surfacing - question for lab managers

[essegn]

We use the conventional production line from SATISLOH (generator + smoothing and polishing machine).

Our daily production is about 250 lenses per day.

We have power off errors at high index material about 20%. We do the high index material on 0.12 dpt step flat back tools.

I know that to achieve more efficient production in power off errors is to produce it on more exact flat back tools steps (0.06 or 0.03).

Due to low volume of production it has no sense (because of price and place for storage) to create the additional flat back tools (like 0.06 or 0.03) so i would like to know the following:

Does anybody have any ideas how to figure out this situation ?

I found out one company that procudes exact flat back tool from teflon and by the all high index materials they will cut the exact tool. In middle europe i was not able to find the cheap teflon material.

My imagination about price for 1 piece of any material that we will produce from it the exact curve is maximum 2 Eur per piece.

I checked the flat back tools at few comapnies and the lowest price for tools is 5 Euros.

Thank you in advance!

Peter

[HarryChiling]

We use the conventional production line from SATISLOH (generator + smoothing and polishing machine).

Our daily production is about 250 lenses per day.

We have power off errors at high index material about 20%. We do the high index material on 0.12 dpt step flat back tools.

I know that to achieve more efficient production in power off errors is to produce it on more exact flat back tools steps (0.06 or 0.03).

Due to low volume of production it has no sense (because of price and place for storage) to create the additional flat back tools (like 0.06 or 0.03) so i would like to know the following:

Does anybody have any ideas how to figure out this situation ?

I found out one company that procudes exact flat back tool from teflon and by the all high index materials they will cut the exact tool. In middle europe i was not able to find the cheap teflon material.

My imagination about price for 1 piece of any material that we will produce from it the exact curve is maximum 2 Eur per piece.

I checked the flat back tools at few comapnies and the lowest price for tools is 5 Euros.

Thank you in advance!

Peter

This may be too simple of a fix, but what is the index of your tools? Are you compensating the tools for index?

[WFruit]

If you haven't already, make sure all the lens data (true front curve, sag, etc.) in your lab software is accurate. Manufacturers will sometimes change their lenses without letting anyone know.

I'm curious to know if you see errors more with any particualr index (1.60, 1.67, 1.74, etc.) or if there's any more/fewer errors on spheric vs. aspheric.

And Harry is, of course, correct to check the tooling index as well. And of course, check to make sure the curves on the tools themselves are accurate.

We're also running Satis Loh (gods help us) although all of our curves are cut to .10 on the conventional generators, and .01 on the Digital generator.

[HarryChiling]

If you haven't already, make sure all the lens data (true front curve, sag, etc.) in your lab software is accurate. Manufacturers will sometimes change their lenses without letting anyone know.

It is always a good idea to sag your lenses before surfacing, you'll find that in many cases this can help to resolve power issues and catch warped lenses before precious time and resources are wasted. A digital sag gauge is a must, also make sure to caliper your lenses as thickness is always off from the average lens manufacturers data this will prolong your cutters life as you won't have the tool crash into a lens that is thicker than expected.

[essegn]

thanks guys..

used tooling index is 1.498. On the conventional SATISLOH line use the only aspherical semi finished materials and we have more power off errors at 1.6 like higher indexes.

I'm sure that some of the base curves have to be adjusted, but i would like to know the how good results could we achieve if we produce the high index material on tooling index 1.498 in 0.12 steps.

WFriut - do you use the same tooling index and all the lenses are in 0.1 tolerance?

What is your power off breakage at high index materials ?

We have the digital SAG meter from Coburn, but i really dont know how to measure with that device because if i measure the same curve more time i will always get different values. The difference is not so much, but i has quite big influence on final diopters. I figured out the best way to produce it and see how the result diopter looks like.

[essegn]

Could somebody explain to me for what is tooling index good for ? For me, it is obvious that for 1.498 material is good tooling index 1.498 and for 1.6 material is 1.6 tooling index.

If we use tooling index 1.498 for all material indexes what does it mean? Is there any possibility to change the tooling index for higher one? If so, what does it mean in real? The new flat back tools?

[WFruit]

Could somebody explain to me for what is tooling index good for ? For me, it is obvious that for 1.498 material is good tooling index 1.498 and for 1.6 material is 1.6 tooling index.

If we use tooling index 1.498 for all material indexes what does it mean? Is there any possibility to change the tooling index for higher one? If so, what does it mean in real? The new flat back tools?

The simplest solution is to tell your lab software what your tooling index is and it will compensate it's curve calculations for whatever the lens index is.

For example, all of our tools are cut in .10 diopter increments, 1.498 tooling index. We use the same tools for all materials, all indexes. Our lab software knows what the tooling index is and compensates its curve calculations based on that. Our off power breakage accross all indexes is very, very low. We usually only run into issues when a vendor changes the true front curve and doesn't bother to inform anyone.

See here for an article on tooling index: http://books.google.com/books?id=Zl4...page&q&f=false

[Wes]

Oh my. This is going to take a lot more explaining than I'm willing to do on a blackberry. If no one beats me to it, I'll write out an explanation of tooling indices for you when I get home.

[Wes]

Hey cool, you beat me to it.

[essegn]

For example, all of our tools are cut in .10 diopter increments, 1.498 tooling index. We use the same tools for all materials, all indexes. Our lab software knows what the tooling index is and compensates its curve calculations based on that. Our off power breakage accross all indexes is very, very low. We usually only run into issues when a vendor changes the true front curve and doesn't bother to inform anyone.

[/URL]

we use 0.125 step tools.. What is difference between yours tools and ours? the only 0.025 diopter difference ?
I would like to know which accuracy we can achieve with 0.125 step tools.

I have one more question: If i set aspherical semifinished material to rX program like spherical semifinished material (if i ignore the aspherical topografy and the only radius is set), does it have influence on final power in the optical centre (in our case geometrical centre) or it will only cause the worse peripheral vision?

How do you measure the front curve? By sag meter and then you will shift the front curve according to result ?

What is your internal tolerance for tools? For example toll has to be -8.53. What is the bigest acceptated tollerance? Or it has to be the exact -8.53?

[HarryChiling]

Well here goes a simple explanation if your using lets say a 1.67 lens blank and tooling in 1.498 (0.12 steps) as an example:

Rx: -4.00 sph

Using the simplified lensmakers equation and assuming you need to use a 3.00 base lens:

The front curve is often given as a TBC (True Base Curve) which is referenced to a index of 1.53, so if you measure it with a lens clock it will read what's on the box. The base curve given in the materials index is:

D(material) = (0.67/0.53)*3.00 = 1.26

Ignoring the thickness the back curve power would be:

Back Curve = 1.26 - (-4.00) = 5.26

So the back curve you need is a 5.26, but we're not done yet, the back curve needs to be computed into the tool index that you will be using:

Tool Curve = (0.498/0.67)*5.26 = 3.91, the closet tool you'll have to that is the 3.87 so that would be the proper tool to use.

If your software has 1.53 as the tool index then back surve would be calculated to:

Tool Curve = (0.53/0.67)*5.26 = 4.16, the closet tool would be a 4.12, so you can see where there is a disparity.

Your software like mentioned should allow for you to input the tool curve, you should have this set to the 1.498 value since thsoe are the tools you have available. Also make sure you have pad thicknesses compensated for in the software or on the tool. If your using aspheric lenses it won't matter what the aspheric value is if the lens has a sag value on the box since that sag value should correspond to the same sag value of the lens, however that sag value is used for calculating thickness not power, the TBC is going to be used for the calculation of power.

The best suggestion is to have an experienced lab man look at your set up and program the values accordingly. Also since your tool index is in 1.498 and the most common or default tool would be in 1.53 that may be where you issue is. In th case of lower index lenses you are putting lower powers in the lenses and the material index is closer to the 1.53 so the error might not be as evident, however in higher indices you will have a larger disparity and higher powers so you'll see a bigger error. Without really seeing your set-up it's hard to diagnose or guess at what the problem might be, but it is definately worth checking into.

Sorry beat you to it Wes, although I am notorious for getting th emath wrong so the example may have a kink in it's armor. ;)

[WFruit]

Oh thank goodness I don't have to explain it. :p

It turns out that I did remember it right, but with lab software being what it is today, so much of the math is taken care of by the computer. All you have to do is make sure the software has the correct values and it takes care of the rest.

To answer one question, using tools cut in 0.125 dipoters, you should expect to cut lenses that are accurate to 0.125 diopters of what you need. As you can see in the example above, the actual calculations are down to 0.01 diopters, and then you round to the nearest available tool. The smaller the increments your tools are cut to, the more accurate you can cut your lenses, due to the value of the tool being closer to need needed back curve. However, most labs, when processing lenses on conventional equipment (as opposed to digital surfacing) use tools that are cut in 0.125 increments (some will use 0.10 increments, like we do, and I've even seen a few in 0.0625. And some insane, and usually smaller labs, will cut tools individually for each job accurate to within 0.01).

[HarryChiling]

And some insane, and usually smaller labs, will cut tools individually for each job accurate to within 0.01

Not really insane for a small lab, cutting tools on foam can save the space required for keeping tool racks full of tools and it increases accuracy. Sure it costs a little more but for small in office labs I actually loved not having to pull tools for every job and since in most small labs high production isn't an issue the extra time needed to cut the tools was not a big deal. If you were to try that in a large production lab then that would be insane since the generator would be running inefficiently by wasteing time, to put the tool in, to cut the tool, and then to take the tool out. These things add up.

I would personally advocate any small operation to use foam laps as the convienience will far outweight the costs. I know a few small labs that have a small raneg of tools and anything outside of that they use the foam tools to make this is a great set up as well. I would always mark my foam laps as well for how many cuts I had in them and when your jobs are arranged properly you could get 3 sometimes 4 tools from one foam block.

[essegn]

Not really insane for a small lab, cutting tools on foam can save the space required for keeping tool racks full of tools and it increases accuracy. Sure it costs a little more but for small in office labs I actually loved not having to pull tools for every job and since in most small labs high production isn't an issue the extra time needed to cut the tools was not a big deal.

I would personally advocate any small operation to use foam laps as the convienience will far outweight the costs. I know a few small labs that have a small raneg of tools and anything outside of that they use the foam tools to make this is a great set up as well. I would always mark my foam laps as well for how many cuts I had in them and when your jobs are arranged properly you could get 3 sometimes 4 tools from one foam block.

Exactly as you wrote.. Due to low production value we have a time to produce the exact tools on each high index order.

Do you know any producer of foam tools you can recommend me? Because i have heard that foam tools are not good, because the material tends to heat up during smoothing and polishing operation which causes distortion on the final lenses.

How to work with foam tools that we are intend for making the exact tools? As you wrote that you are able to cut 3-4 exact tools. For example, if you need the tool of -5,00 and then you will need the tool of -10,00;-6.00. So in this case you are not able to do that on one foam tool.

What is the more efficient way how to decide how many foam tools to have in order to be able to produce 3-4 tools from one foam tool?

I have one more question: We have an analog SAG meter as well. So if i measure the SAG height of semi finished 1.68 it has to be recalculated to the radius (and then to TBC) with index 1.53 ?? and than the TBC from 1.53 index to 1.498 for CR39 ??

I do not understand why the SAG height is good the only for thickness calculation. I have been taught that you measure the SAG height. In case that you measure the CR39, you will calculate the radius (for example: SAG 1,68 ; material index=1.498) and then if you have radius calculated, you will calculate the real diopter of front curve.

Is it correct method ?

[essegn]

Rx: -4.00 sph

Using the simplified lensmakers equation and assuming you need to use a 3.00 base lens:

The front curve is often given as a TBC (True Base Curve) which is referenced to a index of 1.53, so if you measure it with a lens clock it will read what's on the box. The base curve given in the materials index is:

D(material) = (0.67/0.53)*3.00 = 1.26

Is that correct _?

[shanbaum]

While it is common for lenses to be marked with 1.53 curves in the U.S., that is not the rule in the rest of the world, where they are usually marked in material diopters. Similarly, while 1.53 laps have been the rule in the U.S. until recently, in most of the ROW, laps have been index-specific. That's changing due to the proliferation of indices; labs are using laps cut in higher indices and/or smaller increments - either of which increase the "resolution" of the lap set. I've always been a proponent of material-index-specific laps, although the new conformable-lap systems obviate the issue altogether.

While some lenses are marked with sag values, these are usually just conversions from dioptric curves to sags at 50mm; they can be used for power calculations, when the calculator requires a sag to be input, but are not useful for thickness calculations on aspheric lenses.

Harry's calculation was incorrect; 0.67/0.53 = 1.26; multiply that times 3.00 and you get 3.79 - which means, a curve specified as 3.00 diopters in a 1.53 index has an actual surface power of 3.79 if the material index is 1.67. However, it's likely that in your location, you needn't concern yourself with that 1.53 index at all.

Using 1.498 laps cut in increments of 0.125 for higher index materials is going to be problematic. While your software can find the closest lap in the set to the curves required, it can't change the laps - the increments are what they are. On 1.67 materials, the intervals between your 1.498 laps cut in 0.12D are 0.17D. Given that your laps are not going to be exact (if they're within 0.02D, you're doing well), it's easy to see where your lap errors alone can be 0.10D. That's before anything else goes wrong.

I recall there having been a fairly extensive discussion on this topic here on Optiboard a number of years ago, but I haven't been able to find it yet. I wish I could; I think we beat the subject to death and maybe then some.

Harry's right about foam laps - they're a great solution for a low-volume lab that processes a wide variety of materials. They do have to be used carefully. So far as I know, they're only available from Gerber.

[Speed]

It is always a good idea to sag your lenses before surfacing, you'll find that in many cases this can help to resolve power issues and catch warped lenses before precious time and resources are wasted. A digital sag gauge is a must, also make sure to caliper your lenses as thickness is always off from the average lens manufacturers data this will prolong your cutters life as you won't have the tool crash into a lens that is thicker than expected.

Not always.

[essegn]

While some lenses are marked with sag values, these are usually just conversions from dioptric curves to sags at 50mm; they can be used for power calculations, when the calculator requires a sag to be input, but are not useful for thickness calculations on aspheric lenses.

I am confused a bit.. if i found the SAG number at 50 mm on the 1.67 blank. Should it be recalculated to true base curve with 1.53 index or 1.67 index?
not useful for thickness calculations ? i am not sure, but if you have front curve in diopters (calculated from SAG) and according to set thickness, rX program calculates the final diopter. Could you please explain it to me ?

Using 1.498 laps cut in increments of 0.125 for higher index materials is going to be problematic. While your software can find the closest lap in the set to the curves required, it can't change the laps - the increments are what they are. On 1.67 materials, the intervals between your 1.498 laps cut in 0.12D are 0.17D. Given that your laps are not going to be exact (if they're within 0.02D, you're doing well), it's easy to see where your lap errors alone can be 0.10D. That's before anything else goes wrong.

Sorry, but i dont understand.. why the interval between 0.12D tools is 0.17D ? why not 0.12 ?
what does it mean that lap errors alone can be 0.10D ?

[Speed]

I am confused a bit.. if i found the SAG number at 50 mm on the 1.67 blank. Should it be recalculated to true base curve with 1.53 index or 1.67 index?
not useful for thickness calculations ? i am not sure, but if you have front curve in diopters (calculated from SAG) and according to set thickness, rX program calculates the final diopter. Could you please explain it to me ?




Sorry, but i dont understand.. why the interval between 0.12D tools is 0.17D ? why not 0.12 ?
what does it mean that lap errors alone can be 0.10D ?

Many high index lenese are aspheric. Do not sag an aspheric lens. Get the published true curve from the manufacturer for power calculations.

[Darryl Meister]

As the others have pointed out, your lap tools will definitely be an issue. This may or may not be your only issue though. You really need to investigate the issue further in order to understand the potential sources of your off-power errors, especially since the problem seems to occur more frequently in higher index materials (I have a paper that discusses some of the issues associated with refractive available for download from OptiCampus.com).

A few questions to consider:

1. Are the power errors relatively systematic or random? Are the errors biased toward plus or minus powers? Errors due to your tooling increments should be fairly random, for instance.
2. Does it occur with specific lens suppliers, lens designs, or lens materials more frequently than others? Have you verified that your software has the most recent lens specification data from the lens supplier? Do they only occur in high-index materials or do they actually occur in high powers, which are often more commonly produced in high-index materials?
3. If you're using an automatic focimeter, is your device calibrated to the correct reference wavelength and the correct refractive index of the lens material? Have you compared the results of multiple devices?
4. Is pad thickness a potential issue if you are using different fining pads for high-index materials? What about stock removal rates?
5. Have you noticed any blocking, generating, or fining waves on the lens surface after surfacing?
6. Have you verified the calibration of your equipment and lap tools, using something like a lap tool gauge?

Etcetera. Some of the other lab guys could probably provide some other tips and potential problem areas to consider.

I am confused a bit.. if i found the SAG number at 50 mm on the 1.67 blank. Should it be recalculated to true base curve with 1.53 index or 1.67 index?

Does your surfacing software accept a sag value? If so, the software should perform the necessary conversion for you.

not useful for thickness calculations ? i am not sure, but if you have front curve in diopters (calculated from SAG) and according to set thickness, rX program calculates the final diopter. Could you please explain it to me ?

The "sag" value or true curve supplied with aspheric and progressive lenses is not the actual physical height of the lens blank over a given diameter. It is only describes the "theoretical" curvature of an equivalent spherical surface over a very small region of the distance zone that is used to calculate the surface power of the lens blank over this region.

An aspheric front curve, for instance, may be steeper or flatter than the theoretical base curve specified by the lens manufacturer for power calculations. The actual, physical height of the lens blank over a given diameter, which is necessary for thickness calculations, may vary significantly from the sag value specified for power calculations.

Sorry, but i dont understand.. why the interval between 0.12D tools is 0.17D ? why not 0.12 ?
what does it mean that lap errors alone can be 0.10D ?

If your lap tools have been calibrated to a refractive index of 1.498, the interval between each tool is also based on this index. If your lap tools are marked in 0.12 D increments, this increment only applies if the material has the same refractive index.

If the index of the actual material is higher, the difference in curvature between your lap tools results in a greater difference in actual surface power. For a 1.66 lens material, lap tools marked in increments of 0.12 D in an assumed index of 1.498 produce an actual difference in surface power of (1.66-1)/(1.498-1)*0.125 = 0.166 D. Since your potential rounding error is half of this value, your lens powers will be off as much +/-0.08 D, just from your lap tools.

This is actually one advantage to a system based on the US tooling index of 1.530 or any higher refractive index, since it results in more lap tools, as Robert noted. Basing lap tools on the refractive index of the actual material is a good idea as well, since you could stock the cylinder powers in only 0.25 D steps, significantly reducing your lap tool inventory for cylinder powers. But this would obviously require multiple lap tool sets, which is apparently impractical in your situation.

Given your investment in 1.498 tooling, however, I suspect that the easiest short-term solution for increasing the accuracy of your lap tools would be to fill in some of your lap tools with a few 1.498-index tools in 0.06 D increments, focusing first on the prescription ranges that most frequently result in off-power errors for you. This will obviously cut your potential rounding errors in half.

Best regards,
Darryl

[HarryChiling]

While it is common for lenses to be marked with 1.53 curves in the U.S., that is not the rule in the rest of the world, where they are usually marked in material diopters. Similarly, while 1.53 laps have been the rule in the U.S. until recently, in most of the ROW, laps have been index-specific.

I didn't know that, you learn something new every day.

While some lenses are marked with sag values, these are usually just conversions from dioptric curves to sags at 50mm; they can be used for power calculations, when the calculator requires a sag to be input, but are not useful for thickness calculations on aspheric lenses.

I had it flip flopped but yes they work i one sense but not in another, wow 1 year away from a lab and the information goes quickly.

Harry's calculation was incorrect; 0.67/0.53 = 1.26; multiply that times 3.00 and you get 3.79 - which means, a curve specified as 3.00 diopters in a 1.53 index has an actual surface power of 3.79 if the material index is 1.67. However, it's likely that in your location, you needn't concern yourself with that 1.53 index at all.

I knew there was an error in there in the math, just too little time to look it over thoroughly, good catch.

Using 1.498 laps cut in increments of 0.125 for higher index materials is going to be problematic. While your software can find the closest lap in the set to the curves required, it can't change the laps - the increments are what they are. On 1.67 materials, the intervals between your 1.498 laps cut in 0.12D are 0.17D. Given that your laps are not going to be exact (if they're within 0.02D, you're doing well), it's easy to see where your lap errors alone can be 0.10D. That's before anything else goes wrong.

I recall there having been a fairly extensive discussion on this topic here on Optiboard a number of years ago, but I haven't been able to find it yet. I wish I could; I think we beat the subject to death and maybe then some.

I remember that, should be a search away for the poster.

Harry's right about foam laps - they're a great solution for a low-volume lab that processes a wide variety of materials. They do have to be used carefully. So far as I know, they're only available from Gerber.

Great post.

How to work with foam tools that we are intend for making the exact tools? As you wrote that you are able to cut 3-4 exact tools. For example, if you need the tool of -5,00 and then you will need the tool of -10,00;-6.00. So in this case you are not able to do that on one foam tool.

Do it in order:

5.00
6.00
10.00

Flatter to steeper, mark the tools with a grease pencil across the center, you want to remove stock on each new tool you make so if the grease marks are visible you should take the tool down another 1/2mm to 1mm. They also make harder plastic laps which can be used multiple times, we used to use those for common powers in another lab I worked and when you ground the lap down to thin you would mark the power's on the side and place it in inventory with no further cutting. The harder plastic laps had to be checked for accuracy more often but they still offered a permanent solution rather than a foam disposable solution.

[shanbaum]

Here's the thread I mentioned:

http://www.optiboard.com/forums/showthread.php/7157

[Darryl Meister]

Great post

Yeah, I'm going to start calling Robert "The Tool Master." ;)

Best regards,
Darryl

[shanbaum]

Of course, you will not be the first...

[Darryl Meister]

Of course, you will not be the first

LOL. I think I'm referring to a slightly different context than Mrs. Shanbaum was, Robert. :bbg:

Best regards,
Darryl