A spindle walks into a test bar... WHAT?!

[@steamfan7147]

If your compound slide has enough travel, you could re-grind the spindle taper in situ.

[@bobuk5722]

Hi, you've got me realising that I need to give my version of this lathe a very good going over. Well done by the way for finding the challenge, I, and a few others I expect, now know how to test for and also to overcome it. I'm still working on my mill, one shot oiler, DRO etc and now I have to check its spindle! There's a lesson here for me I guess, as I've got the measuring kit I should use it ...... At least got one thing right, I prioritised being able to make accurate measurements early on in my purchase history. Now I find I need to use them!

[@terrymoorecnc2500]

This is why in the custom metalworking business experienced machinists never trust the taper in the spindle even on really good machine tools. I've been doing this since 74 and I won't trust the spindle taper. If you want to turn parts between centers, turn the center in the chuck and you will be able to count on it. I've even seen tailstock tapers that were out on CNC Turning Centers. I'm not going to tell you how to work around that one.

[@CRILIKk]

I did a similar exercise on my small lathe , I found that measuring run out wit test bar first was key & at 3 points & from there I used a Morse taper reamer to fix any issues & got it close enough , once that was sorted I found the spindle housing needed refinement on how it sits on the bed to fix the angles it was incorrectly sitting at a lot of work & head scratching & I got it really close & a lot better

[@daniel635biturbo]

Nice little Lathe video series, I'm also happy that I bought a 1942 Southbend clone Lathe, as it seems to be almost the same amount of work to get them in good fit and finish. Only drawback is that I miss having roller bearings, as I'm stuck with Bronze bearings in mine. (And the inherit difficulty of having no or small spindle play) I also did a small video series but it's in Swedish actually a bit similar to yours. I'm calling my machine Frankensvarven = Franken lathe as it now consists of parts from several old donor machines, and some new Chinese parts 😅

[@Ivin8091]

The bearing run out is not repeatable. Some times error can add up to each other, some times the error can conter out each.

[@adhawk5632]

It was all answered in the last test, dead center test in the spindle, runnout is bad, mark the dead center and recut the point👍👌🇦🇺

[@bobweiram6321]

It's possible to have a perfectly tapered bore and still have runout. While the taper bore can be perfectly tapered and centered, but it may have been bored at a slight angle, which only shows up as you move further away. It's why headstocks are usually line bored and not drilled to ensure absolute parallelism with the bed.

[@nbrworks]

Hi Bob, thanks for your comment. I also thought of that, but if that was the "ONLY" problem, I think we would have repeatable results...? I confess that what puzzled me the most was the randomness in what the indicator was showing (I tested several indicators, by the way).

[@phillcameron8090]

I have the same lathe and issue. Solved by using a Morse taper reamer in the spindle. You shouldn't be able to remove the test bar by hand. If it's making full contact you will need to knock it out from the through hole.

[@phillcameron8090]

I also reamed the tail stock, 100% required unfortunately if your chasing precision

[@neverwipe]

In your first test the cylinder central axis is intersecting to the spindle axis of rotation at different points. When it intersects at the point you're taking the measurement, it shows concentricity, but the 2 axis aren't aligned. The other tests later in the vid are harder to understand. I think what's going on here that explains the consistent indicator showing a drop as to sweep to the tip of the cylinder, and the "non parallel" reading you got when you rotated the cylinder 180 degrees is the following: Given that your dial indicator is vertical, when you're measuring a point on a circumference, the measurement drops when the point you're measuring is misaligned vertically from the central cylinder axis. You're not just measuring the vertical error when you sweep the bar like this, since if there is a horizontal error the measurement will also drop. The fact that the error seemed to consistently follow the bar and not the spindle is very confusing to me. You showed that the test bar was good when you measured the taper to be concentric to the cylinder. Maybe since the tapers where not mating fully, you're seating the bar in the same way, causing the same side of the bar to be aligned with the carriage travel? I'm interested to hear updates on this. Cheers!

[@nbrworks]

Thanks. I agree with you and I think the following of the bar might have been a coincidence. I really think the spindle taper is not good and one day I might try to regrind it. But that will not be right now as I have other projects ;)

[@willrobertson7778]

Cutting a taper precisely on a lathe is difficult and a tiny error in the taper angle can cause big problems - that got me thinking about how to resolve the situation. For very precise parallel holes we use a drill a little smaller than the finished hole then a reamer to shave off the small excess of metal and finally cut the hole to the exact dimensions needed. I got to thinking "Would that work for tapers?" and it turns out that it is possible to buy "Morse Taper Reamers" (or "Morsekegel Reibahle" in German) to finish the taper to precisely the angle required - they're fairly expensive but I think one of them would solve the problem and might come in useful in future. (I'm not old enough to know much about this but I think - from speaking to much older folk - that lathe headstock tapers were one of the last things to be standardised - with some European and US lathe manufacturers using in-house tapers until relatively recently 🙈) The problems that can be caused by tiny deviations in the taper are so bad that there are special plastic tools with flet pads to clean out tapers and make sure that no tiny particles of dust or swarf are there that could mess up the functioning of the taper.

[@nbrworks]

Hi, I thought of that but the spindle is hardened. I'm afraid a reamer might leave a poor finish and then it'll be even worse to resolve. I think the way to go is to grind it, but I don't have a tool post grinder. I have plans to make one but time is limited and for now it's on the back burner. Cheers!

[@willrobertson7778]

@@nbrworks That's an important point I'd overlooked - if it's hardened trying to use a reamer could end badly. There are sets of tools to measure hardness so those might give an indication of whether it's hardened or not (at the expense of leaving a small scratch mark). One option - this is a guess: There's silicon carbide abrasive powder available in different grain sides that can be mixed to a paste - or can be bought as ready-made silicon carbide paste. One option might be to buy something with a cheap but accurate Morse taper on it then put some silicon carbide paste onto that and use it as a hone to grind down the headstock taper to the right angle - it would damage the bought Morse taper - and it might be necessary to use two - but if it's cheap enough it might be worth it.

[@Arthur-ue5vz]

That sucks! That means you're going to have to regrind your spindle bore! That sounds like an undesirable chore. Good luck! 😊

[@nbrworks]

"Going to have" sounds a bit strong, ahah 😅 I'll try to survive not using it, at least for now. The spindle is hardened and it's only in its original form once... before I try to change the taper I have to be sure of what I'm doing. But for now I think I can work my way around it by turning a soft center in the chuck everytime I want to turn between centers. I might even make a jig to get the compound position correctly for that, quicker. And for collets, I have a 5C chuck... so fingers crossed I can hold off for some time that 'undesirable chore' 😁 thanks for stepping by!

[@Tome4kkkk]

Hi again! Could you please share your metro equipment list? I'm especially interested in the prisms, dial indicator stand and the mitutoyo 156-502 base. Dial indicators I already have :) I'd love to get the prisms from a seller who also has 1-2-3 blocks... Living in mainland Europe.

[@nbrworks]

@@Tome4kkkk yes, I can look into that and get back to you soon.

[@Tome4kkkk]

@@nbrworks I'd really appreciate it. I'm going to put it into my purchase queue.

[@Tome4kkkk]

@@nbrworks Noting down a link to your site for now...

[@nbrworks]

I've been busy lately. I'll add them to the website when I have the chance.

[@Tome4kkkk]

@@nbrworks I was more about pinging you rather than rushing :) I've just watched a video by Blondihacks. While prisms are at the top of my list she recommends collet chuck. I think she means usefulness for machining while I need prisms for checking shafts. To be honest a perfect solution would be a fixture with two adjustable conical dead ends, adjustable in vertical direction as well. I made something like that already to check my drill press spindle...

[@be007]

hello, i bought the same 250x550 lathe but have a problem when motor runs in reverse, what is the maximum rpm in reverse on your lathe ? thanks ! cheers ben.

[@nbrworks]

Hi. Mine has the same issue. In reverse it only goes up to half or even a third of the revs. I've been working around it, knowing one day I'll probably swap the motor (the contactor noise is annoying). But if your lathe is brand new, reach out to the seller and see what they have to say.

[@be007]

@@nbrworks i dont think the motor is the problem but de programming of the software because in forward the motor and rpm,s doing good, the only problem is in reverse, so motor is good, pwn controller is bad (i think 🙂). i have reached out to the seller, but no answer yet. kind regards, ben.

[@warrenmaker798]

Funniest video on you tube,,, inc the title 🤣😂

[@yak-machining]

Maybe your lathe bed is twisted

[@nicolaschampagne8988]

regrind it 😈

[@Sigmatechnica]

yeah that does kinda suck :(