Shopsmith Forums

As long as your laser device has the necessary accuracy and repeatability, offsets can be used to adjust for the kerf cut without using the teeth on the blade as a target. Same would be true for distances less than 6". You just need stable know reference points to use as offsets. I would still urge a more rigorous test to ensure the laser does have a .001" accuracy.

I do agree.

I have used mine, as an experiment, to do Shopsmith verification tests. I found it to be quite suitable despite naysayers comments.

I have the Bosch GLM165-10, that is the cheap unit they sell for $69 at Amazon, I was able to get one for $40 so I don't have a lot invested.

I too had better then 20/20 vision for many many years. Then had to use reading glasses and that was good for a lot of years but finally it was full blown glasses. But as I mentioned I do not see a dot it is line, I have that with other laser things I own so I'm pretty sure it is me.

Now with my laser it gives the feet and inches and then a small display of the 32nds as the smallest reading. So a dimension might read 1' 8" 7/32. The 32nds is actually a smaller display so I'm not as happy as I could be with that.

You end up with a floating measurement sometimes when it is having a hard time deciding just how far away the point is. It will say 1/32" then 1/16" then 1/32" then 1/16" but when you push the button it picks one of them to display.

I think that floating thing tells me that I would not trust it too much down to that level.

Ed

I did a quick search of currently available laser measuring devices. Only a few measure in increments of 1/32" or smaller which I think is a minimum requirement for this discussion. The very best one I found advertised an accuracy of 1mm or 1/24'(.04").

https://www.amazon.com/Spectra-Precisio ... 1-fkmrnull

The review article that shows all the devices is at:

https://www.toolsofthetrade.net/hand-to ... ters_o?o=1

I think that kind of accuracy would allow you to use it for most cutting operations but not for tool calibration. I might look into integrating some kind of fixture on my fence that targets the saw blade. Same would be true for the stop block on my miter saw fence. In some cases it might be better than an reading a fixed scale or ruler but I think it would be a close call.

Just found one that is better from Dewalt that has 1/32" accuracy and reads out in 1/32" increments

https://www.homedepot.com/p/DEWALT-330- ... lsrc=aw.ds

reible wrote:I have the Bosch GLM165-10, that is the cheap unit they sell for $69 at Amazon, I was able to get one for $40 so I don't have a lot invested.

I too had better then 20/20 vision for many many years. Then had to use reading glasses and that was good for a lot of years but finally it was full blown glasses. But as I mentioned I do not see a dot it is line, I have that with other laser things I own so I'm pretty sure it is me.

Now with my laser it gives the feet and inches and then a small display of the 32nds as the smallest reading. So a dimension might read 1' 8" 7/32. The 32nds is actually a smaller display so I'm not as happy as I could be with that.

You end up with a floating measurement sometimes when it is having a hard time deciding just how far away the point is. It will say 1/32" then 1/16" then 1/32" then 1/16" but when you push the button it picks one of them to display.

I think that floating thing tells me that I would not trust it too much down to that level.

Ed

Ed,

I don't think it is your eyes. I watched part of a YouTube video of the Bosch GLM165-10 where he shines it on a stucco wall and there is clearly a line being thrown. I am no expert on optics, but I think the model you have doesn't have a coherent dot like mine does. I have a 2 axis laser level and it tends to throw off that sort of faint line with a somewhat solid dot in the middle. What I saw in the YouTube video definitely reminded me of that, so I am guessing similar optics are used in your GLM165-10.

sehast wrote:I did a quick search of currently available laser measuring devices. Only a few measure in increments of 1/32" or smaller which I think is a minimum requirement for this discussion. The very best one I found advertised an accuracy of 1mm or 1/24'(.04").

https://www.amazon.com/Spectra-Precisio ... 1-fkmrnull

The review article that shows all the devices is at:

https://www.toolsofthetrade.net/hand-to ... ters_o?o=1

I think that kind of accuracy would allow you to use it for most cutting operations but not for tool calibration. I might look into integrating some kind of fixture on my fence that targets the saw blade. Same would be true for the stop block on my miter saw fence. In some cases it might be better than an reading a fixed scale or ruler but I think it would be a close call.

Just found one that is better from Dewalt that has 1/32" accuracy and reads out in 1/32" increments

https://www.homedepot.com/p/DEWALT-330- ... lsrc=aw.ds

I looked at the Spectra and it does look interesting. Seems to be sold as a much more rugged unit than mine is, i.e. more for heavy duty construction versus light construction use. That model is over 2x more in cost to units like mine, but does have better accuracy. I couldn't find a suitable video on the Spectra that showed the resolution though and they don't spec it either. That accuracy should be accumulated over the entire range, so in practice your accuracy would be much better (yes, this needs to be confirmed by someone, but I haven't had time yet). I think what is more important is the resolution, or index that the sensor will tick off, e.g. mine will increment in 1/32" increments. If the accuracy is much better than the resolution (as stated in the article I posted earlier so that is my running theory), then from a measurement standpoint all we are concerned with is what resolution, i.e. how small can resolve a measurement. I would be curious to know if the Spectra can resolve units smaller than 1/32". If so, that could make that unit compelling....assuming we can use it for shop/machine setup tasks as has been debated in this thread.

I have to jump into this quagmire.

Resolution - the smallest detectable change.

Accuracy - no deviation from 'true'.

These two things are NOT interdependent, but rather independent from each other.

One would 'hope' that accuracy deviation is smaller than resolution. However that ain't necessarily so.

One can create something with discrete steps that establishes the resolution.

If those steps are not more accurate then the steps, accuracy becomes the defining factor.

Not clear since a better description escapes me tonight.

An example is the Wixey type angular indicator. It's stated accuracy is less precise than it's resolution. .1 degree resolution vs .2 degree accuracy.

JPG wrote:I have to jump into this quagmire.

Resolution - the smallest detectable change.

Accuracy - no deviation from 'true'.

These two things are NOT interdependent, but rather independent from each other.

One would 'hope' that accuracy deviation is smaller than resolution. However that ain't necessarily so.

One can create something with discrete steps that establishes the resolution.

If those steps are not more accurate then the steps, accuracy becomes the defining factor.

Not clear since a better description escapes me tonight.

An example is the Wixey type angular indicator. It's stated accuracy is less precise than it's resolution. .1 degree resolution vs .2 degree accuracy.

I don't think I stated that accuracy and resolution were interdependent for this laser measure device, but if I did that was not my intent. I don't know enough about the internals of this device to make that assumption, which is why I have been treating them as separate entities. In the general case, can we say that accuracy and resolution are always independent from each other? Resolution is dependent on what type of detector/sensor is used and if this sensor/system has multiple ranges, then the resolution is dependent on that range selected. Often times accuracy will be dependent on the range selected, so in some respect accuracy and resolution are often "linked", so if one changes, the other may change as well. So there is some shared causality between them. Perfect example is the specs from the DMM sitting on my bench (see below). From this spec sheet, they don't look "independent" to me.

So, is this a quagmire yet? I mean maybe if I had brought up Schrodinger's cat and asked if it was dead or alive while I was trying to measure the length of its tail using my laser measure, then maybe it has become a quagmire...

Digital = all bets off!(±LSD)

More to the point, an accuracy that is a % of the reading by definition will be directly related. However it would be linear though not constant.

Resolution would I assume to be constant over the range(scale marks if analog).

The quagmire comment was because this resolution vs accuracy babbling has previously occurred.

When digital the least significant digit will always introduce a ± 1/2 factor.

The resolution is a function of the display.

The accuracy is dependent upon the analog to digital conversion as well as the analog stuff itself.

BUT 'we' digress.

P.S. I do not like cats dead or alive. (unless they are still kittens)

JPG wrote:Digital = all bets off!(±LSD)

More to the point, an accuracy that is a % of the reading by definition will be directly related. However it would be linear though not constant.

Resolution would I assume to be constant over the range(scale marks if analog).

The quagmire comment was because this resolution vs accuracy babbling has previously occurred.

When digital the least significant digit will always introduce a ± 1/2 factor.

The resolution is a function of the display.

The accuracy is dependent upon the analog to digital conversion as well as the analog stuff itself.

BUT 'we' digress.

P.S. I do not like cats dead or alive. (unless they are still kittens)

JPG, thanks for taking my comments in stride and having fun with it (as they were intended). I love your +/-LSD comment instead of +/-LSB. As an Analog guy, that is just too funny.