Homemade DRO's?

[johnnyd]

http://www.yadro.de/

Found this over on the "General" forum & thought I would share.:D

[Davo J]

With glass scale DRO's coming down in price every day and the accuracy they give. It makes you wonder if it's worth making one, thats unless it's the challenge of building it your after.
Davo

[macona]

And then there is the delay in the updates from the digital calipers. There is a fast mode where it updates faster but I think you loose accuracy when that happens.

[John Stevenson]

I had a Shumatech fitted to my lathe for a while but got so fed up of it dancing on the display I swapped it for a glass scale unit.

Another lathe problem is they are good for 0.001" reading but on a lathe because of the diameter it then becomes 0.002" at best, 0.004" at worse and to be honest you get get better on dials.

They were acceptable when DRO's were $$$$$ but it fast reaching the point where the kludge up are equalling the glass scale imports.

.

[loose nut]

The cheap digital scales don't compare well to the glass scale DRO's

[The Artful Bodger]

Just how good are DROs on a lathe anyway? Can you skip the caliper and micrometer stage?

I have cheap caliper type scales that I operate from tensioned wires to indicate saddle and cross slide movement and these are at least as reliable as the knob graduations but I still need a micrometer or gauge to finish, are expensive DROs any better than that?

[Jim Shaper]

My heidenhain DRO on the lathe is every bit as accurate as a caliper once you calibrate it to the cut you're taking.

For instance, you're boring a hole. You set your tool, take a test cut, measure your diameter, set the DRO to the measured diameter. Next cut so long as your tool is rigid enough for the depth, or you're not taking such a light cut as to rub and not cut - it'll be spot on the money. Just like setting up tools on a cnc, the repeatability is only as good as the data input.


I couldn't work anywhere near as fast as I do without my DRO.

[macona]

Yep, a good dro is an incredible time saver and scrap saver. No counting handwheel turns! And like Jim says, with a good solid machine you can repeat 10ths on multiple parts. And turning shoulders is so much easier since very few machine even have dials on the carriage handwheel.

[The Artful Bodger]

I would not know just how accurate my cable arrangement is as so far I do not even have a micrometer but in one test I turned a band on a shaft, retracted the cross slide and moved the carriage along then using the cable DRO I reset the tool position and turned back to join the first band. The join was visible but I could not feel it, I figured in that instance that it was 'pretty good' but I have no idea of the numbers.

Maybe I will be able to justify a good DRO one day but meanwhile even this simple system is worth it for the way it avoids gross errors.

[loose nut]

Tiffie is right in one respect on this subject, a DRO will get you 99% of the way there but the Micrometer gets you the rest of the way.

[The Artful Bodger]

Tiffie is right in one respect on this subject, a DRO will get you 99% of the way there but the Micrometer gets you the rest of the way.

If that is the case I will never spend money on a 'real' DRO but just keep on with this steel cable arrangement and measuring tools to finish.;)

[loose nut]

There is a lot more to take into account when using DRO's then just reading the settings and going. I found out the hard way that DRO's will amplify your ability to see any errors you make in the way you use your equipment.

1. Gibs MUST be locked when the axis in question is not being used otherwise movement that doesn't show up on the DRO's can take place throwing off your actual movement compared to the DRO reading.

2. wore or poorly adjusted gibs can add unwanted table movement that can throw off actual readings compared to the DRO readings.

3. Machine spindles on "good" worn old equipment or new cheap equipment may not rotate accurately around the spindle axis leading to errors in the size of milled slots etc. Mill undersized, measure with the mic and adjust the DRO for a final cut, don't just adjust the DRO to the cutting position and think it will be right on the money.

4. you can use the bolt circle function, as an example, of the DRO to lay out, well "bolt circles" but the spindle and or the drill chuck can/will be off and give you an inaccurately placed hole. See Number 5 this is relative.

5. You may not have to worry about backlash in the lead screws if you are using DRO's but if you don't make all your movement to holes or cutting positions on any given axis from the same direction when you move the table it can twist and throw the relationship between positions of different holes or cut lines out.

DRO's are very useful pieces of equipment and one should get the best they can afford but they have to be used properly or errors will get in the way. These are milling machines, frequently of dubious quality, not jig boring machines they are not really meant to hold tenths or better.

[MuellerNick]

YADRO doesn't have jumping numbers, because it has an built in adjustable filter. The filter defaults to no filtering when values change and filters when the value is to some degree stable.
Fast mode switches off the internal filtering of the scale. As there is an external filtering, that doesn't affect YADRO's display.
The resolution you get with filtering is well above 0.01 mm. Certainly, it doesn't increase accuracy. Accuracy of those scales is about 0.03 mm / 150 mm.

I designed YADRO, when glass scales were very expensive. And good quality scales still are.


Nick

[The Artful Bodger]

Considering that machinists got by for more than a century without DROs I will try to follow their example and save the budget for something more desirable or essential.

[MuellerNick]

got by for more than a century without DROs I will try to follow their example and ...

Dispose all HSS-tools, coatet or uncoated carbides, forget about modern steels like stainless or free cutting, calculate hole patterns with sine-tables (pocket calculator not allowed!), ... oh, and forget about the electric motors, they are devils work! ;)

Or how can you allow to have internet-access as a Amish?


Nick

[loose nut]

Considering that machinists got by for more than a century without DROs I will try to follow their example and save the budget for something more desirable or essential.

That is a valid point but be careful not to throw out the baby with the bath water.

DRO's are very useful tools but you have to use them properly just like anything else . Once you get use to them they are a great productivity enhancer plus they (most, not all of the older ones) have very useful built in functions. Just double check the readings with a mic to make sure things are the way you want them.

[loose nut]

Dispose all HSS-tools, coatet or uncoated carbides, forget about modern steels like stainless or free cutting, calculate hole patterns with sine-tables (pocket calculator not allowed!), ... oh, and forget about the electric motors, they are devils work! ;)

Or how can you allow to have internet-access as a Amish?


Nick

Why do any of us still use manual machines when there is CNC to use.

Just because we don't want to use the latest and greatest doesn't mean we have to go back the 1800's. Doing it the "old fashioned "way, for a hobby, has value in itself.

[Jim Shaper]

Loose nut, you've got some seriously crappy equipment if any of those arguments against a DRO are throwing your parts off by more than a thou or 2.

Due to backlash in the lead screws, you have to approach every dimension from the same direction if you're using the dials. However, backlash doesn't present itself with glass scales. The statements you've made expose a clear lack of understanding of the positional stability the scales provide. If twisting of the bed or saddle was so severe as to relocate a coordinate more than a few tenths, your bed would likely sit cockeyed anyway and you wouldn't be able to hold a dimension with or without a DRO.

Same with a spindle - if you've got more than a few tenths run-out, you're not cutting straight lines even if the bed was tight on the gibs. It would also probably rattle pretty bad.

[The Artful Bodger]

Dispose all HSS-tools, coatet or uncoated carbides, forget about modern steels like stainless or free cutting,
Those things are not expensive, some are even free to me.



calculate hole patterns with sine-tables (pocket calculator not allowed!), ... Ummm... thats what I would use anyway, my pocket calculator is only four function.



oh, and forget about the electric motors, they are devils work! ;)
Devils work or not they are cheaper than a steam engine and my wife refuses to do more than a few hours on the treadmill.




Or how can you allow to have internet-access as a Amish?
Not a problem, I can live with it!:D

[The Artful Bodger]

Why do any of us still use manual machines when there is CNC to use. I have spent much of my life writing software, is that what CNC is about/


Just because we don't want to use the latest and greatest doesn't mean we have to go back the 1800's. Doing it the "old fashioned "way, for a hobby, has value in itself. Totally agree, the object of this retirement hobby is to learn something. Of course if I had more disposable money I would look at DROs but right now they are not even on the front page of the list.

[loose nut]

Loose nut, you've got some seriously crappy equipment if any of those arguments against a DRO are throwing your parts off by more than a thou or 2.
.

My equipment will do better then that but most DRO's read to at least 0.0002" and most of the equipment that we have isn't good enough for that kind of accuracy. My point was just because you have the dro capability your mill or lathe may not be up to using it. You really need some seriously sturdy equipment in a climate controlled environment to hold that level of accuracy not a cheap Chinese mill in an home shop where the temp changes all the time. They are good enough for a thou. There are other factors that have to be taken into account like the quality of drills and endmills collets and drill chucks that can cause problems that won't show up on a DRO. These have to be taken into account or problems will show up, hence the check with a mic.

I love my DRO's but they are not the be all and end all of machining.

[Jim Shaper]

Anyone who knows how to machine would laugh at holding .0002 outside of grinding or a very tight (new) cnc. Without ball screws, you just ain't gonna.

For +/-.002", you are saving yourself a lot of fuss and bother going with the electric display option, even on an old worn machine. So that means the cheap DRO options will greatly aid in user ease of producing parts to spec.

[loose nut]

However, backlash doesn't present itself with glass scales. The statements you've made expose a clear lack of understanding of the positional stability the scales provide. If twisting of the bed or saddle was so severe as to relocate a coordinate more than a few tenths, your bed would likely sit cockeyed anyway and you wouldn't be able to hold a dimension with or without a DRO.

This is true again it's all matter of how accurate you are trying to be, if you are trying to use a mill as a jig borer, which probably wouldn't happen in industry anymore but might in a home shop it could be a problem. most of us don't have the equipment for it.

Not having the gibs tightened down can cause a shift of several thou. on the axis that isn't being moved (moving the X axis screw with the y axis gib not tight) and will not show on the DRO, found that one out the hard way.

[loose nut]

Anyone who knows how to machine would laugh at holding .0002 outside of grinding or a very tight (new) cnc. Without ball screws, you just ain't gonna.

For +/-.002", you are saving yourself a lot of fuss and bother going with the electric display option, even on an old worn machine. So that means the cheap DRO options will greatly aid in user ease of producing parts to spec.

See post above,I agree but lots of times people will need to hold half a thou especially on a lathe which is perfectly doable, does that mean you trust the DRO and not check it with a mic.

You seem to think I'm against DRO's I'm not I think there great but there is a learning curve like everything else. That's the point I'm trying to get across.

[Mcgyver]

Anyone who knows how to machine would laugh at holding .0002 outside of grinding or a very tight (new) cnc. Without ball screws, you just ain't gonna.


baloney. i can turn to .0002 when i have and my lathe is neither a grinder or a new cnc....but maybe you meant just milling ?

btw, what has ball screws to do with it? a precision ground acme thread's lead error can be better than/same as ball screws :confused:

[Jim Shaper]

Ballscrews don't have backlash. So your precision ground lead screw has the potential to be pulled into the cut where a bed that's being held by a ball screw doesn't. Simply varying cutting force could cause this.

If you can hit .0002 every time, you're a better man than I am. :rolleyes: When the work heats up, you're chasing dimensions. CNC's can be programmed to deal with the conditions that occur in the work because they do everything the same every time. Unless you're a robot, you're not spitting out .0002 work tolerances. There's no effen way.

Yes, it can be done. There's no way you'd make any money doing it every time manually with a cutting edge, the method simply isn't that precise. Cutter mechanics alone will fail you.

[MuellerNick]

where a bed that's being held by a ball screw doesn't.

But you'll never find them on manual mills/lathes. they arent self-locking, you'd always have to stop all spindels from rotating. By hand, with three (cross, top, saddle) you need special abilities. :D


Nick

[Mcgyver]

Ballscrews don't have backlash. So your precision ground lead screw has the potential to be pulled into the cut where a bed that's being held by a ball screw doesn't. Simply varying cutting force could cause this.

that's operator error, not having things tight enough, or climbing when you shouldn't. in the example i gave, a lathe, how does backlash affect things? On a mill same thing....so long as you approach things from the same direction so the screw is loaded on the same thread flank (SOP) backlash won't matter.

I can hit .0002 when i need to on a lathe, its not difficult. Saying that isn't making any production claims, and I'm not doing this for money.

btw, ballscrews do have backlash, albeit less than an acme. To get rid of it antibacklash nuts are used on ball screws as well.

what affects accuracy, aside from clearance in the linear motion mechanism is lead error of the screw, which can be as good for acme as it is for ball screws. precision ground acme's are also a lot cheaper than the same class of precision ground ball screws....anyway, so far as the accuracy of distance moved vs that indicated on the dial/encoder is concerned, a ballscrew holds no advantage over acme of comparable lead error which was my point, so the fact that the cnc has a ball screw isn't what makes it more accurate than a manual with an acme

[Davo J]

http://www.yadro.de/

Found this over on the "General" forum & thought I would share.:D

Getting back to the original topic, I really don't think it's worth making a DRO when there are these sort of cheap prices out there.

http://cgi.ebay.com.au/ws/eBayISAPI.dll?ViewItem&item=220536875097&ssPageName=STRK:MEWAX:IT

http://cgi.ebay.com.au/ws/eBayISAPI.dll?ViewItem&item=220503372459&ssPageName=STRK:MEWAX:IT

From what I have seen, a good vernier type setup are around the same price.

Dave

[KEJR]

Ballscrews can be bought backlash free. Some of the cheap ones are crap and allow for so many microns worth of backlash. The better ones use ground ballscrews with precision balls that are selected to have a certain amount of interference (preload) on the fit. Ballscrews do have a spring constant to them, so if you load them heavily they will be off by a certain amount. This is why a very light finish cut is often needed for high precision work(among other reasons).

As far as readouts and accuracy, I think the problem is when you assume you dial in a position and think that is what your part will be. The trick with any readout is to measure your part when you start getting close to your dimension and then use the readout to make small incremental cuts (using the readout) to chase your dimension. While this aspect of a DRO is not so different than reading a dial, with a readout you still have your zero reference that you can machine your sloppier dimensions to without having to keep measuring the part.

So, in conclusion, a readout doesn't make your machine more accurate but can help you in making parts faster with less measurements and tool locating.

[Mcgyver]

BThe better ones use ground ballscrews with precision balls that are selected to have a certain amount of interference (preload) on the fit. Ballscrews do have a spring constant to them, so if you load them heavily they will be off by a certain amount. This is why a very light finish cut is often needed for high precision work(among other reasons).

the precision ground ball screws can be quite pricey....but that addresses lead error not backlash (clearance) - you still need an antibacklash nut (ie preloaded style)

[KEJR]

The precision ground ballscrews are precisely ground to a certain depth and the bearing selected to have preload, therefore zero backlash. This assumes that you have a precision preloaded thrust bearing as well. Spring loaded anti backlash nuts are another way to do it, certainly. I quoted the Hiwin ballscrews for my bridgeport and they were < $1000 for the precision screws/nuts and block that holds the two ballnuts. I haven't made the investment yet though.