• This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn more.
  • PLEASE: Read the FORUM RULES BEFORE registering!

Coaxial Indicator Problem

BRIAN

Global Moderator
Staff member
H-M Supporter-Premium
#1
Just lately I have had cause to question the accuracy of my coaxial indicator. so as a test I held the stem in the lathe 3 jaw. and placed the end of the short feeler ( the most accurate ) against the tool post zeroed the indicator -- the cross slide dial and the digital read out. fed in the cros slide to give one division on the Co ax -- said to be .0005" offset,
And the dial and digital read out read about .0025" this did not look good. so to get better accuracy I went to 10 divisions and divided the result by 10 and came up with .0027"per div the long feeler gave a figure of.0064" per div.
P1013299.JPG

The information with the indicator defiantly says :0005" per div offset.

img195.jpg
So confused and just a bit annoyed I placed a dial indicator against the body of the co ax and lo and behold one division on the co ax is .0005" on the indicator. IT's the dial resolution not the indicator.
How is that for a useless piece of information??
P1013301.JPG


The manufacturers have forgotten about the ratio of the feeler arm. Test yours you may be surprised.

I am waiting for a reply from the suppliers
Brian is not a happy boy.
 
Last edited:

Bob Korves

H-M Supporter - Premium Content
H-M Supporter-Premium
#2
The value of each increment on the dial changes with the length of the feeler probe. The actual value of the off center indication does not need to be even looked at to center a part. Just move the handles until you get the least movement on the dial. Pay no attention to the numbers.

What were you hoping for beyond that?
 

Bob Korves

H-M Supporter - Premium Content
H-M Supporter-Premium
#3
A co-ax indicator is not even on the list of tools I want to acquire. A test indicator stuck in a collet, or for bigger work a test indicator on a Noga arm in a collet does the centering job just fine. I suppose if I found a co-ax indicator for dirt cheap I would buy it, but it would mostly stay buried in my tool box while I reached for other quicker and/or more accurate solutions.

Edit: Another issue is that the blurry needle caused by the high rotation speed does not let you see if the range of needle motion is because of offset or because of surface imperfections. By turning an indicator around slowly you can see if there is a nick in the surface or a burr sticking up causing the needle deflection, which can be ignored while sweeping the rest of the diameter. Using the coax, when you reduce the overall needle movement you are centering on the highest point and the lowest point of the surface, often not what you are looking for...
 
Last edited:

BRIAN

Global Moderator
Staff member
H-M Supporter-Premium
#5
Yes I do know how to use a co ax. And do understand geometry.
Looking at the Blake chart it says that at zero radius with the small feeler (as I tested it) one division is .0005" it is not, it is .0027"on this indicator. and this is at it's most accurate. With the long feeler if I get the wiggle down to one division I can still be .0064" out.
Obviously the Blake is in a different ballpark as far as accuracy
You only find this out when the parts you have spent days making do not line up as you think they should.
Brian.
 

Bob Korves

H-M Supporter - Premium Content
H-M Supporter-Premium
#6
OK, you were using the numbers and directions on the co-ax dial for meeting tolerance. I certainly understand that and, yes, the tool should meet the specs given in the product literature. Unfortunately, in the real world tools do not always meet the specifications or our expectations. Again, that is why I use a trusted DTI, set up correctly, to indicate holes and lots of other surfaces. I can also see the idea of using a co-ax indicator for doing production or serial work where it would be a lot quicker than a DTI. Thanks for the product report. Do you have a brand and model name we can tie it to, so we don't buy one like yours?
 

BRIAN

Global Moderator
Staff member
H-M Supporter-Premium
#7
Hi bob I am still waiting for the answer from the supplier before I give any names I feel I must give the man a chance to answer. but he has had my query for 10 hrs so I am not holding my breath.
Brian
 

Bob Korves

H-M Supporter - Premium Content
H-M Supporter-Premium
#8
Good on you, Brian! If more of us would call them and let them know that they have a problem, and that we will tell many other about it, maybe they will change their ways.

Also my apologies, I did not catch at first that you were a global moderator and probably had a clue... 8^)
 

wyldboar

Iron
Registered Member
#9
I don't know a lot and read these forums to try to pick up some information, so I have a question . It looks like your test only checks for half of the indicated sweep and without the other half you would not be able to find the accuracy of the indicator. What am I missing ?
 

John Hasler

Active User
H-M Supporter-Premium
#10
I don't know a lot and read these forums to try to pick up some information, so I have a question . It looks like your test only checks for half of the indicated sweep and without the other half you would not be able to find the accuracy of the indicator. What am I missing ?
I think you're right. I thought that a coax indicator was primarily a nulling device anyway, though.
 

BRIAN

Global Moderator
Staff member
H-M Supporter-Premium
#11
WY thanks for your question.
We do not need to use the indicator to see if it reads as advertised just push the feeler the correct amount and see if the dial agrees.

John. Yes the coax is indicating the null. The point is how wide is the null. On my indicator when the short feeler is used the dial is pushed by a Class One leaver of about 5 to1 ratio. so even when the in it's best mode. at zer0 diameter with the effort at 90° the accuracy is degraded by that amount.The longer the feeler the worse the result is
Brian
 

wyldboar

Iron
Registered Member
#12
I am still not sure I understand, if a 5 to 1 ratio than .0005 would equal .0025 of linear movement, is that correct ?

Sent from my SM-P600 using Tapatalk
 

BRIAN

Global Moderator
Staff member
H-M Supporter-Premium
#13
Yes correct. sorry to cut you loose but it's one in the morning and I am off to bed. see you later
Brian.
 

mark_f

Global Moderator
Staff member
H-M Supporter-Premium
#15
I kind of hate to step in this , but my understanding of these coax indicators is, they are not for measuring, they are for centering only. That meaning they are strictly for centering on a ID or OD of a part or centering on a hole, in which case the numbers are for reference only. I know I have a Best Test indicator with a long and short arm. With the short arm , the numbers are accurate, but if I use the long arm the numbers are off by a factor of 2. They tell you this in the literature and the long arm is not for measuring but more for lining up on a hole or OD where it is for convenience. I understood, and I could be wrong, but a coaxial indicator is not for measuring ( that is why I don't own one now). That is my opinion.
 

wyldboar

Iron
Registered Member
#16
As I understand it BRIAN thinks he has an error with his co-ax indicator and was using a linear measurement to verify if there was an error or not , I was trying to understand the process he was using.

Sent from my SM-P600 using Tapatalk
 

BRIAN

Global Moderator
Staff member
H-M Supporter-Premium
#17
Hi Mark . thank you for joining the conversation, all contributions are welcome.
WY you have it right I was trying to establish if the resolution of the indicator was as stated under the best conditions. Short feeler . Zero diameter. effort at 90°. I do not use the indicator to "measure".
But the thread took off down another path .
The point is my indicator has a resolution 5 times worse than is stated in the instructions and on the dial.
I am interested to see if other (Quality) brands have the same difference. so is any one prepared to do this simple test that takes all of 10 minutes, Or do we just say Bad resolution is not a problem.
Brian.
 

wyldboar

Iron
Registered Member
#18
Well I set things up the same as your pictures, my dro and dial have been checked with a known accurate test indicator and I ran the same test as described about 5 times and the dro averaged .0072 . I didn't use an indicator to measure movement at the end of the arm like you did, not sure it matters? My co-ax is by SPI , if that helps, I have used it for other set-ups and it was always accurate for my needs. You may need more accuracy than I do.

Sent from my SM-P600 using Tapatalk
 

BRIAN

Global Moderator
Staff member
H-M Supporter-Premium
#19
Hi WY
You are correct the second test is not relevant, thank you for running the test
So using the short feeler moving the Coax one division takes .0072 movement of the cross slide
Thank you for your time and interest.
Any others willing .
Brian
 

Bill Gruby

Global Moderator
Staff member
H-M Supporter-Premium
#20
Small comment here. I see someone put the Blake video up. Do not try to compare the Blake with any other Coax. Most are not anywhere near the accuracy and quality of the Blake. If you really want to test them you need both a Blake and an off brand. Yes I am a tad biased on this, I have used many and the Blake is the winner hands down in my opinion.

I can understand the tests and reasons for them. In the end, the indicator with the finest resolution will be the winner It's hard to use something when it is not to spec.

I wish you good luck in resolving this issue Brian.

"Billy G"
 

BRIAN

Global Moderator
Staff member
H-M Supporter-Premium
#21
Hi Billy G,
Thank you for your input.
I am doing this not only for my benefit but because it affects all owners of Coax indicators,
Most owners just say "OH" I am happy with mine, But have never tested it. So I do not find there answers very useful.
Are you happy with what the manufacturer tells you ?? Think VW ??
The test is simple.
Short feeler.
Zero radius.
Ensure the feeler travel is horizontal.
Load the end of the feeler against the tool post a little with the cross slide.
Zero all .
Move the cross slide untill the indicator moves 10 divisions.
Take the cross slide reading and divide by 10.
Compare with what the manufacturer says.

Now who will be the first person with a Blake or any other "Quality" indicator to set the standard.
Brian.
 

rgray

Active User
H-M Supporter-Premium
#22
Mine is an MHC. Don't know much about it the label is mostly unreadable, but the mhc is formed into the plastic case. No name on the indicator or on the face. Says .0005 per division.
10 divisions is .025 on the cross slide. I don't have a fancy dro just an igaging but both it and the cross slide dial agree at .025
 

BRIAN

Global Moderator
Staff member
H-M Supporter-Premium
#23
Thanks to Wyldboar for mentioning the other half it made me think , this is the total movement of the feeler and will allow us to compare resolution , but to compare the axis offset we must divide the answer by 2. this still leaves my indicator having to move .00135" for one division instead of the quoted .0005"
 

BRIAN

Global Moderator
Staff member
H-M Supporter-Premium
#24
Thank you rgray divide the reading by 10 gives .0025"resolution to compare the axis offset divide by 2 = .00125"
Can any one find anything I have missed in this test.
 

chips&more

Active User
Active Member
#26
OK, I hope I have all the answers for you. I have a Blake CO-AX. I just ran a test like Brian did and to my surprise, I got about the same results. I have had/used a Blake for decades and never noticed the discrepancy in feeler travel verses dial readings. I have only used the CO-AX for centering not for measuring anything. Cosine error, different feeler lengths, feeler angle and whatever else can effect quantitative measurement. "The manufacture also points this out". But why such a gross error? I can see the copy indicators having the same error BECAUSE they copied the design of the original Blake indicator. The error was built into the copy from the original, huh. OK, so now where is this stated dial reading/graduations coming from??? I found out that the space from the bottom of the CO-AX case to one of the feet on the spinning feeler holder gives the correct readings on the dial, imagine that! Kinda meaningless, but that’s the deal. Just use the CO-AX as a centering tool and not for measuring and you will be good to go…Dave.
 
Last edited:

Bob Korves

H-M Supporter - Premium Content
H-M Supporter-Premium
#27
Do the dials on the various coax indicators show the value? Does it say something like .001"? Or are they just undefined numbers on a scale?
 

mark_f

Global Moderator
Staff member
H-M Supporter-Premium
#28
I have a question. I just looked at several different brands of co-ax indicators and NONE of them say what each division is. They all say on the dial "AXIS OFFSET .0005"
What does this mean?
 

chips&more

Active User
Active Member
#29
I have a question. I just looked at several different brands of co-ax indicators and NONE of them say what each division is. They all say on the dial "AXIS OFFSET .0005"
What does this mean?
My Blake manual says: “Dial Graduation is divided such that each division represents .0005 inch of AXIS OFFSET (deviation from true location) when feeler is tracing a diameter of two inches. (varies slightly with feeler angle).”
 

mark_f

Global Moderator
Staff member
H-M Supporter-Premium
#30
My Blake manual says: “Dial Graduation is divided such that each division represents .0005 inch of AXIS OFFSET (deviation from true location) when feeler is tracing a diameter of two inches. (varies slightly with feeler angle).”
That is what I am talking about. It does not say each division equals .0005" of movement. it says it is axis offset,:dunno: what ever that means. offset from what and where.
I give up. I'm not going to get one of these.:confused:


I know for a fact, that the dial indicators are accurate if you use the arm straight out. when you bend it ( or I should say angle it) the accuracy changes. if you use the auxiliary long arm that comes with it in the case, the accuracy goes out the window. the same is true with the Starrett indicators. they are accurate alone, but if you start using the fancy attachments that comes with it in the case, the accuracy is shot and by a lot depending on the length of said accessory. I think we are chasing our tail here and to get back to the original question, I think there is no clear answer as they don't say that .0005" is readable movement....it is offset .... and for me that makes the waters muddy. That is what I think.
 
Last edited: