Thread: 12x36 Lathe compound handwheel markings are wrong

In comparison to my last post, this is something you SHOULD look at, if you actually use the compound slide to move exact amounts. I just plunge when threading (rather than the 30 degree method), and use it to cut tapers where the length is fairly irrelevant, so I've never realised this before. It's probably been discussed before, I guess.

I wanted to face 0.2mm off an existing part on the AL335 last week, so I thought I'd use the compound to make it easier to dial in the small amount, given the carriage handwheel is graduated in 0.5mm divisions, and I'm not sure that they're actually accurate anyway, as I've tried to dial in a specific amount off that handwheel before and got some odd result. It'd be nice to blame the single set of graduations that bend the laws of physics to be both 0.02" and 0.5mm at the same time, but from memory it was out more than that. I usually just use verniers to measure the part as I go anyway, as the length between shoulders usually doesn't matter for what I do, and I also have a mount to set up a dial indicator to measure carriage travel if I want.

In this case, I wasn't 100% sure on the exact amount I needed to take off anyway, given I was measuring it with feeler guages and thumbnails, so I figured I'd just use the finer resolution of the compound to be able to dial in really fine cuts and check the fit. Dialed in 0.2mm on the compound, and scratched my head. Just didn't look right. Fiddled around a bit, and decided, well, surely it must be right, must just be the nose radius on the insert playing tricks on my eye. Made the cut, and what do you know, took twice as much off as I wanted. Eh? Backed the carriage off, flushed up the front edge of the compound slide to the lower part, zeroed the dial, wound on 1.25mm, and measured it with a ruler. 2.54mm travel. Okay, I thought, it's just the old diameter/radius deal, dialed in 0.1mm on the compound, and faced the mating shoulder, job done. Didn't think anymore about it until I got out of the shower last night, and went hang on, WHAT THE F.....

Upon inspection today (Easy since the compound is still stripped on the bench), the screw for the compound is 10 tpi. Thus, every turn of the compound moves the slide 1/10 of an inch, 0.1 inches - which is exactly whats marked on the compound dial on the imperial side. So far, so good. The metric side, however, is marked as one full turn being 1.27mm - where it SHOULD be 2.54mm. Unfortunately on my machine, the ink has rubbed off the little plate that tells you how much each division is supposed to be (probably would have helped me realise earlier), but upon checking the photos on H&F website, each graduation is supposed to be 0.02mm. The way the graduations are on my handwheel, every second major line is marked as an increase of 10. It SHOULD be, either every major line marked as 10, or every second major line marked as 20.

Having looked at the photos on H&F websites, this applies to both the AL335 and AL336, but it wouldn't surprise me if this was a common thing on most variants of the CQ6230. I have a vague recollection a few people have had weird struggles with threading using the 30 degree methods in the past, and this might have something to do with it.

Attached a photo straight off H&F website to hopefully help visualise what I mean, if you look at the 70 thou mark on the imperial scale, you'll see that the metric scale shows about 0.9mm, which is only 35 thou. It SHOULD be 180. Now I'm considering turning the numbers off the metric side, and remarking them, cause I KNOW I'll forget that with the amount I use the compound...

*EDIT* Aaaand, I just realised the tailstock handwheel has the same issue, not that I've ever used it - but something to keep in mind.

Weird. I'll have to test the few H&F lathes I have access to, and compare with this anomaly

Try using a dial indicator on a mag base to measure what you are really winding on. Particularly with those dual scale dials, I'm never really confident on how much things are moved. Add some lead screw wear...

Michael

Déjà vu , I posted a thread on my L600 mis-matched dial. Hopefully next week my replacement will be completed. Run it past H&F and see what they have to say.
Regards,
Tom

Originally Posted by topari

Déjà vu , I posted a thread on my L600 mis-matched dial. Hopefully next week may replacement will be completed. Run it past H&F and see what they have to say.
Regards,
Tom

Given I'm the second owner, the machine is ten years old, and the close up pictures on their website appear show the same issue on both the AL335 and AL336 (couple of other machines I checked seem ok), I doubt I'd get anywhere. They're not going to want to ship out a few thousand compound and tailstock handwheel collars to various owners of these machines over time, even if they had some that were marked correctly...

Just wanted to make people aware that this issue exists, if it's in the back of peoples minds someone might be able to help a new poster someday who gets caught by this one. As Michael G mentions, there are ways around it, and they should be used if you're really going for super accurate (although I'm not sure how I'd go setting up a dial indicator on my compound, not much real estate there). DRO would work nicely though.

I discovered a similar discrepancy on my 2006 metric lead screw AL340D 4 or 5 years ago when the y axis on the DRO pooed itself, up until then I had taken no interest in the hand wheel markings.
It turned out when I had to use them and rely on them (W T F is going on with these lengths?) that they were telling fibs.
The hand wheel tells me it is moving the carriage 14.5mm per rotation, investigation and experiment shows the movement is actually 9/16" or 14.29mm.


Footnote. I have spaces between the W and the T and the F because if you don't that 3 letter acronym doesn't appear. F F S!

Given there seems to be more interest in the carriage handwheel than the blatantly wrongly marked compound, I went out and stuck my long travel dial indicator on the carriage, zeroed it with the handwheel at zero, and investigated whether it was metric or imperial. The answer? Neither, really

It's closer to being imperial, as 16.5mm doesn't get quite get back to zero on the handwheel, but it's a little shy of being correct of the imperial measurement either at zero (hits about 16.65ish). But it would make sense that it's nominally imperial, given all the screws are imperial, and the numbers marked on the handwheel are sort of* marked in thou.

* By sort of, the numbers are marked every ten graduations, 10, 20, 30. Zero would be 33. But the little plate that you line up to says every increment is 0.02". Thus, 33 * 0.02 is actually 0.66 inches, or 16.764mm.

What on earth were they thinking? I can't help but laugh. I don't really care much about the carriage handwheel markings, they're too bloody hard to keep track of anyway, and at 16.6 something mm a turn with the fairly ordinary rack drive arrangement (as opposed to the screws on the cross and compound), not exactly suited to precise adjustments. But it's no wonder my approach to machining things on it has evolved as a process of making texta marks on the workpiece as a guide, cutting, and then using verynears to get the actual length I want. At least the cross slide is actually marked correctly (1 out of 4 ain't bad, really!), and takes off pretty much what it says it does....

Well, until you've just been using the mill knee, and forget you're back on the lathe and forget you have to halve the difference between your current and target diameter, cause it measure direct movement, not reduction in radius. Operator is deep in error. But that happens I believe on any lathe, some are marked as to what they take off the diameter, some as to what they take off the radius.

Well ya made me go have a look at the H&F site.
It looks like you have to pay 10 grand to get correct dials.

It's just the same issues on my AL-300 lathe, owned for 20 years & used for hobby work, bike restoration etc. The compound & cross slides have the same confusing 12.7mm engraving for one revolution of the dual-marked scales that accurately moves the slide by 0.10" when checked with a DTI. Same on the tailstock.
The carriage is only marked in imperial; one rotation ccw i.e. towards the headstock moves it along by 0.575".
I find it OK to work in small metric dimensions as long as it's less than one rotation of the top hand-wheels. The mental exercise converting imperial to metric or vice-versa keeps you on your toes. One learns to live with the problem I guess. Measure & cut, measure & cut & repeat ad nauseam....
I only recently discovered that the marked angle setting for the compound was about 1 degree out. Or rather the zero mark/lubber line was inaccurately placed. But then I'd never tried to machine matching morse tapers before, not being a professional machinist. Once I set up an existing MT3 blank & accurately matched the compound angle using a dti, it all became clear. I later established just where the zero line should be, removed the compound, filed off the old line & re-marked it. Trust nothing is the rule I suppose.

I've an AL 336 and same problem. Trying to take an exact depth cut has a bit of a guess factor in it, needless to say an accurate result isn't guaranteed. Was looking at setting up an 50mm dial indicators for both axis but then saw a linear scale with digital readout for about half the price, about $40 for 200mm and $65 1000 mm. So for a bit over $100 digital scales on both axis. Not a full blown DRO but would do most of what I want. DRO's are 3 to 4 times that. These linear scales are made by Shahe and for sale on aliexpress.

Originally Posted by Halifax614

The compound & cross slides have the same confusing 12.7mm engraving for one revolution of the dual-marked scales that accurately moves the slide by 0.10" when checked with a DTI.
.........
I only recently discovered that the marked angle setting for the compound was about 1 degree out. Or rather the zero mark/lubber line was inaccurately placed.

Jesus, you copped the cross slide as well? That's the only one that's correct on mine, I'd have burned the bloody thing if that one was out. It's probably not 100% accurate, but it's close enough that any variance is as likely to come from deflection of the workpiece as the slide, verynears or micrometers have always been my go to if the exact dimension matters.

And you actually got a marking for the compound angle? Mine only ever had the faintest trace of one there... I lightly scribed an "about there" line on it for rough work, the scale is too coarse anyway. Indicating in as you did is a far better way if the angle is critical.