Mobile Delta Unisaw/Shaper Combo

[Ken A Irwin]

My latest project in progress, my full iron top mobile Delta Unisaw/Shaper mombo-combo dohicky. It's kinda a shop made tool/machine, right?

20200428_043538.jpg

[Mitch Fenneman]

That is awesome! What did you change the aluminum extrusion of the fence to? Is it some 80/20 or something like that?

[Ken A Irwin]

it's a Peachtree Uni-T fence rail extrusion, they're local so I bought a whole 8' section of it, which in retrospect ended up being more expensive since when I got home and measured it it was quite a bit short of 8'. They sell it off the shelf as 36" or 43" sections and also use the same rail for their router table fence. The thing I find most puzzling about it is that JessEm specifically makes an adapter kit to mount their stock guides on this extrusion but Peachtree does not carry JessEm and wasn't even aware such a kit existed.

It's kind of interesting that you brought up 8020, it's where this all came about actually. The original plan was to do the all cast iron, 50" rip table with a router table in it and support it with 8020, but by the time I priced it all out it just seemed cheaper and easier (and arguably better) to get a shaper and span the 2. By the time I bought the router table casting, a decent lift, a heavy duty router, fence and a truckload of 8020, the shaper started to look pretty cheap. The grander plan is to put a sliding table on both ends, probably the Grizzly ones, and those likely will be supported with 8020, though I won't know that for sure until I have one in hand. I also plan to convert the shaper to 3 phase with a VFD so I can drive it closer to router speeds when using router bits.

[Paul Haus]

I'm concerned with your statement "I also plan to convert the shaper to 3 phase with a VFD so I can drive it closer to router speeds when using router bits." Are you talking about overspeeding the spindle by roughly double? Shapers typically user larger diameter cutters so it becomes a matter of cuts per inch (CPI) that is hard to duplicate with smaller router bits using them in a shaper collet at a given feed rate. Things like materials, belts and bearings are selected to use within a given rpm range and unless you change everything (in possibly redesigning or remanufacturing all the affected components) you risk possible failure and injury.
Most shaper spindles have a maximum rpm limit due to design and materials, and that's short of the rpms a router is capable of achieving safely. I have heard of some shapers that could achieve those speeds but most are both special and expensive to purchase and maintain.
If you're talking about altering the speed of the spindle cartridge, personally I would not recommend it. My recommendation would be to keep the spindle speed maximum as the shaper designers designated and instead slow down your feed rate to closer match the CPI that matches what the shaper could achieve. The other option would be to buy or build a router table to use the tool that's actually designed to work at those speeds.
Don't mean to be a downer here, rather flagging a potential safety issue that could cause injury (if I'm interpreting your statement correctly).
Good luck and be safe.

[Ken A Irwin]

I'm concerned with your statement "I also plan to convert the shaper to 3 phase with a VFD so I can drive it closer to router speeds when using router bits." Are you talking about overspeeding the spindle by roughly double? Shapers typically user larger diameter cutters so it becomes a matter of cuts per inch (CPI) that is hard to duplicate with smaller router bits using them in a shaper collet at a given feed rate. Things like materials, belts and bearings are selected to use within a given rpm range and unless you change everything (in possibly redesigning or remanufacturing all the affected components) you risk possible failure and injury.
Most shaper spindles have a maximum rpm limit due to design and materials, and that's short of the rpms a router is capable of achieving safely. I have heard of some shapers that could achieve those speeds but most are both special and expensive to purchase and maintain.
If you're talking about altering the speed of the spindle cartridge, personally I would not recommend it. My recommendation would be to keep the spindle speed maximum as the shaper designers designated and instead slow down your feed rate to closer match the CPI that matches what the shaper could achieve. The other option would be to buy or build a router table to use the tool that's actually designed to work at those speeds.
Don't mean to be a downer here, rather flagging a potential safety issue that could cause injury (if I'm interpreting your statement correctly).
Good luck and be safe.

Ah, hmm well that's a lot to address, Everything you mention IS a concern if not traded off for something else, but yes on over speeding the spindle, but no on double, it's probably beyond the practical limits of a VFD/motor combination, 90Hz, give or take, is more realistic, so basically 15K. This sort of 3 phase over-driving is done on metal working machine retrofits all the time without issues, so that machines can be run in a motor design speed and variabled up or down from a set mechanical point from there. Pushed past 150% and you start getting into the limits of the cooling fan and motor bearings, though 120Hz is not that uncommon, but for me anyway it's probably too far for me to feel comfortable with. Over speeding beyond a point raises the RPM of the fan in the motor to the point that it can no longer move the air in the direction it's designed to do and starts spinning a static donut of air in a circle without actually moving any air in or out of that donut which creates instead of reduces heat and that is actually what causes the heat more so than the speed itself. Same sort of becomes true when you drop below 30 cycles or so. A standard 3 phase shaper motor with a VFD will comfortably create a variable speed shaper that runs from about 5000-15000 RPM using 30-90Hz when using the stock 10,000 RPM belt position. There are off the shelf motors, given that the shaper uses a standard NEMA mounting, from Baldor, Leeson, Dayton, etc, that are specifically designed for VFDs that take this into account as well as motors with different base RPMs at 60 cycles, or even the Striatech DC motors that Nova and Rikon use which will go way down to a crawl (50 RPM), but if the goal is 5000-15000 RPM (or 3500-10500 on the lower belt setting) a stock motor should work just fine.

So kind of one at a time, first off the belt, the Delta shaper uses a standard V-Belt which tend, especially in a less frequently used non-production application, to set, which in turn makes them introduce a fair a bit of vibration. Simply replacing V-belts with VX belts (toothed/notched) will reduce this vibration a lot, they are interchangeable, another option is the segmented belts, both these options are rated for higher speeds, and higher loads, transmit less vibration and generate less heat. The spindle cartridge bearings are engineered for production workloads under worst case scenario conditions (within design criteria) meaning significant workloads spinning multi-pound cutter heads on up to 1" spindles with extremely abrupt startup speeds, which are actually worse in capacitor start single phase motors to begin with, the a fore mentioned vibration prone V-belt drive and of course heat which will drive captive lubrication past it's limits. A 3 phase motor alone has much better startup characteristics since it doesn't have a capacitor kicking it's first cycle, add the programmibility of VFD to the equation controlling the startup speed and you eliminate that stress factor on the bearings. This is significant in that startup is where the bearing are probably under some of their worst strain, the lubricant is no where near operating temperature and is abruptly smacked into operational speeds with less than optimal lubrication.

Is it practical to spin the spindle at 15K with a 5" diameter, 3" high exterior door stick and cope stack that weighs several pounds and has an effective outer edge speed way higher than any practical router bit? Of course not, you wouldn't even drive that at the 10K high speed of the shaper as it was designed, you step it down to the low 7K speed, or in the case of a 3 phase/VFD shaper, some more carefully selected speed. Interestingly, the big 4" tall, 1" solid spindles, where you would be able to abuse those specs the most, are single speed at 10K RPM. Are you going to drive the spindle cartridge bearings to failure spinning a radically smaller, comparatively, router bit that weighs a few ounces? No you aren't, at least not in any typical home shop use you shouldn't. This gets into the 3rd factor of heat buildup, you are just never going to get there with the significantly lower loading and reduced duty cycles. I'd go so far as to say that a spindle cartridge spinning a mass a small fraction of it's design criteria even at time and a half of it's design speed should do this better and for longer than any universal motor router like a Porter Cable 7518 ever will, which is kinda the standard heavy router table router. Of the 5 speeds of the Porter Cable, 10K, 13K, 16K, 19K, 21K, the shaper spindle should be able to effectively cover the lower 3, give or take a few...ish. Which should be enough to get a similar finish result under most circumstances which is the whole reason to do this to begin with. Can a shaper spindle cartridge realistically and reliably cover those 2 top speeds, well maybe but no, because it's probably somewhat irrelevant since driving the motor at 120Hz isn't really all that practical, and even if you did, this is where your real bearing problems are going to be probably, the motor bearings are probably a bigger concern than the spindle bearings at that point.

This is a shaper and as such there is really no real reason to run any large diameter side cutting router bits in it, there is plenty of shaper/molder options for such things, what you'd run in it would be smaller, typically end cutting bits, the type of bits that in combination with the router collet spindle have less mass than a 1" spindle all by itself with no cutter head installed. The loading is again, practically insignificant in comparison between a high mass, high load, high inertia spindle/molder head and a router bit that will, more often than not, weigh less than the collet it sits in. So does this push the machine past it's design specs? Well on paper, perhaps, but in reality I don't really see how, not if you address the other major factors that would lead to failure, reduce startup stress through the programming of the VFD, reduce vibration through drive improvement, reduce heat buildup through duty cycle limitation, and reduce lateral load through reduction in mass. So while I get the concern, I think I am approaching this with realistic expectations. RPMs in and of themselves are not the absolute limitation, it's a whole bunch of other factors that are most easily represented as that RPM limitation which itself is not necessarily a specific design limit but a mechanical limit of the 2 speed step pulley and the specific 2 speeds chosen based on that criteria. A bearing spinning at 10K RPM is not running at it's absolute limit with all other design factors taken into account, that would be a recipe for catastrophic disaster, and that applies to the motor's 3450-3500 RPM as well.

Anyway, I've researched this quite a bit between the use of VFDs in machine tool retrofits, to realistic spindle capabilities of the Delta/Powermatic/Grizzly spindle cartridges which are all somewhat related to each other. Have even had conversations with Richard Andrews from AE-Tool on the subject who wrote the most detailed info on this I've been able to find which addresses this issue specifically: https://www.ae-tool.com/wp-content/u...ta-shapers.pdf

And if I go through all this and it feels unsafe, I still end up with a variable speed shaper with less than optimal surface finish, which again is the whole point to the exercise to run router bits at, at least something closer, to THEIR design criteria. I would like to be able to run a Ready2Route CNC fence on this and do dovetails and such on it which is again one of those things that looks excessive on the surface but when you look at what it can replace starts to look cheap. And I just don't know how effective that would be if I can't get the type of finish it requires. Again we'll see.

[Paul Haus]

Part of my concern is IIRC routers can run up to 24K rpms which is more than double the shaper max speed and you said you were planning on driving the shaper at router speeds.. I agree having a variable speed shaper would be a good thing, but even at 15K rpms, I'm still a little concerned about safety. Sounds like you've researched this to a great extent, so other than 'be safe', let us know how it works out.

[Ken A Irwin]

Part of my concern is IIRC routers can run up to 24K rpms which is more than double the shaper max speed and you said you were planning on driving the shaper at router speeds.. I agree having a variable speed shaper would be a good thing, but even at 15K rpms, I'm still a little concerned about safety. Sounds like you've researched this to a great extent, so other than 'be safe', let us know how it works out.

Well, I said I was planning on driving the shaper at closer to router speeds. I would never consider driving it at 2 and a half times it's factory speed, at least not without some very serious references that that is doable. As far as pushing it to 15K though, there is references to doing that and as someone who makes their living in engineering, I have to say that all things considered this would not have been a product that survived in commercial shops for 75 years if running a comparatively dinky router bit on a tight, relatively low hour cartridge at 150% of the speed occasionally of what a heavily loaded, commercially used pre-WWII machine with a technologically archaic motor has been often capable of doing continuously day in and day out for more than half a century or more. Given that this machine and the basic cartridge were designed in the 1930s when the only real way to make sure something definitely worked was just to over engineer by several orders of magnitude, that design was then copied, virtually unaltered, by it's competitors, and continues to be copied still, I'd say it's a pretty safe bet it's over-engineered enough to be pushed this relatively minor amount. And as I said, it's by no means without precedent, at least not at the levels I am considering.

And frankly I'd be way more concerned about the catastrophic failure of a 15lb router mounted upside down from it's design intention spinning at 20,000 rpm with the only thing separating you from the spinney bits is a thin piece of plastic or MDF than I am about that same bit running 3/4s of that speed encased in 400 lbs of steel and cast iron being held in with a bearing package that alone has as much mass as that entire router. I mean if we're going to get into what the designers of things originally intended, over speeding a shaper is no more contrary to the designers original intention than hanging a router upside down from what is basically the side panel of a piece of IKEA furniture, by screws meant to hold a glide plate and jamming wood into the bit. Which of these things is more likely to get you a Darwin award?

Anyway, enough about that, it's not even a path I've gone down yet and certainly not what I intended this thread to be about. When I get to that stage it can be a whole other topic of debate.

[Ken A Irwin]

This has all be done in stages as these things are, started out with just sort of blending the new wings into what was a fairly neglected Unisaw top and getting it all flat and consistent. for a while it was just the saw and a left and right wing (8" wing on the left and 12" on the right). This is where the top was at that point:
20200229_130030.jpg
Now I've added another 12" and another 8" wing and of course the shaper and I am in the process of getting the whole top blended to the level the first pieces were. It's all flat within about 5 thousandths per foot and you can barely feel the seams between pieces with your finger nail. Still a ways to go, the shaper table needs to be drilled to tie it into the Unifence rail, and I have enough spare Uniguard brackets to tie it all together across the back, so the whole table is not only bolted together but also tied together front and back between the Unifence and the Uniguard. How I am doing the Uniguard mounts is a whole thing unto itself. I've counter sunk the mounting bolts and reversed the way the mounts mount so that you can put an outfeed table flush against the back of the saw:
20200427_134414.jpg

[Ken A Irwin]

For reference this is what the shaper looked like when I got it, and the Unisaw wasn't much better and in some ways worse, though I don't have any pictures of it:
20200203073841273_L.JPG

[Ken A Irwin]

Ooo, my cam levers arrived, now swinging down the Uniguard is tool less.
20200430_125525.jpg

[Paul Haus]

Ken
Just curious, do you use things like a tenoning jig much (things that are cut vertically)? Back after I bought my cabinet saw I bought a Uniguard and it always seemed to be in the way, I did furniture restoration/repair on the side and made a lot of furniture parts. Seemed like some days I spent more time fussing with getting the guard out of the way than working. I'm the first to admit the actual blade guard worked well especially on cuts close to the rip fence along with a nicely designed mounting method. Lacking dust collection was also a bit of a negative for me.
I finally ended up saving the mounts and blade guard and scrapping the rest. I designed and made my own overhead arm system that got around most every issue I ran into with the Uniguard and it's been working out pretty good all these years. I even made it with replaceable blade guards so I could get dust control along with the ability to swing the arm out of the way when needed rather than having to remove the Uniguard when it conflicted with what I was doing. The only area I wish I could improve is the lift mechanism, which I made out of 1/4" steel and it works OK. I wish it could have been more substantial but without a mill it wasn't in the cards. I also made a smaller version for the shaper that though I don't use all the time, I do use when the situation permits.
Not suggesting you do this, rather if it doesn't work out then see what can be done to improve on it as far as the guard goes. It can become a very good system with some modification IMO.