in the center of the die but off to one end.
3. I prefer the post and screw type mounting system using a back plate over the hook and screw type. The hook and screw types are usually awkward when it comes to getting everything lined up and in place (and kept in place). And by the time you finally get all that done and are ready to tighten the screws the block has done a lot of slipping and sliding on the IHS and you always wonder if the thermal paste coverage has gotten screwed up.

 

[BugFreak]

Nice looking design there. It looks like your design skills are way above mine so I'll leave that input to you and other more experienced people. I will add some input to your question about looks over performance though. These days everyone wants some obnoxious lighting or wing or something so it will play a part. That being said, a nice clean water block like you have there should be plenty. Especially with the stainless pieces you plan on using. Nothing like a nice shiney piece of stainless to stand out imo.

Good luck in your ventures.

 

[thenrz]

1. I will be targeting AM4 first. An Intel bracket won't be much to do after, but I want to focus on one thing at a time. Do you guys trust the stock backplates and the hook screws or is it better to just go with my original drawing and machine the full mounting system? My current mount is a hybrid option... it has the 54x90mm on center holes and a crossbar, tab, and hole for a thumbscrew with hook bolt. A custom stainless backplate is also drawn. The hybrid mount won't be in the attached screen shots. Took those a few days ago, and I'm too lazy to go boot up my dev rig at the moment... 13 hour day at work... I thing using the stock back plate is fairly standard in the industry. Keep in mind Intel uses a different thread than AMD when you get to that crossroad. If you are referring to the plastic hooks for mounting, I don't know of any other manufacturer that utilizes those, even for air coolers. I think each socket changes slightly and its easier/cheaper to just match the bolt through pattern. I installed an AIO cooler in a couple of work computers I put together, and they used a metal hook bolt and thumbscrew. It was a convenient idea, but you are correct. Especially carrying the additional mass from the heavier metals that comprise my design, bolting to the backplate is really the only way to go.

2. What is the ideal amount of contact pressure between block and IHS? I am using a machined stainless steel mount, I know that pressure will be directed towards the center of mass since there will be no flex and the resultant misdirection of force. Will including springs and stops on the mounting screws be beneficial enough to outweigh the additional cost? I'm not sure what the pressure is supposed to be but I'm sure AMD and Intel have posted these specs somewhere. You might even be able to contact them as a future manufacturer and get this info direct. All the water blocks I've worked with use springs to apply even pressure. I did have an older air cooler that had a spring steel mounting plate that did the same function and lasted way longer than any spring will. Just a thought for you. A good piece of advice to contact them directly, and I have done just that. I am thinking that the general end user experience with a mount is something I will have to iterate through. I have modified the mount itself with relief for a hex bolthead and it is extruded down, in a cylindrical form, to reduce the height of the bottom of the mount to the motherboard. I will end up getting a lot of different sizes and tensions of springs to gradually approach the ideal hardware.

3. How important are aesthetic features... really? If you were to buy a block today, would you look at performance first, aesthetics first, or a blend of the two? RGB sells, it's as simple as that. Obviously that would come later, but how important is it to your purchase? In todays world aesthetics are everything. Manufacturers are adding RGB to their mouse pads for crying out loud. It's gotten out of hand IMHO but that stuff sells. For my money performance is boss, but I'm afraid I'm in the minority here. In truth there's no reason you can't provide both with proper engineering. The design of the acrylic top was always to keep enough extra volume unused by both structure and performance to dedicate to an RGB system... Whether it wraps around the outside or is seated within a relief cut into the underside of the acrylic, both ways would work. I did some searching, and it would only add between $3-$8 to the overall cost of the build in terms of parts. Labor based on actual production would be a different story, but I am working this through in stages. Having hardware that functions at a high level is first and foremost my first stage goal. I think the stacked design will end up being really pleasing to the eye, and seeing the different metals through the clear top with pleasing lines and colors will have aesthetic appeal immediately, in a modern, industrial sort of manner.

4. My design can come apart completely for cleaning, pressure plate change out, gasket swaps, etc. Is this a feature that is desirable? Or does it add complexity that a majority of people won't want to deal with? Should some come sealed? The ability to clean and replace worn gaskets/o-rings is critical for a custom waterblock, not for an AIO system. You mention your block will use three different gaskets if I remember correctly. That seems high. My older gen XSPC block uses two o-rings and my current EKWB only has one. As you know every seal is a potential leak point. I'm not suggesting that I wouldn't buy your product because of it, I am saying that if you sell a lot of them you have the potential for a higher fail rate, which will affect the future of your business. That should be a concern. You are 100% correct that each additional possible point of failure works against me. With this design, and the ability to swap out directional/pressure plates based on application, has a three gasket minimum as the plate needs to seal to the cold plate and both inlet and outlet. This was the reason I went with rigid metals like stainless to hold the structure together. I was very careful to engineer the gasket interface surfaces and channels to be as robust as possible and also firmly hold the gasket in place when the unit is not assembled. I am using a softer Viton fluoroelastomer chemical resistant cordstock to fabricate the gaskets which will deform to fill the gasket channel under the pressure of assembly, so no concession on quality is made at this potential point of failure.

Great points. Got me thinking in a different thread, and that's exactly what I had hoped. Anything else you think of, firmly play the devil's advocate please!

A couple of suggestions for consideration:

1. I'm not sure if they still do this with their most recent CPUs, but for years the IHS on Intel processors was not flat but cupped. It was lower in the center than it was at the corners and sides. When you went to lap them this became evident. And many air cooler cold plates were manufactured with a convex shape to accommodate. A good example was the Thermalright 120 Ultra Extreme. Not sure if AMD does that too. But my question is does you cold plate design compensate for this or should it? This is something I am currently exploring. It wouldn't be hard to machine the contact surface into a convex surface... all it would take is additional time. Would need to drastically reduce the stepovers and use a ball mill as opposed to a couple of passes with an end mill. I see some of the really high end blocks are actually turned on the contact surface rather than milled. I have access to all that equipment, but I don't think the advantage of turning over milling will outweigh the inconvenience of having to find open CNC lathe time. This could change in the future as I hammer out the manufacturing processes and build more fixtures.
2. Does your design compensate for the asymmetrical placement of the Ryzen 2 chiplets under the CPU IHS? They are not in the center of the die but off to one end. Yes it does. This is the exact reason for the swappable stainless direction/pressure plates. The posted design rakes the liquid over the entire surface of the conduction zone and does not favor the physical center. I have a couple of different designs for this, and my prototypes will compare apples to apples in regards to cooling effectiveness on MCM CPU designs. (an aside, remember way back in the day when many of us scoffed at MCM topologies?)
3. I prefer the post and screw type mounting system using a back plate over the hook and screw type. The hook and screw types are usually awkward when it comes to getting everything lined up and in place (and kept in place). And by the time you finally get all that done and are ready to tighten the screws the block has done a lot of slipping and sliding on the IHS and you always wonder if the thermal paste coverage has gotten screwed up. Agreed. I think I am trying to appeal to the least common denominator. too much.. it's an artifact from working in electronics manufacturing as a product designer. I was tasked with making the product appeal to as many as possible. Custom liquid cooling is a niche (but a growing one), and my target audience is different than in my past.

Thank you so much for your input! You made some really valid points, and reiterated how important the convex contact surface really is. That's very high up on my list now.

Nice looking design there. It looks like your design skills are way above mine so I'll leave that input to you and other more experienced people. I will add some input to your question about looks over performance though. These days everyone wants some obnoxious lighting or wing or something so it will play a part. Yeah, you are certainly correct here. After the initial batch of working prototypes, I will start modifying the acrylic section to house an RGB strip. I thought it would be cool to backlight the logo, but it seems that side lighting the acrylic is what sells.That being said, a nice clean water block like you have there should be plenty. Especially with the stainless pieces you plan on using. Nothing like a nice shiney piece of stainless to stand out imo. That's what I'm hoping. It's kind of a different design, but it's engineered to be robust and easy to maintain on the same level as performing with the highest echelon products... and I really feel like it will have a truly unique and aesthetically appealing look and feel to it.

Good luck in your ventures. Thank you!

Thanks a ton for your input. All three of you so far have mentioned the importance of adhering to the market's infatuation with RGB, so to the top of the list this goes as well.


Wasn't able to get much done today. Had a couple of cast molds that required a lot of engineering and setup time on the CNC for revisions and feature milling. I was able to get some toolpaths down for most of side one of the coldplate, though. The stepdowns for the conduction cutouts on the bottom face are really laughable. Using non beryllium alloys at first to avoid the gumminess that plagues some copper machining... hoping for the best because I really don't know a lot about copper in fine detail. Thanks again for all your input, guys. Won't be forgotten.

 

[thenrz]

Had a chance to go through the 304 stainless, 6061-ish aluminum, and beryllium free copper cutoff pieces this morning, as well as talked to the plant manager about suppliers and his contacts. Obviously my target cost per unit will change over time if and when things go to a mode of manufacturing more akin to production rather than prototyping. With the acrylic stock that I'm getting for free this first time around, I should be able to cut enough pieces for 6 prototypes without having to scrounge for less desirable pieces. My total cost is surprisingly low since I am getting the materials at scrap cost. I want to try a mount out of stainless and aluminum (doesn't interact with the fluid, so no galvanic corrosion issues) and see how the rigidity holds up. I found some brass here, too, and I want to see how the pressure plate works with that as well as 304 stainless. Brass can be expensive, but it is also a lot easier to machine. I am really liking the options here, and having a lot of different possible configurations will allow me to test a lot of the permutations. I'm pretty much set on 304 for the inner plate, but the golden color of brass may end up having value in aesthetic appeal. Brass also sits next to copper, as does stainless, in the corrosion index meaning it won't react with copper as well as stainless does. We all know how aluminum and copper like to go for each other's throat.

If anyone has any materials expertise, I'm all ears. I am looking at the weekend (maybe not depending on some pending plans) or next week to start making chips. My half mil cutters and all my watercooling testing hardware starts showing up today into next week. Got some great deals on ebay on a res, pump, and rad... Only the reservoir isn't a high tier brand, and with a 1500lph pump/top and an 80+mm thick 240mm radiator, I didn't spend over $120. Scratch and dent for the win. Cannot wait to see some finished parts start to come together! I have an arduino test bed for temps pretty much set up already, just waiting on the thermocouple boards to show up from overseas... I didn't feel like building them this time around. Going to use power resistors at set wattage points to produce repeatable temps for testing so I will be able to see just how much each permutation and material combination affects overall performance. I will also be working with several different deck heights of the plate above the conduction zone to see what flow rates across the fins will best suit the design. I know that higher will be better with the designed restriction, but I'm sure that there's a desirable spot in the curve. My design is based off of maximizing dead zones where coolant run through the block without interacting in any sort of efficient manner with surface area at the bottom of the block.

Thanks again to everyone who has responded. You guys are great!


Photo: top, 6061 aluminum. left, 304 stainless. right, brass (unknown alloy), bottom, 3.25" diameter low/no beryllium copper (have to find out definites on this one)

 

[trents]

About the convex surface. Be careful with that in regards to how you handle the pressure exerted by your mounting hardware. I think Intel cupped their IHS in order to distribute most of the cold plate pressure to the outside walls of the IHS. Too much center-focused pressure can distort the CPU socket and create contact problems in the LGA grid.

 

[thenrz]

Ugh, just deleted my post...

I'm going to cut my bottom contact pad .010-.015 high dead flat and mount it up at recommended manufacturer's pressure. I will use a kind of spray that the mold fab guys here use to determine contact between parting surfaces. Then when I mount and unmount, I will be able to see exactly where and how much contact was achieved. Then I can try to indicate my block back in the CNC and machine a spherical/parabolic (?) Surface to match.

...all of this arduous work unless I can find the info!

 

[Blaylock]

We use Dykem here at work. You can find better prices than Amazon. This was just a quick Google-Fu.

 

[thenrz]

I have a can sitting next to my 10'x5' CNC. Good stuff for this application, but it's going to be really messy... I've saved the owner of this company so much money, he essentially gave me free reign on whatever I want to use...

 

[thenrz]

So the drawing is officially done for revision 3 of my N2 block. Mating parts are no longer drawn line on line and necessary relief has been added where necessary, but the parts will still assemble without really any gap at all.. Changed the gasket interface gaps to appropriate distances. Almost completely revised the layout of the coldplate. The stack's holes are now drawn to proper diameter to accept a 4-48 bolt with bare minimum relief as to properly align the parts as they are assembled from bottom up, and the top bracket drawn to accept the proper tap. The mount has been modified to accept the 6-32 standard AM4 mounting bolt and additional pads have been added for screw alignment help. The top threaded bracket has had caps added to increase thread engagement and fit the standard size bolt.

A "somewhat isometric" view.
View attachment 209043

I have a spreadsheet with all my material prices and vendors once my free lunch runs out from work. It really isn't bad... not nearly as bad as I had expected using top notch materials. Raw stock is reasonable, and tooling won't be terrible assuming I don't revert to machining in ways that greatly reduce tool life. I should be making chips for this project starting tomorrow even if it is in a limited capacity.

I am thinking about grabbing some LED strips and pods to put under the acrylic and on the pressure plate earlier than expected. I think the backlit logo would look killer... Anyway, worked 11 today and did this all evening. Tired and crabby now haha. Have a good night everyone.

 

[trents]

Any idea what your selling price will be?

 

[Blaylock]

I like what you have here. Curious to see how she performs. I think you are setting yourself up right with your testing platform as well.

If you could make the acrylic labeling customizable, I think you could set yourself apart from the competition. I don't know if anyone else that offers this, but I think that could be a big selling point.

 

[thenrz]

Shhhh, that was my idea all along. It would be a really solid up sell. There is going to be some sort of design or text around the top bracket, too. That'll come after the performance index is nailed down, though. I don't want to waste free machine time at this point. There are actually two machines right now that I could fit a vise in, and hopefully I can make use of that because that won't happen a lot.

And as far as price, I'm really not sure yet. My costs will change as my production size increases. I know I'll get machine time at rock bottom price, but there are stock materials, tooling, machine time, fixturing, finishing, assembly, accessory costs, packaging... What I won't do is charge an arm and a leg. That's not the point here. Of course I want to recoup my initial costs and eventually generate profit, but I'm more about offering something great for a reasonable cost. People pay high mark-ups on custom loop hardware as it is.

Thanks for the input as always guys!

 

[trents]

Do you have a marketing name for the product yet?

 

[thenrz]

Do you have a marketing name for the product yet?

Sort of I guess. I have had the three letters "nrz" in my handles, email, etc since og Starcraft pre-brood war. Feels right to attach that to this whole endeavor. The model "N2" is because this is the second time I've tried to go about this. The first time I wasn't skilled enough in pretty much every area that is required to pull all of this off; I had an idea and no way to get there. For now I'm going to stick with what I have. I really like the name "metalcraft", so I might end up calling it the "Metalcraft N2" block. Feedback on that would be helpful haha.


Well, I made some chips today, but hardly got anywhere. Only got a quick rough done. Had a customer decide he wanted a revision to his CNC mold two days before delivery. Seriously the whole thing was done including all the finishing work, texturing, framing... So I had to get that apart and onto the machine to hog out a fin that they realized at the end of the process that the section wouldn't get enough heat. Oh yeah, and I couldn't go bigger than a 1/4" end mill and the inside of the fin was ~5.5" long. Good thing it was back side and not cavity side... So all I finished was one side of the coldplate's roughing cycle...ugh... Anyway, I have the coldplate and the pressure plate completely programmed, and side one done on the acrylic top. I am cutting the entire first run of parts out of aluminum as a fitment model. I have access to more scrap aluminum than anyone could ever want, and it makes sense to utilize that before getting into the good stuff.

Have to go in super early tomorrow to get it done because I have to remove the vice asap to put on a chunk of billet. I can't take it off now since I have nothing (well, nothing easy...) to indicate with, and I don't want to start the whole recutting business on the first go. Anyway, some photos ahead. I am seriously hoping Autodesk will allow me to have a home license of Powermill; that would allow me to get all the rest of the programming done and get the fixturing laid out much more quickly which would allow me to use all my extra time at work to cut on the machine... that would put me way ahead! I have to come up with mounting fixtures for the mount and the top retainer as they're too thin to vice cut.

Anyway, I'm tired and scatterbrained at the moment. Work 9 then three hours on my stuff. These 12+ hour days most of the week are starting to wear me down.


View attachment 209067
cycle start...

View attachment 209068
side one rough is done... and that's about it.

View attachment 209069
another view of said completed rough

View attachment 209070
got the free acrylic sample. 0.701" thick and really hard and dense stuff. Haven't ever had acrylic this rigid before.



EDIT: The last of the hardware has been ordered. Gasket stock, assembly bolts, the world's tiniest end mills, drill bits, taps, a new tap wrench because I can't let the CNC do all the manly work, and specific glue for the gasket material. I have all the raw materials at work and can use all of their cutters that aren't under 1/16". I would say I'll have an aluminum site model prototype in the next 3-4 days and a full material prototype in 7-10.

 

[Blaylock]

Looking good. I will say you are certainly in a good position. Design and machine time are generally the biggest expenses in these projects and with you being the designer and having free machine time you are well ahead of the game. I'm a designer too and my shop has every tool and material I could ever want. My problem is it's a strong union shop and they can't/won't do anything without a work order. I've designed several versions of cases but haven't pulled the trigger on having any of them made yet. Maybe some day.

 

[thenrz]

The model/pattern/mold (rotational mold) industry is a fly by the seat of your pants sort of thing. We don't do production, so everything we do is a one off. We consistently have to machine things that are so tall that the cutters we are required to use are far too long and too small of a diameter, even in 3+2 or 5 axis machining. We get molds from the foundry here and have to indicate them in after benching with basically no parallel surfaces or any nice features on which to touch off. The aluminum and steel billet sizes we work with sometimes are pretty remarkable. So... doing my project is fantastic because it's all small stuff, no drafted walls, no weird undercuts, nothing difficult. I look forward to it honestly. The programming aspect of what I normally have to do is never easy or pretty. I like to craft perfect toolpaths, but when you're working with the inner cavity of a giant spa mold, there's no time to make it perfect... what isn't just get benched out. We have to compete with China now, and it is making time even more tight. In some instances, just the materials cost in our quote is what China can do shipped to the customer's door. But, their quiality is horrendous, and we routinely have to rework them before they even mold a part. I'm ranting now, but hey...

I have never even tried a case man, that's pretty impressive. What is your CAD suite of choice? I basically am stuck on Solidworks since every mold shop I have worked for has used it. It's the standard in this industry. What's funny is my degree is in computer science, but I couldn't stand coding all day; I have to work with my hands. Kind of fell into the industry. I could make more money doing something else, but I get paid well enough here that the percs of the job outweigh a couple more grand a year somewhere else. I have made a lot of things here for personal use, namely parts for my car, replacement this or that at home... Now that I'm getting pretty good on the CNC lathe, my ability to craft objects has gotten even more adept. There isn't a union going on here, but there doesn't need to be. This company has been around for 81 years, and the owner treats us very well. Most of the office staff and plant manager staff are really cool people who will do anything to help. Like, I got a $45 piece of high end acrylic for free, and I am getting copper, steel, and brass bar stock in the sizes I need for less than even the least expensive online resources and I don't have to pay shipping since they will just ship it with what materials the company needs already. It really is the ideal setup for someone who wants to have access to manufacturing machinery and resources for personal goals.

Anyway, I have the first side of the coldplate mockup done. I figured out how to program at home, so that's going to seriously speed this process up. I'll post a photo later, but for now I'm swamped here and need to knock some of this stuff out.

 

[Blaylock]

I started designing about 25 years ago on old school paper and pencil but have evolved with the industry. I've used Catia since V4M and Unigraphics (now called NX) since version 11 back when GM/EDS still owned it. My shop does a wide variety of things but, for what I'm involved in, is mostly prototype stamping dies. We used to do a lot more production dies but that work has flown overseas with the majority of the industry.