Rivista Italiana di Gemmologia #2: Available NOW in English - September 2017; See Gemological Articles below for full details!
Welcome to the GemologyOnline.com Forum
A non-profit Forum for the exchange of gemological ideas
It is currently Mon Oct 16, 2017 6:18 pm

All times are UTC - 4 hours




Post new topic Reply to topic  [ 162 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6, 7, 8 ... 11  Next
Author Message
 Post subject: Re: “Amateur Gemstone Faceting” book by Tom Herbst
PostPosted: Sun Dec 14, 2014 4:57 am 
Offline
Established Member
User avatar

Joined: Mon Apr 15, 2013 6:47 pm
Posts: 17
I've had the books for a week or so and I also picked up a tip in the first ten minutes of browsing. I'm definitely suggesting the books for our faceting classes-at this price, just about everyone can afford them.

Really, Tom, thank you for the clear and concise approaches that make this book so accessible AND enjoyable.

I'm also singing the praises of Jon Rolf as I polish a wickedly large sunstone with the Greenway. Blessings on those who make our faceting experiences more pleasurable.


Top
 Profile  
Reply with quote  
 Post subject: Re: “Amateur Gemstone Faceting” book by Tom Herbst
PostPosted: Sun Dec 14, 2014 2:31 pm 
Offline
Moderator: Lapidary Arts and Tools
User avatar

Joined: Tue Jun 12, 2007 6:41 pm
Posts: 5475
Location: Massachusetts, USA
Quote:
and decided to sell a PDF on his own site to avoid the Kindle graphics problems.


Confirming that Kindle's graphics situation is hopeless. Currently reading a book that contains some detailed illustrations. The illustrations would have been helpful, but-
As others pointed out, PDF has its own problems.
The book belongs in the shop, anyway.

_________________
www.gearloose.co


Top
 Profile  
Reply with quote  
 Post subject: Re: “Amateur Gemstone Faceting” book by Tom Herbst
PostPosted: Mon Dec 15, 2014 6:14 pm 
Offline
Valued Contributor

Joined: Fri Sep 12, 2014 6:04 pm
Posts: 138
Location: North Carolina
Just got my new books in! Man, they look great, now to start reading!


Top
 Profile  
Reply with quote  
 Post subject: Re: “Amateur Gemstone Faceting” book by Tom Herbst
PostPosted: Sat Dec 20, 2014 1:34 am 
Offline
Valued Contributor

Joined: Sat Oct 25, 2014 11:42 pm
Posts: 135
I just got the books and jumped right to the last section because it interests me the most.

I could shed a tear for what he has done with his true digital angle adjustment system...TRUE DIGITAL! Not facetron "digital", not potentiometer "digital" but actual digital sensor (you could argue this one but its pretty solid), micro controller, and CODE! All well within the grasp of anyone who is willing to learn a thing or two.

I looked through his design, its pretty damn solid (always room for improvement but that goes for anything). With his setup, an incredibly accurate machine (0.009 or even 0.0045 degrees accuracy) could be achieved easily depending on how much you wanted to spend (200$ range now I'm thinking).

And that Britta water filter drip tank...just brilliant. That's one of the first things getting upgraded on my machine.

At those prices it truly is a gift to the cutting community. He has much respect for that (if he had a colored deluxe version I would be interested in it too!).

So much to learn and so little time...


Top
 Profile  
Reply with quote  
 Post subject: Re: “Amateur Gemstone Faceting” book by Tom Herbst
PostPosted: Fri Jan 02, 2015 7:08 pm 
Offline
Established Member

Joined: Sat Dec 01, 2012 5:27 pm
Posts: 41
I have a set of Facet Books that have made my best sellers library.


A picture is worth a thousand words.

Image


Top
 Profile  
Reply with quote  
 Post subject: Re: “Amateur Gemstone Faceting” book by Tom Herbst
PostPosted: Sat Jan 03, 2015 3:40 am 
Offline
Gold Member
User avatar

Joined: Fri Oct 19, 2007 5:04 pm
Posts: 1559
Location: Walnut Creek, CA
We have the same bookshelf. :D


Top
 Profile  
Reply with quote  
 Post subject: Re: “Amateur Gemstone Faceting” book by Tom Herbst
PostPosted: Sat Jan 03, 2015 6:12 am 
Offline
Valued Contributor

Joined: Mon Sep 15, 2014 6:21 am
Posts: 200
Great books, thoughorly engrossed . I like the "triboligist " Comments . Dry , like a dad joke... But , nice to see it in a faceting book . Not the norm . Good to see gearloose gets a well deserved wrap . Here's my adapted eyeware ( thought I was the only one ) so many ways to look at your work , but I'm stuck on a hand loupe... 10x gemoro . But would love it if someone could produce something similar in a headset . I heard someone has just developed a small magnifier( by accident ) that will be accessible to phones and other applications ( mini microscope lens) so will see . [url][URL=http://s1003.photobucket.com/user/kingsolomonsound/media/image-63.jpg.html]Image[/url][/url]love the toothpaste And other cheats .still reading :)
Barney


Top
 Profile  
Reply with quote  
 Post subject: Re: “Amateur Gemstone Faceting” book by Tom Herbst
PostPosted: Sat Jan 03, 2015 2:51 pm 
Offline
Gemology Online Veteran

Joined: Sun Oct 29, 2006 11:43 pm
Posts: 514
Location: North Carolina
Folks,

Could someone with a better understanding of these things take a look at the Digital Facet Angle Encoder info in Vol2 starting on page 391, and see what electrical and encoder stuff is needed, and also available. If possible maybe a shopping list?

I looked up the encoder mentioned on page 398, HEDS-6545, and it is listed as obsolete or just not available. Looking through the specs at sites listed in the book, I don't see the things Tom says are needed in any of the specs sheets, such as Counts per Revolution. The info may actually be there but I don't know enough to interpret it.

I am thinking of putting this setup on an Ultra Tec as I have an older one to experiment on. If I'm successful I'l relate how it was done to all. Imagine a Digital Angle AND a B/W on the same machine = no more excuses, unfortunately.

Thanks for you help, and Thanks to Tom for the books!

_________________
Bob K.


Top
 Profile  
Reply with quote  
 Post subject: Re: “Amateur Gemstone Faceting” book by Tom Herbst
PostPosted: Sun Jan 04, 2015 12:38 am 
Offline
Valued Contributor

Joined: Sat Oct 25, 2014 11:42 pm
Posts: 135
Facetor wrote:
Folks,

Could someone with a better understanding of these things take a look at the Digital Facet Angle Encoder info in Vol2 starting on page 391, and see what electrical and encoder stuff is needed, and also available. If possible maybe a shopping list?

I looked up the encoder mentioned on page 398, HEDS-6545, and it is listed as obsolete or just not available. Looking through the specs at sites listed in the book, I don't see the things Tom says are needed in any of the specs sheets, such as Counts per Revolution. The info may actually be there but I don't know enough to interpret it.



I've done a bit of research on this (I've only skimmed that particular section in the book, I haven't read up to that far yet but I will). The best I could dig up in terms of optical encoders is this one company, "US Digital". It seems that they offer the most economical precision encoders that would be suitable for such a project (put it this way, if you want to go to the extreme, a single ultra precision laser optical encoder can outclass the cost of even the most expensive faceting machine).

Their "S6 Optical Shaft Encoder" is probably the place you want to start in selecting your encoder. Those ones come assembled in one unit, case and all, with a shaft sticking out one end (mind you, prudent use of Tom's favorite adhesive, bathtub silicone, is highly recommended...everywhere on everything that doesn't move. Precision encoders are precision devices!). You are probaly looking want the following options:

2: any shaft will do, match it to the machine manufacture's standard. American machines probably use inch standard, rest of the sane world, metric.
3: include the index, its important for zeroing
4: select single ended output, differential is kinda not for us right now because it increases complexity but significantly improves reliability (we don't need to go that fast)
5: probably default torque because the viscous damping fluid they provide will help absorb high speed shocks which may or may not affect missed encoder counts on the micro controller side.

Now, there are a few complicating factors your encoder resolution selection (well...it is digital electronics, not "magical, solve all your problems" electronics, analog had it worse btw).

The first option is CPR, or Cycles Per Revolution. Its a measure of how finely your encoder can resolve positions. The higher the CPR, the finer your system can read angles. 10k CPR is an insanely fine resolution for an insanely low price. To further sweeten the deal, you can multiply the CPR by 4 to get the "real" maximum accuracy of your system for 40k positions per rev or equivalent of 0.009 degrees of resolution (read his quadrature decoding section, its an awesome explanation).

But there's a few catches here. First of all, encoders aren't perfect and just as in gear systems where you have "backlash", encoders do suffer from this problem too, accuracy will be +/- half a quadrature count in either direction, higher resolutions will have better tolerances. Second, you have the rounding issue. At 10k CPR, multiplied by 4 for quadrature decoding, you have an accuracy of 0.009 but since faceting diagrams are only displayed in hundredths of a degree (which is also rounded) you have a bit of rounding uncertainty to deal with but its minor.

The final but most serious problem is with mostly the Arduino itself. The Arduino can do software quadrature decoding but it can not do it well. It never was designed for this task and thus it falls short. The problem here is that if you swing the arm around too quickly you will overwhelm the poor little Arduino with more pulses than it can track. This means missed readings and your angle measurement is invalid and you need to re-zero it. The state of the art for Arduino encoder decoding is somewhere at 127k quadrature pulses per second, with a 10k CPR, this is 127k/40k = 3.175 rev/second. This means you can rotate at a maximum velocity of 3.175 revs/s before your Arduino will start missing steps. This is fast but if for whatever reason you drop the arm, it could potentially exceed this rotational velocity and invalidate your angle, unfortunately there's no way of knowing if you missed a step so you could be happily cutting away at a totally wrong angle and be none the wiser (until you hit the next meepoint). To further complicate things, the 127k read limit is the absolute max, it means your Arduino will be doing nothing but handling encoder steps and leave you with zero time to do anything else like say display the angle reading.

You can mitigate this problem by lowering the CPR on your encoder to something more sane but then you trade off accuracy.

The easiest way to get around this is to offload quadrature decoding to specialized hardware such as their quadrature decoding ICs (with these you can easily operate in the Mhz range, plenty plenty for what we are doing). Fortunately these are cheap, a few bucks, unfortunately they increase the complexity of the system and you will need some more serious electrical and computer programming skills (not beyond anything of anyone's skill here, as long as you are willing to learn). If you go this route, your system is pretty bulletproof.


Top
 Profile  
Reply with quote  
 Post subject: Re: “Amateur Gemstone Faceting” book by Tom Herbst
PostPosted: Sun Jan 04, 2015 7:38 am 
Offline
New to the Forum or The Quiet Type

Joined: Thu Nov 13, 2014 4:20 am
Posts: 9
Hi All,

Thanks for the insightful comments and feedback regarding the optical
encoder. Some remarks:

Facetor wrote:

> If possible maybe a shopping list?

This appears on the top of page 392, with references to further
details on subsequent pages.

> I looked up the encoder mentioned on page 398, HEDS-6545, and it is
> listed as obsolete or just not available.

I typed "HEDS 6545" into google, and the first hit was for Avago
Technologies, which seem to carry this part (Having said that,
it appears that this company changed ownership from Agilent
sometime in November, and their online listings are still
a bit confused). It seems that Nanotec, the company from
whom I purchased my encoder, no longer carries this
model. The closest replacement part is the NOE2 encoder, but there is
a cryptic message on their website that this encoder is "only
available together with a motor and must be mounted by Nanotec."
Hmmmm.

As Faken2 points out, perhaps the best place to find a suitable
encoder is at US Digital (also see Chapter 15.8.11). I checked the
current offerings, and any of the following kit encoders should work:
E2, E3, E5, E6 (your choice will depend on how you want to mount
it). For what it is worth, the models closest to the one in the book
are E3 and E6. If you order a single unit, expect to pay in the
neighbourhood of $100 for the whole kit.

Table 20-2 on page 398 of the book explains how to configure the
various options (counts per revolution, index channel, single-ended
outputs, etc.). You can find the US Digital products and an online
"configurator" here:

http://usdigital.com/products/encoders/ ... rotary/kit
(just select one of the devices, for example E3, and scroll down to Purchasing)

Please let me know if anyone identifies additional sources of suitable
encoders. I will update the website with this information (look here:
http://www.facetingbook.com/corrections.html)

Faken2 wrote:

> Their "S6 Optical Shaft Encoder" is probably the place you want to
> start in selecting your encoder.

These will work as well. The "E" models assume that there is a shaft
that goes into them (or through them), while the "S" models
incorporate their own shaft, which can be mechanically coupled to the
faceting machine. It all depends on how you mount the encoder. For
my Scintillator head, a through-shaft coupling (E model) worked best.

> mind you, prudent use of Tom's favorite adhesive, bathtub silicone,
> is highly recommended

Yes! Another convert! (Incidentally, during my day job, I work with a
telescope containing a couple of 8-meter (27-foot) diameter, 15 ton
mirrors worth millions each. How are they mounted to the telescope?
You guessed it, good old bathtub silicone...).

> 10k CPR is an insanely fine resolution for an insanely low price.

Agreed. Unfortunately, my encoder model had a maximum CPR of 2000,
which gives 0.045 degrees per step...still pretty darn good, however.
I assume the higher CPR devices will require that much additional care
in mounting and centering (although the encoders ship with a tool to
do this).

> you have "backlash"...encoders do suffer from this problem too,
> accuracy will be +/- half a quadrature count in either direction

True enough, although this opto-electronic backlash is tiny compared to
typical mechanical backlash.

> The final but most serious problem is with mostly the Arduino
> itself. The Arduino can do software quadrature decoding but it can not
> do it well.

I have to (mildly) disagree here. Faken2 is quite right that high
speed rotation can cause the Arduino to lose track, but we simply
don't spin our quills that fast. I worried about this a fair amount
and did a lot of testing, both with an oscilloscope and in the "real
world". As mentioned on page 411, I set the hard stop and despite some
rather rapid and silly looking quill waving, I never managed to lose a
single step. Nor have I had a single instance of an angle error due to
the encoder in five years. And this is with my 2009-era Arduino. The
new boards are faster and more capable. You can google "Arduino
encoder" to get a sense of how people are using them in this
application.

> The state of the art for Arduino encoder decoding is somewhere
> at 127k quadrature pulses per second, with a 10k CPR, this is
> 127k/40k = 3.175 rev/second.

And for the 2000 CPR device, this is 15 full quill revolutions per
second, or sweeping it through its 90-degree operating arc sixty
times per second. That's fast enough to make Buddy Rich jealous...

> you will overwhelm the poor little Arduino with more pulses
> than it can track. This means missed readings and your angle
> measurement is invalid and you need to re-zero it.
> ...if for whatever reason you drop the arm, it could potentially
> exceed this rotational velocity and invalidate your angle

Agreed...but the layout of the encoder "re-zeroes" the angle every
time you lift the quill above horizontal (see Figure 20-29). If you
drop the quill, you will do that at least once to see what is left of
your stone ;-).

> To further complicate things, the 127k read limit is the absolute
> max, it means your Arduino will be doing nothing but handling encoder
> steps and leave you with zero time to do anything else like say
> display the angle reading.

The encoder service is done with interrupts (p. 402), giving them
highest priority to minimize the chance of lost steps. As Faken2
points out, this relegates other services, such as the display, to
lower priority. For what it is worth, aggressive waggling of my quill
does not produce obvious flickering or lag in the display.

> You can mitigate this problem by lowering the CPR on your encoder to
> something more sane but then you trade off accuracy.

Yes! In fact, I find that the 0.045 degree accuracy from a 2000 CPR
encoder is more than enough for published gem designs on real-world
(i.e flexure prone) machines. Note that you can also increase the
effective CPR if you have a fine angle adjustment. For example, the
micrometer head in Figure 20-17 lets me "split" encoder steps by
another factor of about 4, giving ~0.01 degree accuracy.

> The easiest way to get around this is to offload quadrature decoding
> to specialized hardware such as their quadrature decoding ICs

Interesting! I will have to look into those. And please let me know
about your experience with higher CPR encoders.

Best wishes,
Tom Herbst
tom@facetingbook.com


Top
 Profile  
Reply with quote  
 Post subject: Re: “Amateur Gemstone Faceting” book by Tom Herbst
PostPosted: Sun Jan 04, 2015 8:06 am 
Offline
Valued Contributor
User avatar

Joined: Thu Oct 04, 2012 1:57 am
Posts: 138
Location: Germany
Now it's getting interesting :)

I use an UltraTec V2 analog. The problem with my machine is that the readout
is driven by gears and the smaller gear - a brass cheapo - is worn out, causing
considerable slope.
Mounting an optical encoder is tempting, but I am not really sure where to mount it, trunnion axis
doesn't work on the UltraTec, does it?

I am thinking of changing that gears to another ratio (it's 6 to 1 now) to a ratio in the vicinity of 2 to 1.

Let's say I use a ratio of 1.8 to 1 (15 and 27 teeth) and an encoder with 5000 cpr. That would give me 9000 signals per round,
which translates to a reading of 0.01 degree without any rounding.
But still there would be the backlash of the gears plus the optical backlash :(

Any ideas where else to mount the encoder on an UltraTec??


Top
 Profile  
Reply with quote  
 Post subject: Re: “Amateur Gemstone Faceting” book by Tom Herbst
PostPosted: Sun Jan 04, 2015 9:10 am 
Offline
Moderator: Lapidary Arts and Tools
User avatar

Joined: Tue Jun 12, 2007 6:41 pm
Posts: 5475
Location: Massachusetts, USA
Quote:
use an UltraTec V2 analog. The problem with my machine is that the readout
is driven by gears and the smaller gear - a brass cheapo - is worn out, causing
considerable slope.
Mounting an optical encoder is tempting, but I am not really sure where to mount it, trunnion axis
doesn't work on the UltraTec, does it?


Do not drive an encoder through gears. In addition to backlash, there is cyclic error.

_________________
www.gearloose.co


Top
 Profile  
Reply with quote  
 Post subject: Re: “Amateur Gemstone Faceting” book by Tom Herbst
PostPosted: Sun Jan 04, 2015 9:34 am 
Offline
Valued Contributor
User avatar

Joined: Thu Oct 04, 2012 1:57 am
Posts: 138
Location: Germany
Didn't know this.. thanks for the heads up!

edit: This should be the case with the factory-made DAD as well...
(just thinking ..)


Top
 Profile  
Reply with quote  
 Post subject: Re: “Amateur Gemstone Faceting” book by Tom Herbst
PostPosted: Sun Jan 04, 2015 9:56 am 
Offline
Moderator: Lapidary Arts and Tools
User avatar

Joined: Tue Jun 12, 2007 6:41 pm
Posts: 5475
Location: Massachusetts, USA
I went through this issue decades ago when I built these:
Image

We used SloSyn steppers, and I was beaten over the head with a cudgel about the issue of using a gear driven encoder in a control loop.
This discussion needs to include a description of the difference between an absolute and an incremental encoder.
A hybrid polymer pot has infinite resolution which is limited only by the number of bits in the A/D. It is an absolute encoder. When designing a product with an integral goniometer I insisted on absolute encoding. I had a very limited package envelope and an encoder the size of a hockey puck was a nonstarter.
A absolute encoder's output is always the same at a particular shaft position. It does not count pulses.
A digital absolute encoder has an optical track for each binary decade. That's why they are so big. Final resolution depends on the length of the binary word. They do not require zeroing, and are immune to power failure amnesia. There is no processing, just a simple BCD to Display format decoding.

_________________
www.gearloose.co


Top
 Profile  
Reply with quote  
 Post subject: Re: “Amateur Gemstone Faceting” book by Tom Herbst
PostPosted: Sun Jan 04, 2015 10:59 am 
Offline
Gemology Online Veteran

Joined: Sun Oct 29, 2006 11:43 pm
Posts: 514
Location: North Carolina
EVERYBODY,

Thanks to all for translating something that might have been Sanskrit into something almost English.

Please continue the discussion amongst you who understand what's under the covers!

Just to show how much this is not understood, can this approach show negative angles? If I set it at 90 degrees as
described in the book, and then lift the quill up above that, what happens? It's about a 180 degree swing from straight up to straight down.

And I do wonder what happens if I drop the quill. It usually causes the angle setting to be moved and needs to be reset
on the existing machine. Happens pretty fast too, but maybe not 1/60th of a second?

Jon,

You said "Do not drive an encoder through gears. In addition to backlash, there is cyclic error."

On a UT there is that gear driven shaft that would be the simplest solution, it even has a place to put the encoder or what.
How bad would the error(s) be with the existing gear setup?

Thanks all!

_________________
Bob K.


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 162 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6, 7, 8 ... 11  Next

All times are UTC - 4 hours


Who is online

Users browsing this forum: No registered users and 7 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
Gemology Style ported to phpBB3 by Christian Bullock