Tool Post Grinder. Dovetail Version.

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

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


Active Member
Aug 29, 2009
Reaction score
Well, after doing some research and viewing others handiwork here, I decided to build my own version of a tool post grinder.


Design parameters are everything IMO. Everyone has different skill levels, different machinery, different budgets, and different purposes to suit. With that, nobody's design is perfect.

1. I have a small lathe. A Grizzly 9X19 to be exact.

2. I hate anything that is a hassle.

3. I don't think I will be needing a tool post grinder every day. Rarely at best. But, I do need one occasionally, and therefore it should be simple and quick to use.

4. Factoring in #'s 1, 2, and 3, I don't want to spend a lot of dough on a TP grinder.

So here is what I came up with.

The basic idea was to use the existing dovetail tool holder, to mount and use the TP grinder. The thought of having to take the entire tool post off of the machine just to set up a grinder doesn't do it for me.

On that note, I see A LOT of garbage tool posts installed on these lathes here. :-X FWIW, I will post a "how to" if needed on adapting the 100 series dovetail tool posts onto this small lathe. Yes, they fit perfectly, and it is very simple to install, and it is like night and day with the resulting usefulness of this machine. So technically, this is step 2 in the upgrade to the tool post. It's an expansion pack for the original modification. The 100 series tool posts are a measly $100 on sale at Enco, and they come with 5 tool holders, and their net worth is a million bucks in usefulness. Add a few more holders at $23 dollars each, and it's drop and go. Just as I want my TP grinder to be.

So, the basic design revolved around the dovetail tool post. Specifically, the 5/8 round hole boring bar holder is all but useless. I have 1/4 inch, 1/2 inch, and 3/4 inch, boring bars, and... it's useless as... (insert colorful metaphor here)

Searching for parts was easy in some ways, and tough in others. In fact, I think I am looking for some upgrades already.

Here are the parts I settled on for the initial design.

The motor is the trickiest part. After some errant purchases at the local "flea market store", I found myself staring at an old starter motor for model airplanes, which I have stored, and moved, and tried to give away to other R/C fliers for years. So, it has been an unneeded "Spare" lying about for way too long. The revelation was in my tunnel vision of A/C motors. Why not 12 Volt D/C? I put some tape stripes on the pulley, and put a tachometer on it, and it was running at 5,600 RPM's with no load. Hmmm. Game on.

So, I started with a blue print. This is my blue print, I drew it, I dimensioned it, I own it. If I patent it, everyone who wanted one would just build their own anyway, copying it. Nobody will buy it, because everyone wants to make their own anyway. Therefore, if you want a DXF, I'll send you one. Hack away builders!

(I will post it here, having trouble converting it to postable image)

Then I started turning the shaft and things.

For die cutting the threads, this is what I use. Simple, and parts on hand. Put the die in a socket. As you thread it on, you can lock the tail stock, and the die will slide up the inside of the socket. You can just turn out some more lead on the chuck if needed. You can watch it, and you already have the sockets on hand for any sized die.

I bored the dovetail holder.

I found it broke through some holes, sized for using the 6202 bearings I got at the Ax Man (for free with purchase of my new mill chair), lol. But, no big deal, if it bothers you, use smaller bearings. The trade off will be a smaller shaft size with less rigidity though. The metal in these things is kind of cheesy and cast like, but not really here nor there. :big:

I used the 4 jaw for the first time, and I was less than confident that the bearings would be bored true. I did face the first end, so that hopefully the other end would square up against the chuck better. These holders are ripped out en-mass at the T Aiwan machining company, and are not square by any means. Just indicating in a hole with a corn cob finish is a thrill, lol. But, it worked great.

Pulley design was a coin toss. I had some tooth pulleys, and a belt that was longer than I wanted, but the friction would be low, and the pulley adaptation would be tricky. O-rings work, but milling and lathing at the same time for round grooves is not an option here yet. And the starter motor already had a v-grove in it, but I knew loads would go up turning a v-belt. ....

The v-belt won out, but it can also run an O-ring too. More on that later.

The cutoff shown here is an insert design, it fits in the tool post, and is reversible from right to left. The trick is keeping blade pressure up, and it cuts like gangbusters with little chatter at all. Turn slow, push firmly. Reasonably priced too.

Anyway, here are the basic pieces ready for assembly.

Some things to note here. I turned a shoulder on the shaft for a press fit to the bearing, and turned the rest of the shaft about .002" under sized. Then I knurled the bearing area to provide a press fit for the second bearing. This way, the bearings slide easily up the shaft to the point where they need to be, without fighting you the whole way. Two shoulders would work too. But .... A. I screwed up. and B. I wanted to try my knurling tool anyway. :p

Here is what it looks like assembled, lying on the bench.

And the drive end.

So, now all I have to do, is remove the tool from the tool post, and drop this assembly in place, lock the dovetail, and plug it in.... well, actually it's more like .....go get the battery from the boat that is in storage, and the battery charger, and.... lol. But, I have a plan for that! :big:

Business end. I can see what I am doing, and nothing gets in the way of anything else.

Except for. 1. the cord is precariously incomplete and strays a bit close to the chuck for my liking. 2. The wheel attach bolt (arbor/adapter) needs to be a button head cap screw, so it doesn't kill me when it hits the chuck, like it would now. and 3. It just ain't done quite yet. ;)

The drive end. You will note that I have left some tail stock on the shaft for plans B, C, and D, if needed. lol. Gear ratio pulleys, etc...

And a front view, for size reference, et al....

Now, as for some pending verdicts.

The v-belt really soaks up some power and RPM's. I think it will only turn at about 3,500 RPM's with this heavy of a belt. It also draws a whopping 10 amps with it installed.

Not to fret, using a large O-ring in the same v-groove, the RPM's raise to about 4,500, and it only draws about 7.3 amps.

However. It is apparent that this is an intermittent duty motor. How do I know this? Well, it is designed for starting model airplane engines, then you go fly for 15 minutes. Also, you'll note a distinct lack of cooling holes in the motor, and the absence of a cooling fan. Hmmm.

The case halves are riveted together, so, although designing and making a centrifugal fan on the pulley side would be "easy peasy lemon squeezey", the disassembly of this motor to cut holes is another story.

On the other hand, this is a DC motor, thereby, I can purchase a 15 or 20 amp Variac (variable DC transformer), and run this thing at voltages of up to (probably) 24 volts and have speed control to boot. However, increased volts means increased amps, and .... increased heat. My "Good" starter is rated at 12-24 volts.

One could argue, that this motor was free, and had one foot in the garbage can anyway, so why not test the limits. And in fact, I intend to do so. However, breaking it, by disassembling it to add cooling holes, before I even get to use it might prove disappointing at best....

The other issues are learning which wheels I want, etc... The 2 freebies I have are coarse, and somewhat less coarse IMO. I am not getting a finish anywhere near the factory finish on a shoulder bolt, for example.

But... I am grinding. and it takes 2 seconds to change from a cutting tool, to a tool grinder. And that's saying something.

Any, and all, ideas and comments are welcome. If you hate it, you can even bring that too... lol


Good job T

When I built my Tool Post Grinder I was surprised how often I started to you use it, when you've got it you start to find work for it.

I found with mine that the power of the motor is important, started off with a sewing machine motor rated at 180 watts at 3000 rpm, at a 1-1 pulley ratio it was fine, but when I tried lowering the ratio to increase the RPM it just didn't have the power to get going, ended up using a motor from a woodworking router.


Thanks Stew. Your insights have helped a lot here!

I did look at my router with dismantling eyes, and then promptly gave it a reprieve of sentence, when I saw the 30,000 RPM's listed. I also looked at my Roto-Zip, and then offered it the same clemency, for the same reason.

My son produced a dishwasher motor from under the crawl space. :big: Which he apparently dissected his mother's dishwasher as it hit the scrap heap, and is hoarding this kind of junk at my house, lol. But that was a scary motor. Open holes, no safeties, exposed fan, and the shaft had at least 1/4 inch of end play, and plastic ring mounts. It was a 3750RPM water pump motor, and was quiet, but clearly designed to fit into a caged housing. It had issues.

But I am still open to motor suggestions. This baby has some good torque (It is designed to start up to 1.8 CI engines after all.) But being the cheaper model, it says nothing about 24 volts, only 12, lol. Heat is the real question here. Speed increase by raising voltage, to the point of blowing wheels up, is of little doubt. But she's going to get hot... But for $79 dollars I could buy a brand new super high torque model like my good one. And when you think about it, that ain't bad, for a brand new 12-24V DC motor that meets the needs. ....

Maybe I'll try some heat removing mods on this one. If they work, pick up a used ultra high torque version on eBay, along with the variac, and mod that one the same way. ??

Hi Ted

This is the router I eventually used:-


I was given it by a friend as the guide fence had broken, but the motor was still good its rated as

850 watts variable speed between 11400 and 28000 rpm

I made the grinder on the large side as I'm planning to use it as a cutter grinder some time in the future, but thats another project.

You can follow my build her



T, this is a quick & dirty version that I built so I could grind the hardened valves that I am using in a small engine I am building.

The motor is a 110V. I found it at Axman on university ave. I don't know what the rpm is but it seems right for the small grinding wheel I used. When I spotted the motor, I checked for end play. There was none. They tested the motor at the check out desk and it had a lot of torque and seemed plenty fast for what I wanted. Cost -- $4.00.

I mounted the motor in my AXA QCTH by making an aluminum cradle that fit the AXA and held in in place with hose clamps.

I ground the seats on 4 valves in just a few minutes. Worked great.

This might be a source for the motor you are looking for.

Actually, the 2 motors I bought, were from Axman. I've been there twice recently. But the two I bought were quiet, and smooth, but I grabbed the bare spindles and pinched them between my fingers, and instead of burning them, I was able to stop the shafts easily.

My die grinder at 11,000 rpm was promising, until I remembered that it sounds like the circular saw from Hell... :big:

So, what are you guys using for dressing your wheels? I've seen 1/4 Carat replacements for as low as $23, but, I am not sure how I would hold it. A mag base? A holder for the chuck?

Do you dress a Borizon wheel? I forget. As little as necessary, I'm sure of that. It's been 20 years since I've done any grinding on a monoset, or the like, what wheels would you recommend? I have a pink and an orange Norton, both on thier last legs.

My primary project needs are to turn down the OD of my turbine wheel blades, (after I bore it, which I may break down and call in a favor, and have it wire cut, as it's hard to clamp with no hole) made of inconell. And, my grade 8.8 turbine shaft bearing shoulders. Suggestions, I hate to waste money on stuff that wasn't what I thought it would be.

Super post and pictures - it has inspired me to do something similar (when time permits).
Just one point - a Variac works on AC, there is a clue in the trade name Vari(able) ac (alternating current). With the voltages you are using on the motor some form of electronic control would be more suitable perhaps?
Hope that helps

Nice post!

These tool grinders do like to throw grit onto the ways as people always bring up.

I'm making a new set of way covers for my mill that it occurs to me would be convenient for covering lathe ways too. Sorry, no pics yet, but I can describe.

I'm using some of those really powerful neodymium magnets you see everywhere. I basically cut 2 aluminum strips and drilled them for a zig zig of those little magnets, about 1/2" between each magnet. Now you can take 2 of the strips and sandwich them together to hold your way cover material. The strips will stick to each other to clamp the material, and they also stick to your machine. Real quick and easy to put on or take off.

The way cover material I'm using is a shower pan liner available from Home Depot. It's a very supple rubber of just the right thickness.

I think this sort of thing would work well for TP Grinder work too!



PS Bunch of TP Grinder ideas on my page here:

Latest posts