DTI vs DI

Topics include, Machine Tools & Tooling, Precision Measuring, Materials and their Properties, Electrical discussions related to machine tools, setups, fixtures and jigs and other general discussion related to amateur machining.

Moderators: GlennW, Harold_V

Mr Ron
Posts: 2126
Joined: Tue Dec 22, 2009 12:36 pm
Location: Vancleave, Mississippi

DTI vs DI

Post by Mr Ron »

Which is the preferred tool for measuring runout on round stock mounted in a lathe chuck? a dial test indicator or a plunger style dial indicator. It would appear that either one works and although I do possess both types, I always use the DI because it is always handy and the DTI is in it's case and requires assembly before I can use it.
Mr.Ron from South Mississippi
earlgo
Posts: 1794
Joined: Sat Jan 29, 2011 11:38 am
Location: NE Ohio

Re: DTI vs DI

Post by earlgo »

For me it is the DTI. One is always at hand on the lathe for just such occasions. There is a little adapter to hold the stem in the tool holder.
indicator and holder.jpg
And the use is very fast and easy.
indicator in toolpost.jpg
Results may vary.
--earlgo
Before you do anything, you must do something else first. - Washington's principle.
User avatar
GlennW
Posts: 7284
Joined: Sun Apr 15, 2007 9:23 am
Location: Florida

Re: DTI vs DI

Post by GlennW »

I use a DTI unless it does not have enough range, then it's time to use a DI. Either if which can be used with my holder.

I keep it handy and ready to drop onto the QCTP.
INDICATOR.jpg
Glenn

Operating machines is perfectly safe......until you forget how dangerous it really is!
User avatar
BadDog
Posts: 5131
Joined: Wed May 17, 2006 8:21 pm
Location: Phoenix, AZ

Re: DTI vs DI

Post by BadDog »

DTI is cheaper, smaller, and readily available in higher resolution (though you have to watch cosine error if you want accurate quantification). They also have a large variety of easy/fast mounting and positioning/orientation options, often including fine adjustments. They also have a format that makes it convenient to sneak in among things that may be in the way.

Considering all the above, I find I rarely use a DI compared to DTI. Basically only when longer ranges are required, or actual quantity beyond a range of a couple thou (total indicated) is necessary. And that latter point is sometimes mitigated by having a reference option (like gage block stack on surface plate). That way my DIs mostly stay safe and snug in their protective storage. Thus they stay clean, work smoothly, and have little risk of crystal face (or other) damage, but are ready when/if needed.
Russ
Master Floor Sweeper
pete
Posts: 2518
Joined: Tue Feb 10, 2009 6:04 am

Re: DTI vs DI

Post by pete »

Starting from bar stock a .001" DI should be good enough for round stock in a 4 jaw independant since +- .001" shouldn't make a bit of difference. Re-algning a previously machined surface then your part and accuracy requirements dictates what level of perfection is needed. That said there's also super accurate .00002" or better dial indicators and fairly rare .001" dti's if you search. So that's still not a given on which to use and when. They make .0005" DTI's and the .0001". To avoid buying more indicators I bought the 10ths DTI's. I think the DTI's can be positioned a lot easier into tight areas so might use a 10ths indicator then even if I thought + - .001" was good enough.

Robin Renzetti on Youtube showed a custom built top nut on his quick change tool post that had a .375" diameter post on the end so real easy to slap a snug and the arm from a magnetic base on that and then quickly position whatever DI or DTI on the part that I thought was a really well thought out idea. Having one already mounted in a qcth would obviously be even faster though.
SteveM
Posts: 7763
Joined: Mon Jun 27, 2005 6:18 pm
Location: Wisconsin

Re: DTI vs DI

Post by SteveM »

I have a dial indicator mounted to an AXA toolholder. I find it easier to use with the larger dial.

As it is mounted fixed horizontal, I don't have to worry about cosine error, although do you really need to worry much about cosine error when all you are trying to do is center the part? If you are at a 30 degree angle and it says it's centered, then it's centered.

If I needed the precision, I would get out a dti. I have them in thou, half thou and tenth. Need to set up a dedicated AXA toolholder with the DTI mount on it.

Steve
User avatar
tornitore45
Posts: 2077
Joined: Tue Apr 18, 2006 12:24 am
Location: USA Texas, Austin

Re: DTI vs DI

Post by tornitore45 »

A DTI is a good "nulling" tool and sensitive but is awkward to position.
IF the DTI is mounted on a flexible arm holder (NOGA) is difficult to set on center and withing the limited range.
IF the DTI is mounted on a tool holder it can be set once but then it requires a lot of cross-slide cranking to get positioned on the part.
Unless you can center the 4 jaws by eye within the DTI range is a pain to get there due to the limited DTI range.

The DI has a lot more range.
I resolved the cranking issue by making a permanent mounting arrangement on a Quick Tool Holder.
The DT mount on a (radial) slide that has sufficient motion to position the tip anywhere between the spindle axis and the max turning radius regardless of the cross-slide position.

My 4 Jaw centering procedure is as follow:
1) Mount the part by eye by using the ring on the chuck but also looking for the vertical jaw to sit on the high crown of the part when mowing the front (horizontal) jaw. This initial step gets me within 0.15" Total Instrument reading.

2) Place the DT and its slide on the QCTP. Adjust the slide to preload the DT about 0.2" not really critical, just do it.
3) Turn the chuck until the DT shows maximum distance.
4) Crank the cross-slide to zero the DT. You need no more than one turn to reach the closest zero.
5) Rotate the part until the DT read minimum distance. In other words the part is closest to you.
6) notice the reading, divide by 2 and crank the cross slide to the 1/2 error point. This position the DT to read error symmetrically about zero.
7) I place the BIG chuck key in front and the small chuck key/knob in the back. Always push the part from front to back.
The BIG key does the muscle job, the knob in the read just back off and snug up.
8) Now turn the chuck to have one set of jaws horizontal and the DI showing positive deviation ( the part needs to go back).
9) Work both keys to zero the DT. Snug the part but not too tight.
10) Turn 90 degree to the position that reads positive error. Repeat 9).
11) In an ideal world the part is now perfectly centered, reads zero on all 90 degree orientation of the jaws.
12) In reality the part is only weakly clamped and as far off a 0.003" and not even symmetrically.
13) Iteration of the process one more time gets the part where the DT needle just barely move. At this point the clamping is serious, the rear key is not backed up, any small movement is done by torquing the big key on the front jaw. At this point you decide if this is 0.001" job or a 0.0002" job. Anything sloppier could have been done on the 3 jaws but changing to a 3 jaws chuck is worth only if I have many parts to make that have no critical requirements. The 4 jaws is the default chuck.

Seems complicated at it took some time to have the pattern develop into my neurons, now is an automatic set of motions and the part is centered in about two revolutions.
Mauro Gaetano
in Austin TX
User avatar
SteveHGraham
Posts: 7788
Joined: Sat Jan 17, 2009 7:55 pm
Location: Florida

Re: DTI vs DI

Post by SteveHGraham »

Perhaps I am wrong (first time for everything), but I feel like you can see sub-thousandth wobble on a DI just fine when indicating round stock, even if you don't know exactly how big the runout is. A part doesn't have to be wobbling by 0.001" to have a problem which is obvious on a dial indicator, so you should be able to true it up to well under a thousandth. That's pretty good for most work. Usually, you don't really need to know whether it's running out 0.0002" or 0.0003". If the needle is barely moving, you know you're close enough, so you don't need a test indicator's increased accuracy.

I like dial indicators because they're easier to deal with.
Every hard-fried egg began life sunny-side up.
User avatar
tornitore45
Posts: 2077
Joined: Tue Apr 18, 2006 12:24 am
Location: USA Texas, Austin

Re: DTI vs DI

Post by tornitore45 »

Often a piece of unknown lineage or even virgin stock will be out of round 1/2 a thou, at that point is good enough if the needle wiggle a bit.
Mauro Gaetano
in Austin TX
User avatar
BadDog
Posts: 5131
Joined: Wed May 17, 2006 8:21 pm
Location: Phoenix, AZ

Re: DTI vs DI

Post by BadDog »

I have a couple of 0.001 DTIs, along with 0.0005, 0.0001, and 0.00005 (pretty useless). I don't find any of those hard to find at all, and I got all of them fairly inexpensive, so I chose the one best suited to my needs.

No need to worry about cosine error when just minimizing deviation, like centering up in a 4 jaw, or aligning to an axis, only when trying to quantify measurements. Like using a DTI referenced to gauge blocks and the declaring the part is 0.004 off using a short finger DTI and starting already off horizontal. If you want to write down 0.004 deviation on a spec or QA sheet, or more likely to know you need to "take another 0.004 off", there might be better ways of knowing that's an accurate number. Like the use of a carefully oriented DI, preferably on a comparator base. A DI on center and aligned carefully (within reason) horizontal on a lathe cross slide could also be accurate, but it would have to be both on center and reasonably close to horizontal (or rather paralel to ways) with the cross slide to be able to get true accuracy. But with DTI precision or repeatability isn't an issue, but accuracy isn't really the forte of DTIs.

Regarding the fiddly DTI, yes I agree. That's why I prefer to use a reasonably steady and convenient arm like the Noga, with a fine adjust at the DTI mount. That makes it pretty easy and easily good enough for my needs in aligning the vise, mill spindle, or lathe parts on axis and/or centered. I just mount it on the mag base Noga, plop it somewhere conveniently located, lock the arm in an articulation that clears and gets the DTI roughly located, then use the fine adjust to get it in range. Takes seconds, and works great. Sometimes that puts it oriented so I can make adjustments with the slide, and I don't have to swap tool blocks and crank in/out to get it done.

Another mark agains the DI is they are much more likely to have problems with sharper surface deviations. Holes splines and t-slots aren't a big deal at all for a DTI as long as the pivot leads the surface probe. And it can be more help averaging out rough castings too. But a DI even with a fat hemi-sphere tip gets significant side loading unless you set it carefully for very minimal travel. Certainly not an indictment in itself, but again, the DTI takes it in stride. Even if you completely pooch the initial setting all you get is a finger deflection and at worst the body shoved aside.

As I said earlier, I almost never use a DI unless I actually need to know that the number I determine is accurate. DTIs are much cheaper and more forgiving in daily use. So I would very much disagree that DI are easier to deal with for daily use. Even without the mechanical function advantages of a DTI, just the form factor makes it better for my daily uses. But maybe that's just my use, or just my ignorance...
Russ
Master Floor Sweeper
Mr Ron
Posts: 2126
Joined: Tue Dec 22, 2009 12:36 pm
Location: Vancleave, Mississippi

Re: DTI vs DI

Post by Mr Ron »

Thanks. It appears it's 50/50 between DTI and DI. I guess it's whatever works best for you.
Mr.Ron from South Mississippi
User avatar
SteveHGraham
Posts: 7788
Joined: Sat Jan 17, 2009 7:55 pm
Location: Florida

Re: DTI vs DI

Post by SteveHGraham »

Everyone has their preferences, but I like dial indicators when I'm centering round work because I have all sorts of travel to work with. Get the tip of a 1" indicator anywhere close to where it has to be, and I'm all set. I don't have to be within the tiny arc a test indicator covers. Once I get going, if I have to move the work a long way to get it concentric, I can leave the indicator where it is. I don't have to worry about maxing out the dial and moving the indicator to start over, unless I have huge runout.

You also get flexibility as to where the body of the indicator goes, which is nice if you have to work around stuff. It's helpful on the mill when I use a rotary table. Maybe I would feel different if my magnetic bases weren't so crappy.

I've never worried about cosine error when indicating stock for concentricity. You're looking to see whether the dial moves at all, not how much it moves. You don't need to think about the difference between 0.0003" and 0.0003" times the cosine of 5 degrees. My coax indicator came with long tips that multiply the measurements by some wild numbers, much worse than big cosines. The indicator has to be much less accurate with long tips, but I've never thought it until today, because all I was looking for was a steady reading.

For fun, I looked it up. The cosine of 5 degrees is 0.99619469809. Multiply that by one thousandth and subtract it from one thousandth. Not a big number at all. Makes a tenth look like the Sears tower.
Every hard-fried egg began life sunny-side up.
Post Reply