[xcad]

I am not sure if this is the right place to post but this is for Max-Mod. The file YTT1.3.2.zip on the www.max-mod-shop.com web-site is broken.
Best regards

[MaX-MoD]

Hello xcad, thanks to report this.
I re uploaded the zip so it should be fixed.

Best regards
Max

[xcad]

Thanks Max-Mod, I tried again but he file still will not open in either winzip or 7zip. File size is 1286kb. I have done 3 downloads, all seem identical none will extract. Anyone else having the same problem?
best regards

[keebler303]

Thanks Max-Mod, I tried again but he file still will not open in either winzip or 7zip. File size is 1286kb. I have done 3 downloads, all seem identical none will extract. Anyone else having the same problem?
best regards

Works fine for me with winRAR.

Matt

[jalessi]

Works fine with 7 Zip

http://www.7-zip.org/download.html

[MaX-MoD]

Hi again,

xcad, maybe you have been victim of a little joke from your favourite browser's cache
I experienced the same problem which was fixed by cleaning the cache of firefox.

Best regards
Max

[xcad]

Many thanks everyone, extracted it now from the version posted.
Best regards

[Iron-Man]

Max-Mod

Here are some pictures of my latest updates to the dsPIC Servo project. I have made several changes to the original Glaister schematic and Jan 2008 software. The filter caps on the encoder input are not necessary if you have shielded cables. Filtering is done in the software. The RC network that is formed sets a low limit on the max encoder speed, and will limit the overall speed of the encoder input. I found out that by removing these, I could dramatically increase the speed of the servo system. The reducers are using Litton-Clifton precision motors with a 3:1 reduction ratio. Torque will be approximately, 49.5 oz-in continuous and 375-oz-in peak. This should be more than adequate for a Sherline Mill/Lathe. These motors have no cogging with my current set up. I will build two prototype systems using the IRAMS10UP60B module coupled with the dsPIC (Version 3 Chip) and the TMS F28335 DSP. The OPA 549 is a great amp, but I want to drive more powerful motors.
The final step for the reducers is prep and powder coat. Candy red.
https://www.pendrypowdercoatings.com...0ee62e36060dc9
Max-Mod, I also noticed and made the change to capture.c {capture every rising edge }.

I will post video soon.

Iron-Man

[MaX-MoD]

Hey there,


this looks to be a clean design, congrats!
But hey! finally you didn't ended with the red-hard H-bridge module? :rainfro:

But if I can make a small remark, avoid the use of vias on power traces. Depending on the manufacturing process the copper at the holes wall can be very thin (0.5oz, sometimes less) while the geometry will keep heat inside the hole. Under high current the via will heat up and eventually blow up.
Some manufacturers specify their vias for 1Ohm, sometimes more (that's enormous!!!).
You can use uncoated vias so the soldermask won't cover them. Then, you can put a drop of solder in it to avoid any potential via blowing problem. Alternatively, you can use multiple vias to reduce the total resistance.

As you said, the encoder inputs have electronic filtering and error-healing feature inside the dsPIC so it's preferable not to put an additional RC filter. Encoders have either differential or open-collector outputs that are less sensitive to electrical parasites.
Oh and have you tried the "step input" test?


PS: Impatient to see the "red candy" running

Good luck,
Max

[beliefdrive]

Max-Mod

Here are some pictures of my latest updates to the dsPIC Servo project. I have made several changes to the original Glaister schematic and Jan 2008 software. The filter caps on the encoder input are not necessary if you have shielded cables. Filtering is done in the software. The RC network that is formed sets a low limit on the max encoder speed, and will limit the overall speed of the encoder input. I found out that by removing these, I could dramatically increase the speed of the servo system. The reducers are using Litton-Clifton precision motors with a 3:1 reduction ratio. Torque will be approximately, 49.5 oz-in continuous and 375-oz-in peak. This should be more than adequate for a Sherline Mill/Lathe. These motors have no cogging with my current set up. I will build two prototype systems using the IRAMS10UP60B module coupled with the dsPIC (Version 3 Chip) and the TMS F28335 DSP. The OPA 549 is a great amp, but I want to drive more powerful motors.
The final step for the reducers is prep and powder coat. Candy red.
https://www.pendrypowdercoatings.com...0ee62e36060dc9
Max-Mod, I also noticed and made the change to capture.c {capture every rising edge }.

I will post video soon.

Iron-Man

hi iron-man.
great look pcb.

i thinks dspic too enough for DC servo , i don't thinks need use tms F28335 DSP

b.r

[Iron-Man]

Thanks Max-Mod for the input on the vias. I did have a quote for 2-oz copper but decided against it. I still have a few changes to make in the board design. No, I have not tried the step input test yet. Also, I decided not to go with rad-hard parts at this time.

Thanks for the input Max.


Iron-Man

[moraru]

Hi Max,

I have one question, is enough to write dspic-servo.hex in dspic?

Thanks

Moraru.

[Iron-Man]

beliefdrive

Thanks for the input.

The TMS 28335 is for a high end hobby/industrial product. I want to design a system that can use multiple motor types.

Iron-Man

[beliefdrive]

yes i think 28335 is high end, it can control 2 brushless servo same time

[MaX-MoD]

Hi Max,

I have one question, is enough to write dspic-servo.hex in dspic?

Thanks

Moraru.

Yes. Thought you may prefer to burn the ingenia bootloader so that you can avoid the use of ICD/picKit and update the firmware thanks to the ProgKey.
You can also look at the my "unified" version of the code, here: http://www.max-mod-shop.com/index.ph...yapsc&Itemid=4


Hey guys, I glanced at the TMS 28335 specs: that's serious!
I guess the biggest of the family with their 18PWM could control 3 servos at a time without any problem
By the way, even AC servo control for a single motor on dsPIC33F is largely possible.
I implemented control on dsPIC33Fj128MC506, which handles FOC operation along with position control (PID) and resolver decoding at a 15KHz rate
Now my biggest problem is to find a 180V DC source with respect to ground...


Best regards
Max

[beliefdrive]

to Max.

I think dspic30f6010A can control ac servo . did you try it?

thanks

[MaX-MoD]

Hi,

didn't tried it but it's quite sure. There is small difference between 33F and 30F. Mainly:
40MIPS vs 30MIPS
3.3V vs 5v
90mA vs 200-300mA (!)
~5$ VS 10$
and much better ADC (200Ksps 12bits works well on 33F contrary to lots of 30F models)
They share the DSP architechture, PWM, QEI etc. so the code is easily portable from one to another.

If you can handle the ADC problems on 30F and if you are happy with a slower control loop, the 30F6010 (and other smaller 30F devices) are suitable for AC motor control

For my part, I now stick to 33F models instead of 30F models.


Best regards
Max

[beliefdrive]

thanks max, your infomation is usefull, i'll try with dspic33

i have a question, what is prolem with adc of dspic30?

thanks again

[MaX-MoD]

Hi,

I advise that you carefully read the errata of the chip you want tu use. Some of them have missing or faulty features.
On the 30F (don't remember which part exactly, think it's 6010) the ADC sample rate was less than excepted and the precision in 12b was poor. Also conversion couldn't be triggered by the PWM.

Each part number (or "subfamily") have their own problems, which are now pretty well documented for the 30 and 33f series.

[beliefdrive]

thanks max, i'll read errata carefful

thanks alote