When locking the head after tramming puts the mill seriously out of tram

[Bill PhinnParticipant @billphinn90025]

I’ve just had the displeasure of spending nearly two hours tramming my Warco WM18 (it’s now within 0.01mm over a radius of about 7 inches, which will do).

Nine tenths of the displeasure was caused by the fact that tightening down the two locking bolts after tramming always put the perfect tram I’d just carefully dialled in out of whack by anything between 0.5 and 1mm – the amount things became out of whack being proportionate to how tightly each locking bolt was finally tightened down – something which, as you can imagine, is difficult to gauge without a torque wrench, and there’s no way you can get any normally-proportioned torque wrench up inside the head of a Warco WM18 and on to the upper of the two locking bolts.

Having to counteract this tram-altering effect at the final hurdle and hit the desired accuracy when everything was tightened down seemed practically impossible initially, and it was only by many tedious lockings and unlockings and tappings of the head that I finally hit the numbers I was looking for with everything locked down.

If anyone can suggest a tramming technique that would involve less trial and error in the present circumstances, I would be delighted to hear from them.

[DC31kParticipant @dc31k]

From your description, this appears to be an engineering problem not a technique problem.

Human beings are smart enough to overcome deficiencies in most things, as you have proved by the hours of your life you have just wasted.

Spend one more hour applying that same smartness and assess WHY the current setup does not work as it should. Think about how it could be re-engineered to function better. Heal the injury, not just go to the forum shop for a better plaster.

It is hard to recommend anything at all from the sparse description you have given. Unless a person had the exact same machine, we do not know the details of what you are doing nor how the mechanical system works. All you have told us is that you have to tighten two bolts that are somewhat inaccessible and that you believe the torque to which those bolts is subject has a high correlation with success of the overall process.

[JasonBModerator @jasonb]

Yes hard to say without even the basics such as is it moving sideways when tightened which would be due to head rotating or is it front/to back nod which could be down to the mating surfaces?

A good reason not to tilt your head unless you really need to.

[Brian WoodParticipant @brianwood45127]

I don’t know the machine at all but might I suggest fitting a lathe faceplate to spindle fitting hardware; pulling the combination hard down onto the table, with the tramming bolts loose, and clamp the faceplate down to the table while you bolt things up again.

Regards  Brian

[John MCParticipant @johnmc39344]

I had similar problems with my mill, a Cormak ZX7045, similar to other suppliers offerings, close to Warco’s Super Major and similar to the OP’s WM18.

First thing I did was remove the paint and filler from under where the three head clamping bolts sit.  Also made some much more substantial washers.  This made the clamping far more positive.  Doing this helped somewhat but was not the answer.

The photo shows how I finally achieved accurate tramming, self explanatory I hope.   The device works for vertical and a few degrees either side.  The round stop bar can be unclamped and moved out of the way for bigger angles.

I think the problem is caused by a couple of things.  Firstly, the head is poorly located on a very short spigot so tends to lean forward when the clamping bolts are released.  Therefore one is immediately fighting the alignment in two directions rather than one.  When tramming the head I leave the three clamp bolts just nipped to stop the head leaning forward.  The two set screws are 5/16″ UNF, quite capable of moving the head even when the clamping bolts are still holding the head against it’s mounting face.

Secondly, the excessive mass of the head doesn’t help with the minute amounts of movement to achieve the accuracy of tramming wanted.

 

[SillyOldDufferModerator @sillyoldduffer]

I have a WM18 which, like Bill’s is hard to tram.   Though mine doesn’t move out as far as his does –  0.5 and 1mm is a lot.

I put it down mainly to the extreme simplicity of the tilt arrangement.   To tilt the head, two bolts are slackened.  One serves as the axle, allowing right left movement.   The other is a clamp, causing nod as it is tightened or loosened.   Problem is, the two bolts interact as they are adjusted, and tiny movements translate into noticeable tram errors.

What works best for me, with no science and not much experience behind it, is to slacken both nuts by the absolute minimum needed to allow head movement   This, I guess, prevents most of the nodding error, front/back.  Then re-tighten each alternately whilst checking no movement has occurred.   The method is not sure-fire:  on one occasion worked first time, others took several frustrating attempts.   Now the tram is correct, I avoid messing with it.   The need to tilt the head can often be avoided by tilting the work instead.   Being able to tilt the head of WM18 is of dubious value because the set/reset time is so high.

Last time the subject came up on the forum, someone (sorry I forget who), suggested a different method.   Basically, it was to lower the head flat against the table (or precision blocks), so that the tram is physically enforced.  Then tighten the nuts.   Makes sense in that head has less opportunity to move.   Haven’t had occasion to try it yet, because I avoid moving the head unless I really, really must.

Bill’s adjustment finishing up to 1mm out seems extreme.  A factory finish problem maybe?   Check the bolts, nuts, holes and surface for burrs or dirt.   But I suspect the real cause is that the tilt feature is cheap to provide and works ‘well-enough’, except it’s a pig to adjust.

Dave

[Clive FosterParticipant @clivefoster55965]

Besides the inherent engineering issues of a heavy head with all its weight overhung on a short spigot it sounds as if something is either not quite flat enough or the mating surfaces aren’t sufficiently clean. A bit of crap rolling around between two milled, rather than scraped, surfaces can create surprising amounts of error.

Another gotcha is if the ends of the drilled bolt holes haven’t been chamfered to ensure nothing round the edge has risen when the drill starts or breaks through. More usually associated with billet materials rather than cast iron but it can happen.

Time to pull the head and give it a very careful going over. Make yourself a stout support post that can be fixed to the table and grabbed in a collet to hold the head as you remove it using the table feed to pull it off the spigot. Use it to hold the head whilst you are working on it too rather than chasing around the bench. Probably need an extra support at the spigot end too.

Ultimately all the affordable machines tend to suffer from this sort of problem. There is limit to how much effort that can be put into a machine without taking it well beyond the price you can afford. It’s a great pity that the hobby machine tool market is way too small to support Tier 1 Automotive Supplier style factories with their ability to produce very accurate components at silly low prices. Even with professional market machines, like my Bridgeport, you still have to be careful with the right technique to quickly get things just so.

Tramming accurately is never a trivial job. Need brain in gear and temper in cool, calm and collected mode. I’ve been known to walk away for a time out before the red mist descends and I break something.

Concerning the torque wrench issue it’s relatively simple to make a bending bar type by welding the end of a spanner onto a suitably slim bar with a pointer above, or along side it. As it’s a one value thing calibration can be very ad hoc. I’ve used the ending bar principle but in a rather different area.

Allegedly you can hook a spring balance or luggage weigher onto a spanner and work out the pull need for given torque. The only guy I ever met who had tried that was quite eloquent in explaining its impracticality.

Clive

[SillyOldDuffer]

On SillyOldDuffer Said:

[…] I suspect the real cause is that the tilt feature is cheap to provide and works ‘well-enough’, except it’s a pig to adjust.

Me too 

This sounds like a perfect example of the trade-off required to serve the hobby market.

I think it was John Stevenson who coined the phrase “fit for purse” to balance the ideal of “fit for purpose”

The likes of Aciera do these things so much better .. at a price.

MichaelG.

[Brian WoodParticipant @brianwood45127]

Dave,

The method of lowering the head to the table to hold the tram until bolts are tight was used frequently by one of the professionals on a Bridgeport mill. He often had to work in complex angles and this was his quick and successful method.

My faceplate approach [above] is a variation where quill reach might be limited and it has worked well for me on my Dare Westbury mill, another round column machine.

But JohnM [also above]  exploits the design very neatly in my view and his method gives micrometer control to correcting nod

Regards   Brian

[JasonBModerator @jasonb]

I thought John Mc’s screws were to adjust rotation which would be sideways tram not front to back nod?

[noel shelleyParticipant @noelshelley55608]

The spring balance torque wrench idea that Clive mentions does work – eg required torque 500ftlb = 4ft bar and 125lb load. when doing up wheel nuts on BIG vehicles my weight is the load on a bar of calculated  length, using a degree of experience and common sense.

As to the problems of a hobby mill tramming is one of price versus good engineering/ usability. For the amount of spring it would seem that there is a lack of flat surface to pull down on, there by the torque applied to the bolts varies the setting. I think one has to decide whether to try and work round it or engage in a re engineering project ? Good Luck. Noel.

[Brian WoodParticipant @brianwood45127]

Jason,

He said the method works for vertical tram, ie nod, to which I concur, and a few degrees either side for sideways tilt. I would agree with you that is rather limited but it is the easier direction to correct in any case.

I still prefer my faceplate method that can set tram in both directions in the one operation.

Regards   Brian

[Tony Pratt 1Participant @tonypratt1]

All the mills I have ever worked on have this problem to some extent or another, just slacken enough to tram the head then start to tighten the nuts adjusting the tram as you tighten OR if you are smart enough adjust the tram with a bias such that when you tighten the nuts the head will be square.

Tony

[Dave WoottonParticipant @davewootton]

At one time I worked for a company that imported the VM-E type mill from Taiwan and branded them as their own or supplied them to other sellers branded with their own name, the inspection of these was very rigorous and it was found a large number of the batch would behave in the same manner as the OP’s machine when tramming. The milling head was fitted to a T shaped casting that fitted to the column, this looked to have been nicely machined and looked  fine, but when blued up showed some shringage had occurred after machining, enough to make tramming very hit or miss, the shringage was mostly in the area where a vertical core of the T section was, probably the result of not seasoning the casting properly before machining. Distorting the mating faces as the head bolts were finally tightened fully. The answer was much tedious scraping and checking, given to the youngest and newest fitter ( Me- it was a long time ago!). It may be a bit of a pain to do but if possible I would suggest as Clive says above to blue the mating faces to make sure they are flat and true and not distorting as things are tightened up. Good luck with it please let us know how you get on.

Dave

[larry phelan 1Participant @larryphelan1]

Delighted to hear that I am not the only one having problems like this, nice to know that I am not alone ! In my case, I tried putting shims in, taking shims out, putting them upside down and turning them all around. Made no difference that I could see, so I put it all back together again and brought the dog out for a walk. Temp by then was about 5deg short of exploding.

Seems like these things are simply made to a price and a standard, full stop.

[RobinParticipant @robin]

0.01mm -gasp- a dream come true 🙂

I cleaned out the assembly connecting the beam to the head and the nod tram went from +0.22mm to -0.14mm

Obviously I now have to give the assembly connecting the beam to the column the same treatment

This is going to be fun 😀

best

Robin

[Bill PhinnParticipant @billphinn90025]

Thanks a lot to everyone for your very helpful replies.

I should say at the outset that the re-tramming was unavoidable after I had tilted the head to gain access to the stripped high-low gear so that I could replace it. A separate thread discusses this. I intend to post an update there when I have time. Use of a tilting vice removes the need for me to tilt the head at other times, or at least has done so far.

It was the first time I’d had to re-tram the mill; from the factory it was trammed to within a thou over a sweep of 14 inches. I’ve improved on that a little now with 0.01mm over the same distance.

Several of you (Clive F., S.O.D., DC31k, John MC, Noel, Tony, Dave W.) have drawn attention to the design and/or finish of the machine being likely to be behind the present problem. Clive’s recommendation to pull the head off and give things a thorough looking over is something I definitely intend to do at some point, but since I’ve only just had the chance to get my mill working again after six months of it being out of action, I’m loath to prioritise the time for head removal and re-engineering just at the moment.

As Dave S.O.D. suggests, a downside to this type of mill is the head-locking mechanism it happens to have. Others (e.g. YouTuber Olly of Olly’s Workshop in his side-by-side comparison of a Sieg SX3 and an Amadeal mill much like mine) have pointed out that the two locking bolts on my type of mill (one vertically placed over the other and them being rather close together) probably does not make for as secure a head locking system as the one-bolt-either-side locking system found on, for example, Robin’s mill and John MC’s and the Sieg SX3. I’ll try to photograph mine up inside the head later so people unfamiliar with my typeof mill can see what we’re dealing with.

My main takeaway from this experience, besides the excessive labour involved in tramming the mill and the feeling that my mill is almost certainly “not quite right in the head”, is the concern that, if locking down the head causes this much movement so easily, in use the head may be inclined to move out of tram far more easily than is desirable. I’ll certainly be checking the tram again very soon to see whether things have moved. There was no indication it had moved away from the factory setting before I tilted the head, though I’d not checked the tram for about six months previously.

I’m aware there’s a lot more content in people’s replies to respond to than I’ve responded to in this post, for which I apologize, but I do aim to come back to the thread later when I have more time.

[Clive FosterParticipant @clivefoster55965]

Bill

And one more thing.

Verify that you have clearance at the sharp corner where the spigot and main casting meet.

Common practice is slight undercut at the spigot joint and a small radius at the end of the hole.

Only takes a tiny machining error for sharp corners on both to engage a tiny bit before the faces. So the bolts have to pull things down for the faces to meet as they should.

Clive

[Clive Foster]

On Clive Foster Said:

Bill

And one more thing.

Verify that you have clearance at the sharp corner where the spigot and main casting meet.

Common practice is slight undercut at the spigot joint and a small radius at the end of the hole.

Only takes a tiny machining error for sharp corners on both to engage a tiny bit before the faces. So the bolts have to pull things down for the faces to meet as they should.

Clive

Whilst you were typing that very comment, guess what I was doing. 🙂

Mill Tramming Tool (& Use on a Centec With Riser.)

Bill

[Bill PhinnParticipant @billphinn90025]

Thanks a lot, Clive and Bill. Some valuable information and advice there.

Being a glutton for punishment, I trammed the mill all over again today, this time using the full length of the table (about 26”) to read from. Yesterday’s 0.01mm over a 14” swing had transformed itself into 0.04mm over the longer distance, so more fiddling was embarked on to get it down to a truer 0.01mm.

This time I didn’t undo the two locking bolts as much as yesterday and I used a jack under one or other side of the head to tilt it what I’d roughly calculated would be the right amount each time. The max deflection on tightening the bolts was around 0.2mm, which was easier to compensate for than the bigger deflections of yesterday. I’d replaced the factory-fitted grade 4 hex head set screw in the upper position with an Unbrako cap screw so that I could leave the long ratchet in the hard-to-reach screw head between adjustments.

It remains to be seen how long the present tram holds good. I’m sure people can see for themselves how disturbingly close together the two locking bolts are as well as the lunar surface the washers have to bear on to. Far from ideal.

The last two days have revealed something else I’d long suspected was the case, which is that the quill lock on this machine skews the quill quite a lot – about 0.1mm when measured at each end of the table. This is one of the reasons I was already in the habit of using the head, column locks and Z axis DRO as much as possible rather than the quill lock and DRO.

 

[Charles LamontParticipant @charleslamont71117]

Put some spring or Belleville washers under the nuts. This will allow them to be slackened enough to allow rotation of the head while still maintaining the mating faces in contact.

Same trick is also a huge help with faceplate setups in the lathe.

[Author]

Posts