Tronic's AMC motor motion-controller with pwm/servo output

Re: Tronic's PWM/servo control Interface with encoder/pot input

hmm, my interface looks different.

Looks like your layout is mirrored? :bug:

Regards
Christian

Re: Tronic's PWM/servo control Interface with encoder/pot input

Oh noooo!!!!
Ferrari_man's boards are really etched mirrored!!!!!!!! I noticed the mouse next to them...!!!!

:shipwrecked:


Regards, Thanos

Re: Tronic's PWM/servo control Interface with encoder/pot input

Yeah I know... first time... .... but does it matter?? if I just turn the parts .... ??

Ferrari_man

Re: Tronic's PWM/servo control Interface with encoder/pot input

No. In best case doing so, you will burn everything up! Even the voltage regulator is opposite placed!!!

What you can do... Just unsolder carefully all components, and solder them again in a new correctly etched PCB! Its relatively easy to use a suck pump to remove the solder when it's melted and remove the components without destroying them.

Regards, Thanos

Re: Tronic's PWM/servo control Interface with encoder/pot input

Hi everybody,

Erich has sended me more photos from his boards that now are complete and ready to test:









As you can notice he used metal distance holders to attach the cooling aluminum stripes on the mosfets to avoid have the plastic one's melt under great heating if anything goes wrong. This will ensure that there is always firm contact between the cooling stripes and the mosfets!!!!

Its a master-piece! Good Job Erich! (I know that he keeps visiting the forum and reads this 8) )


Regards, Thanos

Re: Tronic's PWM/servo control Interface with encoder/pot input

They are not solded yet... so it is no problem... I just have to make 2 new boards right?

Ferrari_man

Re: Tronic's PWM/servo control Interface with encoder/pot input

Hi again,

As from today I decided to cease the use of 24Mhz crystal on the AVR motion controller v1.5 because its the source of many problems. Some of them include:

1. Sometimes the AVR doesn't boot correctly (crystal denies to start oscillating and locks)
2. There is an small deviation error to the serial communications (small, 0,02%, but its there)
3. Sometimes programming with Ponyprog fail with no obvious reason (the crystal again is here guilty for that)

The bad part is that I even faced myself some bad 24Mhz crystals that caused the above, in particularly 1 of the 4 crystals that I bought! And as I had complains from other that is also happening to them, I decided to decrease the top value crystal that can be used safely on my AVR controller to a 22.1184Mhz crystal.

So you should get a 22.1184Mhz crystal from so on for the AVR motion controller v1.5.

Don't throw away the 24Mhz crystal! You can still try it to see if works fine for you. And I'll probably use it on the next new Ultimate AVR motion controller 6DOF that has ATmega644 chip. With nominal clock support of 20Mhz the ATmega644 can handle the 24Mhz crystal pretty well.

And to keep you from whining about what firmware you should use now with the 22.1184Mhz crystal I made new firmware for both coupled and uncoupled 2DOF motion. Also as a bonus these two firmwares have the ability to display the incoming USO data and the potentiometer readings on the LCD on normal operation without delaying at all the motion control functions! Great for debugging issues during calibration and first tries!

You can switch the LCD display data, ON and OFF by using the DIP switch1. You can even do this when the board is powered as the firmware will detect the change of the switch and respond accordingly.

PWMBETA002LCD_OP.zip

PWMBETA009FR3XLCD_OP.zip

Good luck!

Regards, Thanos

Re: Tronic's PWM/servo control Interface with encoder/pot input

Hi guys,

Is there an online shop where i can order all the parts that i need for the avr-controller and h-bridge?

Best for me if there is one in Holland...

Re: Tronic's PWM/servo control Interface with encoder/pot input

Hi Thanos, :hi:

1 question and 1 favour request.

I am about to order the PCB etch locally. I see there are many versions of the boards both on this thread and your latest zipped file.

I am looking at a folder named single_PDF_PCBs layers, and documents named:

avr_motion_sim_PCB_single_1_5.pdf
Dual_simple_mosfet_hbridge_1_3b.pdf

Are these the most up-to-date documents that I should etch?

If so would you be able to PDF them both on the same page for me, it would save me $25 as the company would have to charge me for 2 lots of filmwork if I supply them 2 seperate PDF documents.

If you do I'd greatly appreciate it!

I'm hoping to be able to propagate some of the boards by this weekend if they can get the document by tomorrow :

Evan

Re: Tronic's PWM/servo control Interface with encoder/pot input

Hi Evan,

I wasn't sure about the exact version you wanted, so I prepared both terminal connectors/no terminal connectors layouts. The versions 1.2b and 1.3b are the terminal connector versions!

The only difference between DSMhb1.2 and DSMhb1.3 is that one uses molex connector and the other IDC connector.

AVRmc_single_DSMhb_NO_TC_version.zip

AVRmc_single_DSMhb_TC_version.zip


Regards, Thanos

Re: Tronic's PWM/servo control Interface with encoder/pot input

Hi Thanos,
Macho Gracious for the files and being so quick! :

To tell you the truth I had no idea either what I required as well! And even after printing all 4 documents you just provided I couldn't spot the difference between the 'No TC' and 'TC' other than the HBridge molex/idc layouts between versions v1.2 and v1.3b.
Sooo I went ahead and sent AVRmc1_5_single_DSMhb1_2_bottom.pdf to them. Hopefully it should be sweet.

Please correct me if I am wrong before tomorrow morning (my time) so I can stop them etching it if I have to!! :lol:

Re: Tronic's PWM/servo control Interface with encoder/pot input

Hi Evan,

Yes Its a little difficult to see the differences if not very expierenced in it. The AVRmc1_5_single_DSMhb1_2_bottom.pdf you chossen has no Ternimal connectors for the mosfets, so you have to solder them directly on the PCB! In this case DONT forget to first solder all the resistors and solder last the mosfets to avoid burning the mosfet gates.

If you want to use Terminal connectors for mounting the mosfets without soldering them, then send them the AVRmc1_5_single_DSMhb1_2b_bottom.pdf (with the b marking)

Regards, Thanos :thbup:

Re: Tronic's PWM/servo control Interface with encoder/pot input

Aaah I see the difference now, the mosfet holes are spaced further apart to accommodate the terminals. Well I bought the terminals for it so I may as well use 'em!! whew thanks for that! :clap:

Re: Tronic's PWM/servo control Interface with encoder/pot input

Hi Thanos,

The only crystal I have managed to get so far is a 20.0MHz Crystal, can I use this or should I really try to source a 22.1184Mhz crystal?

I should get the etched board later today, it will be undrilled but the etching company said they will include a handful of old drill bits which are still good enough for DIY work for free!

Evan

Re: Tronic's PWM/servo control Interface with encoder/pot input

Hi Evan,

No, you need the 22.1184Mhz crystal or else the serial connection will be impossible! The timing of the serial communication depends on the crystal to synchronize to the 115Kbps!

Regards, Thanos

Re: Tronic's PWM/servo control Interface with encoder/pot input

Okay will do. I'll mail order it.
I got the board already and also half a dozen drill bits that look like large dentist drill bits in the parcel as well!

Hi to all,

We have two more verified AVR motion controller v1.5 users in the club! Erich and Rovor have their interfaces in fully working condition from today morning despite the difficulties they faced! :clap:

I'll post more details about the steps it took, to help them fix their interfaces, so this information can help a dozen other A.M.C users and relief me from having to reply, all the time, to emails asking for help

I'll give some new firmware, as well, with lower but more reliable 18.432Mhz crystal.

Regards, Thanos

Hi to all,

With the help of Erich, I present you a way to fix the Device missing or unknown device -24 or the Write failed messages when using Ponyprog2000 to program my firmwares on your AVR boards:

I think the most of the USO connection problems occured by using 24Mhz crystal. Cristian (Egoexpress) had exactly the same problem. Everything worked but no serial connection with his 24Mhz crystal! I remember that I fixed that by using the slower 22.1184MMhz crystal. The problem does not always appear. Just some atmega8535 chip batches seem too unstable above 18.432Mhz frequencies.

But I got some emails from Erich too that he had problems getting his AVR start with a 22.1184MMhz crystal some times and he overpassed is by putting a 18.432Mhz crystal and swaping it later back to the 22.1184MMhz one. He also mentioned that he have to do some settings on his ponyprog to be able to program correctly the with slower crystals.

From ERICH:

>Pleased to report some positive results, and
>one peculiar problem. Start with the problem--the
>AMC is working now. FWIW, I would like to give
>you a quick rundown of all I had to do as it
>might be of assistance to others.

>The problem is that 22.1184 MHz seems to be JUST the
>outer limit where the circuit is JUST unstable.
>This happens when I start up first thing in the
>morning for 3 days now. The program does not run.
>However when I remove change the crystal to 18.432. it
>starts up immediately! Then I power down again and
>switch to the 22.1184 and it works beautifully---all
>day. Left overnight, there is the same problem and
>solution, next morning.

> As usual, you were the trigger to the solution. I
>kept spoiling 8535s--your clue, run with 4 MHz
>crystal---did that but no direct result. Read Lancos'
>PP Doc. Some sketchy explanations from Lancos--so I
>went to
>the website and d/ld updated Doc (attached here
>for your quick reference). Made 4 changes in the
>.ini file, two suggested in the Document pdf, and
>2 others in the same Doc., in the FAQ section:--

>These from the main Doc.:-

>SPIBusSpeed=VERYSLOW---this was set to NORMAL.
>There are 6 options from ULTRASLOW to TURBO. I
>set to VERYSLOW

>ATMegaPageWriteDelay=70---originally 30 (I think)

>And these two from the FAQ section:--

>AVRByteWriteDelay=30------originally 20
>AVREraseDelay=100---------originally 50

>With these settings, a new 8535 and a 4 MHz
>crystal, I was able to set the fuse bits and
>write the program and run it.

>Changed crystal progressively, from 4 to 14.xxx,
>18.432, 22.1184 and 24.0. The 24 did not work so
>stepped back to 22.1184. Worked well!

>Next morning--problem to run 22.1184, unless I
>first changed back to 18.432. It seems to work when
>its given one shock into oscillation from cold.

>Obviously, I cannot use the lower freq. crystal
>with the current SW, as you advised EvanF in the
>forum.

>By the same procedure above, I was able to
>salvage and program 2 other 8535s earlier stuck.
>They too seem to become unstable after 18.432.

So I suggest all you to buy a 18.432Mhz (and a small 4Mhz crystal for the reasons mentioned above) and program in it, the attached PWMBETA002LCD_OP18.HEX firmware.

Here is the INI erich mentioned:

PONYPROG2000_FIX.zip


And some new Firmwares to use with different crystals:
14.318181 Mhz crystal:
PWMBETA002LCD_OP14.zip

18.432 Mhz crystal:
PWMBETA002LCD_OP18.zip

22.1184Mhz crystal:
PWMBETA002LCD_OP22.zip

24 Mhz crystal:
PWMBETA002LCD_OP24.zip


Another fix you can try is to change a bit in the fuses, by putting a check mark on the CKSEL0, like in the photo below:

This will alter the rising time of the crystal oscilator, and fix the sensitivity of RX port so it does not reset the chip in some cases.


You really must be patience with it some times! I had faced some of these problems in the past but I was able to overcome them easily. Remember why I changed the Max232 section from the simple one to the more complicate one cause I could not receive serial data too? Sometimes a solution that works for me its of no use to others since different materials are used in first place!


Regards, Thanos

Hi,

I decided to make some video guides to help people understand the basics about my AVR motion controller.

I have made four video parts so far, that shows the cable connections, the ponyprog programming procedure, and the connection with x-sim software.

Here is the link to download all these videos:


http://rapidshare.com/files/106829488/A ... 1_to_4.zip

I hope you enjoy them! :thbup:



I will make more later that will concern the connections on the DSMhb1.2 and simple motor tests and position feedback examples.

Regards, Thanos

thx tronic
it is helpful

sorry for my English

reagards christian