Horizontal router table: Bit above or below table?

[atgcpaul]

I asked this on another forum but, in my opinion, I didn't get enough concensus on the best way to proceed so I'll put this question up to you all.

I'm making a horizontal router table for some long crown molding. The ones I have seen (DIY and MLCS) have the bit rising
up from the table and the work piece passes over the bit. This seems fine when making raised panels because there is still a
lot of flat surface resting on the table. When the stock gets narrower, as in molding, there isn't much flat contact between
the table and the piece as the bit rises. Think about how your outfeed table on a jointer is higher than your infeed. That
doesn't happen with a horizontal router unless you shim the outfeed.

So I plan to have the bit above the table so the flat back of the stock always sits flat on the table. The bit is then lowered
incrementally to cut my full profile. I have two featherboards on the fence to keep the stock flat and to help counteract the
force of the bit wanting to throw it back out. I also have featherboards on the table to push the stock against the fence.
Finally, the router bit is shrouded by a small box for safety and chip collection is done from above rather than below. This is
how my planer does it and I think this is how most commercial molding machines are setup. I will also be able to feed stock
from right to left as I do with my regular router table rather than in reverse.

Kick back is my primary concern but I plan to lower the bit in small increments and I have featherboard help.

Am I missing something? Are there other reason why I don't want to have the bit above the work?

Thanks, Paul

[Greg.B]

I'm pretty sure you can NOT do this. This is kickback at its finest. It would work until something in the wood causes it to hang, and it will not be pretty. I think MLCS makes a table and indicates you should only use the bit under.

I do understand your problem however.

[Greg.B]

Here is a link to their product line and a video.

http://www.mlcswoodworking.com/shops...oriztab_anchor

[cgallery]

I also think having the wood trapped between the bit and the table would result in cannonball-like kickback.

[atgcpaul]

I also think having the wood trapped between the bit and the table would result in cannonball-like kickback.

I understand the piece would definitely be trapped between the bit and the
fence. How is this different when the bit is below the table and piece is
trapped between the table (and bit) and a featherboard pressing down from
above? Is it that the featherboard, in theory, has some give to it?

Paul

[cabinetman]

There are other considerations. With the bit above the stock, any deviation in the flatness of the stock will change the profile. On this procedure, having the bit below, a downforce will maintain the profile. Depending on the profile, there's no out for the stock if the bit is above other than running it through, or turning off the router. Whatever procedure you chose, if you doubt any part to be unsafe, rethink it through before starting. You should have sufficient infeed and outfeed support for the lengths of stock you're running.

Because of the size of some mouldings, large profiles are usually done on a shaper with a powerfeeder. Powerfeeders can be used on router tables and table saws.
.

[Popeye]

Rule number one in routing with a router table is don't run the work between the bit and the fence, what you are talking about is the same thing only horizontal. And that means as has already been mentioned kickback in the biggest way. Personally I'd run the bit vertically in a conventional router table. Pat

[Brian G]

Paul,

I agree with those that discourage having the bit above the stock.

{Edit: I re-read your post, and you mentioned shimming as I have done below}

Presumably, you'll be running multiple passes. To compensate for the stock removed, pass several inches of stock over the bit, and then stop. Make a shim (laminate, 1/8" hardboard, etc.) that will make up the difference in stock removed. The concept is similar to what you noted a jointer outfeed table does relative to the infeed.

Do this with each pass.

[cgallery]

I understand the piece would definitely be trapped between the bit and the
fence. How is this different when the bit is below the table and piece is
trapped between the table (and bit) and a featherboard pressing down from
above? Is it that the featherboard, in theory, has some give to it?

Paul

A further difference is that as the wood passes the bit, material is machined away. If kickback starts to occur, the wood will move back slightly, but then the bit will have no more wood to dig into because it will already have been machined away.

[atgcpaul]

A further difference is that as the wood passes the bit, material is machined away. If kickback starts to occur, the wood will move back slightly, but then the bit will have no more wood to dig into because it will already have been machined away.

I am definitely not going to tempt fate and do the above table bit
manuever. However, I'm still not getting how one is safer than the other if featherboards
are holding the piece down. During a kickback on the bit-above-table, the
wood will move back to a point where wood has been machined away, too.

With the bit-below-table setup, should I not use featherboards to hold down
the wood and only use handheld pushblocks?

Paul

[Thom2]

one HUGE issue that I haven't seen mentioned is that when you raise the bit above the stock, rotation of the bit is opposite. If you would try to feed the wood from the same point as you would with the bit below the table, you would indeed be creating a finger eating wall piercing pitching machine. No joke .. that setup WILL do damage to you or your shop.

That aside, with the bit below the table and featherboards in place, you'll have a "sense" of any variation in stock thickness before it gets to the bit as the featherboards will provide a sort of "feedback". Also, featherboards will flex, absorbing a lot of the energy of kickback as the stock tries to raise and push back.

Raising the bit above the work, even if you do feed from the opposite end, with featherboards in place, you're only sense of feedback is going to be when that stock binds to a point that the bit has no choice but to pitch it back at you. Even with featherboards in place, kickback in that scenario is going to be 'straight line' .. the stock cannot raise or lower, it is TRAPPED and has nowhere to go but one direction ... BACK AT YOU (or the wall, the window, whatever is in it's path).

NEVER EVER TRAP A WORKPIECE BETWEEN BIT AND FENCE

In this case, the router being horizontal, the fence is your tabletop. Please do yourself, us, your insurance company, your wife, your kids, and your pride and pocketbook a favor, DON'T USE THE BIT ABOVE THE STOCK

[crokett]

What Thom said. The bit above the stock is equivalent to climb cutting with a handheld router and is prone to kickback. I learned this lesson a few months after I got my first router when routing dados. I got lucky and only put a dent in the shop wall, no fingers were damaged.

[leehljp]

I am going to jump in with a different take below.

I have the "Below" horizontal router and use feather boards. But on long boards, (8 ft and longer) I still managed to get some irregular cuts due to vibration. I tried making the feather boards tighter but then it is too tight and I cannot not push the board through in a consistently smooth flow. It is very hard to get the balance between too tight, just right, and a smidgen too loose, which allows vibrations and irregular cuts. Placement of the feather boards are about 2 inches in front and behind the blade. I did not have a problem with keeping the boards flush with the fence.


AS to "above cuts/over the top of the board", It can work, and work fine, but it must be in a very controlled environment like a production system has. To do this, you will need a speed controlled motorized roller feeder system - very similar to the roller feeders on a thickness planer and on a thickness sander.

See below for links to how to handle this situation. However, for the home and occasional user, the below will be much more expensive than you probably want to deal with.

http://woodworker.com/cgi-bin/FULLPR...ller%20Feeders

http://woodworker.com/cgi-bin/FULLPR...ller%20Feeders

http://woodworker.com/cgi-bin/FULLPR...ller%20Feeders

http://woodworker.com/cgi-bin/FULLPR...ller%20Feeders

The Steel City Feeder at $549 is the cheapest.

[cabinetman]

AS to "above cuts/over the top of the board", It can work, and work fine, but it must be in a very controlled environment like a production system has. To do this, you will need a speed controlled motorized roller feeder system - very similar to the roller feeders on a thickness planer and on a thickness sander.

See below for links to how to handle this situation. However, for the home and occasional user, the below will be much more expensive than you probably want to deal with.

http://woodworker.com/cgi-bin/FULLPR...ller%20Feeders

http://woodworker.com/cgi-bin/FULLPR...ller%20Feeders

http://woodworker.com/cgi-bin/FULLPR...ller%20Feeders

http://woodworker.com/cgi-bin/FULLPR...ller%20Feeders

The Steel City Feeder at $549 is the cheapest.

In this particular example of rabbeting the 12" edge of a 4' board on a horizontal set-up, with the bit above the stock, the use of a power feeder would have drawbacks.

It couldn't be positioned above the bit, as there is a bit doing its thing.

I would be ineffective ahead of the bit, as the stock would pass the bit and have no feed or control after the bit.

It would be ineffective after the bit, as there would be no infeed control.

Considering how narrow the stock is, and the length of the roller assembly, there's not much contact with the stock. Using the same example, with the bit below the stock, the roller assembly could be placed more centered over the bit. This may produce a "skipping" due to the shortness of the stock and the space in the table allowing for the bit.
.

[leehljp]

In this particular example of rabbeting the 12" edge of a 4' board on a horizontal set-up, with the bit above the stock, the use of a power feeder would have drawbacks.

It couldn't be positioned above the bit, as there is a bit doing its thing.

I would be ineffective ahead of the bit, as the stock would pass the bit and have no feed or control after the bit.

It would be ineffective after the bit, as there would be no infeed control.

Considering how narrow the stock is, and the length of the roller assembly, there's not much contact with the stock. Using the same example, with the bit below the stock, the roller assembly could be placed more centered over the bit. This may produce a "skipping" due to the shortness of the stock and the space in the table allowing for the bit.
.

Well, I agree that you are right in what you wrote. However, anyone that knows how to set those up would also know that it would take 2 - and infeed and an outfeed. I just didn't want to frighten everyone with the $1000 plus unit. I should have mentioned this in the first post.

My dad had a couple of units in the late '80s and early '90s that he used for ripping on a 12 in radial arm saw. (The only one I have EVER seen on a radial arm saw. Worked well. But back then there was an expectation that people who knew what "add on" feed rollers were - they could "figure out" how to and where to custom mount them without having to have a written manual.

It is kind of like a customized car or chopped one - you get the basic material and make it yourself.

However, with duals feeders, it should work if my memory of my dad's feeders are still half way correct . AS to narrow moulding, it depends on the narrow moldings. I do not know the technicalities, but I think that if it has as much as 1 inch flat surface after the cut should suffice - as long as there are a couple of feather boards on the side to hold against the fence.

In the end, yes it can be done for certain situations and done safely, but it is not practical from a financial point of view IMO.