grinding

[harry a saake]

Question for all the lab people, why is it with all of the super modern equipment in the marketplace, most of the labs still have a problem grinding the OC of a flattop in the correct place, which on a 28 mm lens would be in the middle at 14 mm and at whatever the requested ht. is above the line.

Years ago in the old labs , during the era of glass lenses, we used nothing more then a pink slip of paper that had the base curves and the final generator thickness, and the lenses were laid out exactly on the markers as to the way you wanted them to come out, and they usually did.

today i see 8.5 by 11 sheets of paper that have more numbers then a hooker on 43rd st, some of it understandable, but the rest which seems to be a lot of unneeded garbage.

i understand some of these systems use prism rings to aproximate where the OC will wind up being, which seems to me to have somewhat limited accuracy and i have to wonder why anyone would use that type of system, is it so hard to lay out a lens 3 in and 3 down?, or is it all about mass production and the hell with accuracy.

[jameselex]

Hi Harry:

when I block the lens "off center", I get the oc exactly as I wanted, no problem, with " on center" blocking , most of the time they are good, some times the OCs are off a bit, some time a lot. my Rx software is playing with me all the time, you enter the same job twice and it will give you two different results! and look at the curve you know it is not right, so I beat it a lot with my head.

[rbaker]

For the same reason that the Fryalators at McDonalds have timers. Welcome to the brave new world of unskilled people producing mediocre products for ignorant consumers.

[HarryChiling]

It is easier for their people to make mistakes when they have to block lenses differently from job to job. Consider how many mistakes can happen if your people are not paying attention one day. Therefore geometric center then using prism to move the optical center is the prefered method. A dummy can be taught how to block geometric center, and the lab software will try to do the decentering for him using prism. Their was a discussion on the use of prism to decenter on this site not too long ago.

[harry a saake]

I,m amazed that we all seem to be in agreement that these systems are dumb at best, Harry, of course i know what you mean but back in the old days we used to give a person a specific job and they did not move to the next level until they demonstrated they could do that. worked pretty well and put out some well rounded opticians,but that was then. I would love to hear what shambaum has to say about this. you around bob

[Jacqui]

is it so hard to lay out a lens 3 in and 3 down?, or is it all about mass production and the hell with accuracy.

You hit it on the head. we block on the OC and have no problems at all. One problem most labs that do this have is not paying attention to the overhang. Rermember the old cribbing pliers?? the thought behind them was to even the weight distribution so that one side wouldn't fine out faster than the other. We still will, at times, do a deblock after generating and reblock on geometric center to fine and polish.

[Darryl Meister]

Assuming that your software is reasonably accurate, the biggest issue you would probably face with on-center (geometric center) blocking is in how you're determining the location of the geometric center. If you're using a marking device to dot the geometric center, for instance, the seg inset and drop of the lens blank must be exactly right, which doesn't allow for any manufacturing variance. On the other hand, if you use an actual lay-out blocker to set the seg at the specified seg inset and drop, your results should be every bit as accurate as with off-center (optical center) blocking -- especially if you're not cribbing the lens after off-center blocking. However, this extra step negates some of the benefit to on-center blocking.

[Chris Bowers]

What happens on your flattop, vertically, is that when we grind the back curve that’s anything but parallel to the front surface we induce prism in the bifocal. If there isn’t a lot power differential between the eyes this prism is pretty much bilateral. In the horizontal meridian due to the inset we also have an induced error. We can grind the lens to eliminate this error but it produces cosmetically ugly jobs on lenses that are different powered right to left. Most people measure the seg location and then assume that there is a direct correlation with the bifocals OC.

The work tickets of the modern lab are incredible in accuracy and all of the information is pertinent to the fabrication of the job. We never did have the brainpower back then and even today to hand write calculations for all of the new materials and styles of today.

I believe that both manufacturing and retailers produce the best value that the consumer has ever seen. I do not have any facts to back this up but comparing apples to apples and factoring for inflation the commodity product you spoke of is probably less expensive today. I am a lab person and very proud of what we do.

[RT]

Darryl:

Using a layout blocker would be accurate only if the semi-finished lens blank had zero power. If the lens blank has power (i.e. front and back curves are different), then the fact that the seg is not at the optical center of the lens means that there is some amount of prism at the seg location when looking through the lens during layout. Remember, eyeglass lenses work because they bend light, and we can't pretend that doesn't happen in the raw semi-finished blank. So when you look through the lens blank to lay it out, the seg is not actually where it appears to be.

This prismatic effect is most noticeable on either very flat or very steep base curves for most lens designs. Some blocking systems, such as the Gerber Step One / Eclipse blockers have the ability to compensate the layout image to offset this prismatic error. However, many older manual layout blockers do not have this ability.

If you were laying out a typical ST28 on a 10 base front curve (with a 5 base back curve), the prismatic layout error would be about 1 mm--already straining ANSI tolerances. Bases closer to 6 have less error, but the error is still there and is real.

[Darryl Meister]

Good point, Rick. At least as long as the layout grid is viewed from the back side.

[jofelk]

Question for all the lab people, why is it with all of the super modern equipment in the marketplace, most of the labs still have a problem grinding the OC of a flattop in the correct place, which on a 28 mm lens would be in the middle at 14 mm and at whatever the requested ht. is above the line.

"grinding the OC of a flattop in the correct place"

You request an optical center height of 6mm above the seg (ft28), and the Rx is -2.00 OD, +2.00 OS. We surface the pair and the OC on the finished lenses IS 6mm above the seg, reading the near Rx, 14mm into the seg results :
.................OD> 4^ base down
.................OS> 4^ base up
You can do the math and figure where the center is. I'll do a slab off.

[hipoptical]

Question for all the lab people, why is it with all of the super modern equipment in the marketplace, most of the labs still have a problem grinding the OC of a flattop in the correct place, which on a 28 mm lens would be in the middle at 14 mm and at whatever the requested ht. is above the line.

"The computer said it's right..." and of course, computers ALWAYS have the correct data input and mathematical formulas. We also know that the monkeys pushing the buttons NEVER line ANYTHING up wrong, and they ALWAYS push the right button. I don't know what you're talking about, Harry.




OK, maybe I'm just a tad sarcastic. The REAL answer is that the monkeys who push buttons today are in no way a comparison to the professionals who used to actually be trained in the field of optics and opthalmic lens production.

[RT]

Question for all the lab people, why is it with all of the super modern equipment in the marketplace, most of the labs still have a problem grinding the OC of a flattop in the correct place, which on a 28 mm lens would be in the middle at 14 mm and at whatever the requested ht. is above the line.

Really??? You want the OC to be directly above the center of the seg? No inset? All the time??? I guess you just want us to ignore those PD values that some opticians send in.

Maybe, just maybe, those "monkeys" with 20 years experience in the lab put the OC where they do because the "baboon" ordering it didn't specify where he/she really wanted it.

:D

[Jim Stone]

That's it! I am getting out of the Optical business.

[hipoptical]

That's it! I am getting out of the Optical business.

Now Jim, you know this business is the REAL Hotel California....
You can check out any time you like- but you can never leave!

[Fezz]

Now Jim, you know this business is the REAL Hotel California....
You can check out any time you like- but you can never leave!

Yeah..Great. Just as long as there is "pink champagne on ice".....


Fezz
:cheers:

[harry a saake]

RT, when i posted the first post i meant to say that the oc should be 1.5 mm from the middle of the segment line and at what ever height is called for. If you want a 63/60 pd and you inset the segs where they measure 60mm, then each oc should be 1.5 from the center of the segment, giving you your 63/60.

What i was complaining about was all to often, i will see these oc,s in or out way more then what they should be, which i find strange with all the modern computerized equipment they have available today, they cant seem to grind the distance oc,s of flattops any better then we did without all that equipment back 40 years ago.

[shanbaum]

I,m amazed that we all seem to be in agreement that these systems are dumb at best, Harry, of course i know what you mean but back in the old days we used to give a person a specific job and they did not move to the next level until they demonstrated they could do that. worked pretty well and put out some well rounded opticians,but that was then. I would love to hear what shambaum has to say about this. you around bob

Sorry, Harry, I didn't see this post 'til just now. I've been busy driving on the wrong side of the road.

If you're talking about the old-style production lab - which, for the most part, did work as you describe it (that was my intro into the business, long, long ago, in a galaxy far, far away) - I don't think it was effective at producing opticians. It produced people who gained experience at dealing with a set of tasks; over time, as they encountered a variety of special cases, they developed rules of thumb for handling those cases. But I met a lot of people back in those days who had risen through the ranks of the lab who knew everthing about getting jobs through the shop and next to nothing about optics.

That lack of knowledge was, for a very long time, reflected in the technology we used for fabrication, and it still is to some extent - but to a far lesser extent. I certainly wouldn't characterize modern systems as "dumb" (well, with a few exceptions). But even these still have a number of features which, if not carefully managed, will lead to less-than-accurate results.

With regard to your specific complaint, there are several places in a "normal", "modern" system where unwanted prism can be introduced. As Rick noted, almost all blockers (both finish and surface) suffer from through-power effects; while we have a compensation feature in the Step One and Eclipse blockers, my impression is that most users don't use it. A bit of debris on the blocking ring, block, or generator chuck can do it; but the worst culprits are the finers and polishers. Even when maintained properly (including the frequently-neglected pins and block centers, where present), the forces involved in any of these are way less than ideal. It's also common practice to fine (especially) for very short periods of time - periods much shorter than the time it takes the machine to produce an even amount of work across the entire surface (to the extent that they do that at all).

One of the side effects of the technology that has been and is being developed for free form surfacing may be to reduce these kinds of effects even on toric surfaces. All of the various systems in this area depend on making only very small changes to the generated surface - such that if the changes were of such a magnitude as to introduce prism, the lens might be unusable due to localized power errors. That's because these systems use "small-aperture" polishing tools - tools that are smaller than the lens surface. Getting those tools to perform a sufficiently equal amount of work across the lens surface is "the trick".

Your experience in the B&L lab organization may well have been different; I always had the impression that, for one thing, those labs frequently had more knowledgeable opticians around than did the "big box" labs in which I grew up.

[shanbaum]

What i was complaining about was all to often, i will see these oc,s in or out way more then what they should be, which i find strange with all the modern computerized equipment they have available today, they cant seem to grind the distance oc,s of flattops any better then we did without all that equipment back 40 years ago.

I should add: think back, Harry; 40 years ago you could block a flat-top on the PRP and still not have any overhang on the fining tool with those 55x60mm blanks, and you fined for 15 minutes...

[harry a saake]

Hello robert, greetings and salutations, i hope we will see you out at vision expo west.

Well as far as the 15 minute fining goes, yes of course you had to as it was glass, and i might add if you remember it was frequent to have overhang, especially with your square duallens(for the newbies thats an executive today), but we had a marvelous little machine, in fact made by coburn, which was an automatic cribber, and of course that solved a lot of problems, dont know if they do that any more today, and of course that was better then hand cribbing which i have also had that pleasure.

basically it sounds like your stating that most all of this is due to operator negligence, and very poor maintenince, which i think is a sad commentary if thats the case.

is there no way robert that these systems can be set up where the operator would have to use all the steps you stated that were being left out?

[shanbaum]

is there no way robert that these systems can be set up where the operator would have to use all the steps you stated that were being left out?

Sure - you can have procedures in place that make it more likely that process variables (including machine states) will be measured and controlled routinely. That's the least costly "way".

[harry a saake]

Robert, being that simple, then why would anyone not want to do that, cost, ignorance, poor training??

[shanbaum]

Robert, being that simple, then why would anyone not want to do that, cost, ignorance, poor training??

Yes.

[RT]

Thanks, Harry and Robert, for painting all labs and lab workers as cheap, ignorant, and poorly trained. If Shanbaum hadn't bought the drinks Friday night, I might be mad.

Since the typical rate of jobs remade due to lab error is 1/3 the rate of jobs remade due to Dr/Optician error, what adjectives would most properly describe those professions?

I think that the better observation is that the task is different than it was 40 years ago. There are different requirements in terms of materials (i.e. more than one) and their cost, designs, and expected turnaround time. I know it makes for better reading to try and blame Corporate Optical for all the ills of the industry, but that just doesn't fly. Stats on lab remakes vs. Dr/Optician remakes haven't changed in 10 years, whether privately or corporately owned.

This thread is starting to sound like Mel Gibson out for a weekend drive ("Labs caused all the wars in this world...").

[shanbaum]

Gee, I thought I had described the machinery used in such a way as to possibly get in trouble with my employer, so I don't think it's fair to suggest that I was trying to shift blame. The accusation of "poor training", for instance, is as much our fault as it is the customer's. And I certainly wasn't trying to place blame on "corporate optical". I like coporate optical.

With regard to cost, I was thinking that a lot of the difficulties that labs face can be tied to the requirement that such innovations as we produce are constrained by a requirement of backward-compatibility. It's all well and good to come up with a new technology, but what happens when you tell a lab that they'll have to replace all of their (fill in the blank)? And of course, the belief that labs won't spend a lot of money on new technology has only recently been challenged, by the general acquisition of coating plants, and free-form systems. We thereby end up being stuck with some inherently lame technologies.

And with regard to ignorance, I was thinking of the fact that labs generally have not implemented the kinds of process controls that I've seen, for instance, in film processing - where there was usually a substantial (and costly) effort put into measuring both process parameters and results pretty much constantly. Are there any labs that routinely run test Rx's and measure their deviations from established norms (the way film processors run test strips)? Maybe there are, and I just haven't seen it, but would you agree it's not the rule?

Of course, that may not even be the best way to deal with process control, and perhaps labs are doing more in that area than they used to, but it certainly used to be the case that the discpline of process control was simply absent from the ophthalmic lab. I was always impressed when there was a well-enforced maintenance plan.

I don't mean any of this as a criticism so much as an explanation.

And as far as doctors' re-do's go, I refuse to dig a deeper hole.