Hobby DRO

Hey Splat!

Those igaging scales were on sale! I just got a few!
Nice!!

I can't wait to hook'em up!

Bernie



Hey, a New Jersey buddy! Bernie, you can get the Igaging units from Grizzly. Check the link I have in my previous post.
 
Arvidj


I have the igaging scales on my mill G0704 mill. Thay are much beter than dials, but fall short of a real dro witch I cannot justify the coast of right now. I have a tablet and Yuriy's program and interface will be a great improvment. And being based on android alot more versital than the Shumatech. And if you have ever used a hole circle function on a DRO system it is way easer and faster than eny outher way. Just my 2 cents. And I am not knocking your statment. Just that for some of us this is going to be a great cheap choice.
 
Arvidj


I have the igaging scales on my mill G0704 mill. Thay are much beter than dials, but fall short of a real dro witch I cannot justify the coast of right now. I have a tablet and Yuriy's program and interface will be a great improvment. And being based on android alot more versital than the Shumatech. And if you have ever used a hole circle function on a DRO system it is way easer and faster than eny outher way. Just my 2 cents. And I am not knocking your statment. Just that for some of us this is going to be a great cheap choice.

Agree on the added functionality of a real DRO. I have or had the Schumatech 350, the 550 with the LCD and a SINO. All are much easier than having to noodle thru the math. And as I indicated, an Arduino, probably more than one, along with Yuriy's stuff to work with incremental encoders sounds like a future project.
 
Yuriy,

Please do not take my comments the wrong way. I am very impressed with what you have done and even asked a couple of questions on your site related to interfacing glass scales with quadrature incremental encoders rather than the absolute interface provided by the iGauging scales.

I was simply pointing out that the goal may have needed clarification and here might be some questions to think about to determine if 'just make chips' or 'chips and more' was the goal.

I hope you are not offended by my comments and I am sincere when I say 'welcome to the group'.

Arvid,
Oh no, I really appreciate your comments. The reason I'm posting this stuff is to get some feedback and/or sanity checks.

AS far as the quadrature encodes go, though ... :) That's my next "project". I want to add a glass scale to the cross slide of my lathe, and a tachometer to the spindle (right now I only have one iGaging scale), and that means hardware that can support mixed scales. A "default" Arduino is out of question for that, though. So far the cheapest option I could come up with is either NetDuino (uses STM32F4 MCU @ $60/piece) or LPCxpresso at half the price, but no FPU and/or DSP). Since working with quadrature encoders is VERY similar to what's required for calipers/non iGaging scales, I will add that at the same time as well.
The only problem is the cost of the glass scales. Right now buying a set will get me killed, so I'll have to wait a bit :)

Thank you
Yuriy
 
Last edited:
Yuiry,

Here is what I was thinking about for the glass scale incremental interface.

http://www.avagotech.com/docs/AV02-0096EN

For those in a hurry there is already a shield available ... http://www.robogaia.com/two-axis-encoder-counter-mega-shield-version-2.html ... but I would prefer to make a shield with two IC's on it to accommodate 4 axis. The IC's are about $11 each and they can share the same TTL crystal oscillator so a home-made shield might come in under $30, and certainly under $50 if it were made it on a ready-made shield breadboard.

I think they could be made to work in one of two ways ...

In both solutions the Arduino polls the HCTL-2032's [I am assuming there would be two IC and therefore four counters for a three or four axis system] to retrieve the four bytes of 32bit counter information from each of the axis.

One solution would then have the Arduino convert the 32bit information into the format that is currently being used to transmit data to the Android tablet, giving the impression that the iGauging scales were being used, albeit with some pretty outrageous value.

The other solution, and IMHO the better one, would be to allow a second format where the coordinate information would be a hex representation of the 32bit counters. The conversion from the hex values to what ever the Android is using internally ... Big Decimal, Floating Point, what every ... is then done by the is then all done inside the Android tablet. The values would still be outrageous but I do not think the Android cares as, if I am not mistaken, you said it is simply thinking in incremental terms anyway.

Obviously I have spent all of 30 seconds on the analysis, but at least that is where I would start.

Arvid
 
Yuiry,

Here is what I was thinking about for the glass scale incremental interface.

http://www.avagotech.com/docs/AV02-0096EN

For those in a hurry there is already a shield available ... http://www.robogaia.com/two-axis-encoder-counter-mega-shield-version-2.html ... but I would prefer to make a shield with two IC's on it to accommodate 4 axis. The IC's are about $11 each and they can share the same TTL crystal oscillator so a home-made shield might come in under $30, and certainly under $50 if it were made it on a ready-made shield breadboard.

I think they could be made to work in one of two ways ...

In both solutions the Arduino polls the HCTL-2032's [I am assuming there would be two IC and therefore four counters for a three or four axis system] to retrieve the four bytes of 32bit counter information from each of the axis.

One solution would then have the Arduino convert the 32bit information into the format that is currently being used to transmit data to the Android tablet, giving the impression that the iGauging scales were being used, albeit with some pretty outrageous value.

The other solution, and IMHO the better one, would be to allow a second format where the coordinate information would be a hex representation of the 32bit counters. The conversion from the hex values to what ever the Android is using internally ... Big Decimal, Floating Point, what every ... is then done by the is then all done inside the Android tablet. The values would still be outrageous but I do not think the Android cares as, if I am not mistaken, you said it is simply thinking in incremental terms anyway.

Obviously I have spent all of 30 seconds on the analysis, but at least that is where I would start.

Arvid

Arvid,
This is an interesting solution. I never thought about using a dedicated IC. One major plus I can see from the get go is that the Arduino can still work in the "stupid" mode (don't need 8 separate pin interrupts. timers etc.) On the other hand there is more assembly required. which appears to be a problem with many users (I get a good number of emails complaining how complicated the current hardware is).
I'll try to get a few of those chips and experiment a bit (I have some rotary encoders that can be used for testing, I think).
I started playing with LPC's Mbed board ($60 and can be programmed using an online IDE), but that's not going to be done soon...

The current data format will work fine for this application. It can handle numbers up to 32 bit wide, and the application has a way to set CPI settings for each axis. I will need to add one more axis, though.

Thank you
Yuriy
 
I do not know what type of machine you have or how heavy of a user you are but I believe you will quickly become unhappy with this system, let me tell you why. Lets start by assuming you have a 7 x 12 late or mini mill and would like to improve the overall accuracy of your work. You can buy plain scales (IGaging is a good one, or Grizzly) and mount those on your machine. These type scales come with a remote read out so if you would like to have X and Y axis on your lathe you can mount the remote readouts in a convenient location be good to go for less than $100 for both axis.

Lets presume you have a larger piece of equipment and thus are more of medium to heavy duty user. You are probably going to want your display head close at hand because you will be using features such "relative position", "absolute position", "radius", "diameter", "bolt hole define/use", etc. While this company lists those functions it would appear from the ad that you must access them via the touch screen unit which in turn is connected to a PC. I have no knowledge of the quality of the touch screen unit but I do know that for $210 plus the cost of a PC you can buy a real DRO head. In this day and age of imbedded processors using a PC to provide the computational power does not make much sense. For two to three hundred you could buy a Shumatech head and be much happier.

If you are budget concience and a light duty user, check out the IGaging or other linear scales as a solution. I used that arrangement for several years and was very happy with it, especially when compared to try and keep track of how many turn I made on a feed knob. DRO heads such as Shumatech will read all different types of scales including the ones that I have mentioned.

That's my two cents.

Jim

According to the website for Schumatech, they are currently out of stock for the 550 board. I joined the Shumatech Yahoo group to find out more. The way they handle the manufacturing is to have people sign up on a list, then when the list gets long enough to justify a production run, they put in an order at the factory that actually makes them. The most current post I could find on the board was several weeks old and stated they were about 30 signatures short of a production run. There was no indication how long that might be. I went ahead and put my name on the list because this sounds like the right solution for me. I don't have to have this product badly enough to pay lots more. But I still hope the wait isn't too long!

Walt
 
Hi Walt, you have discovered one of the issues with Shumatec that some people find objectionable, it is not a "store" with inventory. Scott Shumate who is the guiding hand behind the Shumatec product has done a marvelous job at designing and producing these units all at no benefit to himself. I originally purchased one of their earlier units (350 type) and I liked it so much that I bought two of the 550's when they were available, I believe I waited almost a year. Much of the delay was because it was a new product and there were some issues with parts sourcing and vendor schedules, etc.

I think if you just put your name on the list and go off and do something else someday you will get an email of the schedule and your off and running. The product (550) is very solid and full of functionality. My units take a beating and have performed flawlessly for over two years. I am sure you will be very pleased with the unit and if not there is a waiting group of people who will be happy to take it off your hands.

Jim
 
I was just out looking at his yahoo grouip. He hasn't had any new 550 parts made since at least Nov 2011 when he said that he didn't have the time to work on this side business nor the money to float the inventory required to order a minimum production run of boards. Since that time, he's been compiling a list of people wanting boards and needs 60 to justify placing an order. He has 30 people on the list after almost a year and a half. I'm not even going to bother considering this option in the future.
 
Yea, looked at it today and it would sure be a good option for me, especially price wise, but from what has been said here I think it may just be easier for me to make one using a PIC.
 
Back
Top