Jessem stock roller safety

Help Support

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.


Established Member
3 Dec 2020
Reaction score
I have a Wadkin AGS10 table saw and am trying to improve the accuracy and consistency of the results ... while remaining intact. I’ve upgraded the fence to the Incra TSLS and added the Jessem clear cut TS stock guide rollers and also improved my infeed and outfeed setups, and I’m seeing the benefit.

I’ve also been trying to understand the safety issues and have been reading the UK HSE info and also the very helpful guidance from Suva, the Swiss safety authority
Using a circular saw-bench

As a private individual I am free to follow or ignore HSE rules but my view is that they’re generally there for a reason.

So for those of you who don’t know the Jessem Clear Cut TS rollers, the idea is that you fit two of them, one before and one after the blade and they push the workpiece downwards and (because they have 5 deg skew) rightwards into the fence. Just to be clear the roller that sits beyond the blade has to sit on the right side of it so it is pushing into the fence not into the blade. They don’t just add accuracy but they protect against kickback and ejection (because the rollers roll only one way).

They seem to work a treat in conjunction with the Incra fence.

But according to the HSE and Suva guides I’m doing it all wrong: the Incra fence is full length with no longitudinal adjustment, so I ought to be using an auxiliary fence that only goes as far as the cutting point on the blade. But if I did this then the Jessem rollers, or at least the one beyond the blade, would be useless, indeed dangerous. And I like the full length fence I’m using, I’ve set it up dead true to the miter tracks; if I used an auxiliary fence I’d have to be jolly sure it was true, otherwise it might be angling either the workpiece or the offcut into the blade.

So my first question is:

1) do you guys think that Jessem rollers are a sufficient safety feature that it is acceptable to use a full length fence?

Next, I want to replace the crown guard with an overhead unit; this would allow me to use a riving knife that doesn’t project above the blade (I’ve bought one of the Suva scimitar shaped ones). Overhead units like the Suva S91 seem grotesquely expensive so I’m going to build my own. It seems that the crown guard has three functions: (a) to avoid leaving the blade exposed; (b) to be a physical block to kickback; and (c) to act as a dust extraction point. The commercially available overhead extractors seem to be designed to exert downwards force on the workpiece as they glide over it, providing function (b), but I would have thought that the Jessem rollers do this job. So my second question is

2) Is it OK to build an overhead crown guard that covers the blade and provides dust extraction, but doesn’t have a massively beefy system to hold it, as long as it can be held hovering just above the workpiece?

Finally, the HSE seems to permit table saws to be used for part depth cuts for rebating and grooving as long as they are suitably guarded, eg, with a Shaw guard, ideally to the standard specified by BS EN ISO 13857:2008. Well I have no idea what that standard is. So my third and final question is

3) Do the Jessem rollers permit me to do part depth cuts, or do I need to add additional protection? What form should that protection take?

Thanks for bearing with me if you’ve read this far!
I think many products with a US origin presume people have done a number of daft to things, like removed the riving knife, not used a sub-fence, no crown guard etc... they have a different attitude to safety to put it mildly. There may not be that many people using th

I can see the Jessem rollers being great for narrow stock, but further away from the blade they are holding the edge of the piece and not the bit near the blade that could lift up etc, though that’s the bit you want to keep your hands well away from anyway, so maybe no difference

A riving knife just blow the height of the blade is fine, the guard can sit above it, so long as it’s strong enough to catch the bits breaking off the workpiece

IMO if you're cutting ply or MDF then the full length fence isn't an issue and you can use both rollers.
If you're ripping proper wood then you should be using the short fence and you don't want the second roller.
Also for cutting real wood you could make a short fence to attach to the Incra and also fit T Track so you can use your Jessem Clear cut rollers.
Whether the Jessem rollers equate to a shaw guard, I think is difficult. They would keep the stock down on the table to stop it becoming a projectile but they do not cover the blade as a shaw guard would. However if you are also fitting a SUVA guard which is not reliant upon the riving knife for a fixing then the SUVA guard would provide the cover to the blade and the Jessem rollers hold the stock down.
There is a risk with the Jessem rollers and a short fence that as the rollers are designed to push the stock against the fence, if there is only a short fence the rollers may push the stock into the open space beyond the fence which would be dangerous. You need to make sure that the rollers do not continue to act beyond the short fence or pivot the stock.
Many thanks, that's very helpful indeed. If for real wood I am to use a "short fence", would it be a good idea to actually use quite a long "short fence" extending backwards onto the infeed table? and maybe I could use my two Jessem rollers both on this, which perhaps would reduce the risk of pivoting the stock.
It would give you something to experiment with and it is easy to reduce the length of a fence and not so easy to extend it!
If you fit a sliding sub fence on your rip fence, you can simply release and reposition it forward when ripping live timbers that are at risk of springing when cut.
Can the Jessem guides be hinged up out of the way ? That would disengage the rear one when needed.
Re the "floating" crown guard. Picture a kick back where the back edge of the blade picks up the stock and accelerates it upwards, or a worst case scenario where you lose balance and bring your hand or even face down on the blade, do you really want to trust something flimsy ? The more robust the better appeals to me.
Re the "floating" crown guard. Picture a kick back where the back edge of the blade picks up the stock and accelerates it upwards, or a worst case scenario where you lose balance and bring your hand or even face down on the blade, do you really want to trust something flimsy ?
There's a post somewhere on the site from Steve Maskery describing an accident he had with his floating crown guard. I can't remember the detail but I remember that he redesigned it as a result, would be worth doing a search of his posts, it's after the workshop build.

Edit Here's the thread
Last edited:
Thanks, very helpful, plenty of food for thought.

Adding a sliding fence is easy enough; as long as it is no higher than the Incra fence then the Jessem rollers can work over the top of it (easily folded away when not in use); or the sliding fence can stay in place all the time and I could move the Jessems over on to it.

But either way I will no longer have a Jessem roller beyond the blade when ripping real wood; at the moment with full length fence that front roller gives me such a lot of confidence and security, making it very hard for the front rising part of the blade to start to lift the leading end of the workpiece and cause kickback.

The Steve Maskery episode was very informative. MY first thought on a floating guard was to make it dynamic, ie, it is always being pushed down, with a spring or gas strut; the stock pushes it up, and then it comes down again as the stock exits; but I felt there would be a lot of fettling required to make that work cleanly. So given that I had the Jessem rollers doing the pushing down job I decided that a static floating guard would be better in my situation. Steve Maskery's experience to my mind supports this: harder to spot spot a situation where a dynamic crown guard will move to a position during the cut where the blade is fouled than to set up a static guard so that it is clear of the blade.

But I'll chew this over for a day or two; one advantage of the dynamic guard is that it presses down at both ends of the blade during the cut.

Not really related to the question I've been asking here is whether I am paying enough attention to the offcut; it's all very well clearing the workpiece away from the blade asap; but presumably the offcut also has the capacity to become a lethal projectile ....
The value of the JessEm guides is that they are BOTH protective (prevent the workpiece moving backward, which is part of the kickback process), and they are also corrective (keep the workpiece flat against the fence, which reduces wandering and rough cutting).

This is mine ...


By the way, you will note that this is quite a wide panel. It is not a narrow board. The rollers are more likely to get in the way with a narrow board, so using the JessEm with narrow board is going to be more difficult. I get around this by have a long dowel as a push stick.

The improved cut from the JessEm comes primarily from its reducing sideways movement. This is important since the reason for a short fence is to permit movement, allowing workpiece to exit the blade and avoid the back of the blade. It is contact with the back of the blade which causes kickback. A riving knife also acts to prevent kickback by protecting the rear of the blade. Now, if the workpiece is able to continue along a fence and never leave it and move towards the back of the blade, kickback cannot happen. A long fence with the JessEm will do just this. No short fence is needed.

An even better saw cut comes from a slider with a parallel fence. This is when one rips on the wagon of a slider. My tablesaw is a Hammer K3. This has both the slider and a ripfence. Here is the set up for the parallel fence ...



The reason that this is relevant in this discussion is that is is yet another way to reduce movement in the workpiece in a saw cut. Reduced movement can produce a glue-ready finish ....

Here is the board ...


Close up ...


But ...



Regards from Perth

Thanks Derek, very nice saw and very nice results.

You certainly have it right that I'd like that standard of glue-ready cut, and I'd like to achieve it consistently. Before the Jessem rollers arrived I was nowhere near that standard, I had to cut oversize and finish by handplane.

I don't have the slider option. My Wadkins AGS10 came with what I felt was a rather hopeless aluminium outrigger with minimal range. I've replaced it with a static left-hand extension which I use in concert with a crosscut sled---works ok, but obviously no help ripping.

Sounds like you disagree with the uk locals who posted above about using a short fence for real wood even with the Jessems. Now I'm new to woodwork machinery but not so new that I'm oblivious to the sometimes acrimonious debates about safety regimes in different parts of the world, and I really don't want my thread to go in that direction ...

But just so I get the full picture, do you know what the official safety line is down under? specifically is a short fence suppposed to be used? and is there an exemption if you use stock rollers?
I've nothing specific to add to the debate, except to say that you should be able to read all British Standards documents online via your local library (mine allows remote online access during covid). I'm extremely impressed at the level of thought/care you're putting into this safety question, best of luck.
JoshD, when you cut real wood the process of cutting releases tension in wood fibres and sometimes the piece of wood finishes up looking like a banana. I have cut pieces that bow and twist a long way 1 inch 25mm of bow and twist or more. This usually happens in large stock 50 to 100mm thick and can occur even if the width you are cutting is wide. When this happens the wood against the fence will go either away from the fence or towards the fence. With some woods, no amount of Jessem guide will stop this . With a long fence this is a problem as the wood will jam between the fence and riving knife if you are lucky and the fence and the back of the saw blade if you are not (the guide may well stop the wood from becoming airborne but it will not always stop the jam) With the short fence this is not as much of a problem as the fence does not extend to the back of the blade and so if the wood bends away from the blade it is no problem. If the wood bows towards the blade then it is a problem no matter which fence. Even if the wood will not become airborne it could still damage your saw blade.
All that said, I have a Felder so it comes with a long fence and I use the long fence 90% of the time for cutting real wood. However when cutting some timber I know it is prone to bowing and twisting so swap to a short fence.
They are not binary questions. Long or short fence? Guide or no guide? It is about having the appropriate set up for the situation.
Thanks PAC1. I cut some green sweet chestnut sleepers over the summer that banana'ed like that (to make 5 bar gate). Of course you're then no longer cutting to final dimension, you have to cut oversize, then re-establish face and edge, then cut again or plane.... Can be an iterative process. If it will play ball and not bind up I try and use the bandsaw for that (charnwood b250, not the beefiest of machines).

So what I am talking about here on the table saw is making accurate cuts in wood stable enough to be able to make an accurate cut in it, if you see what I mean .....


I've nothing specific to add to the debate, except to say that you should be able to read all British Standards documents online via your local library (mine allows remote online access during covid). I'm extremely impressed at the level of thought/care you're putting into this safety question, best of luck.
Thanks for kind words. I bought my machinery last year, and I came to the conclusion a few months ago that my TS set up was neither accurate nor safe. Clearly TS is the most dangerous machine in the shop (responsible for 2/3 of accidents), and I decided there was no sense in using it at all unless I could make it accurate, and that even if accurate I needed to understand and control the hazards.
Btw I emailed Jessem themselves before Xmas to get their thoughts on my original question. Rather disappointed they never replied.
Yes you do but it takes no time at all.I think they're pretty good, applying pressure both vertically downwards and horizontally towards the fence, but really work best with a full-length fence.
I have Jessem stock rollers on my router table and think they are great. They hold down, prevent kick-back and pull the work against the fence.