MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

uGrids and MI Grids DIY builds
MUFF WIGGLER Forum Index -> Music Tech DIY  
Author uGrids and MI Grids DIY builds
henrry
Just finished the build and programming of three uGrids but encounter some issues.


The programming was done quite easily with a Mac running High Sierra and an AVR Pocket Programmer on the hardware side.

On the software side, i used this useful tutorial to generate the grids_bootloader.hex and grids.hex files using Terminal: Music Thing Modular Tutorial

Then I used the AVRFuses software to flash first the grids_bootloader.hex and second the grids.hex files. The fuses seem to be set automatically on the fuses page.
On the program preferences the choosed Programmer should be usbtiny. Link to the software: AVRFuses

I hope I didn't forget to mention any usefull additional information for those who are new to those builds.

One unit has a clock issue and Tap button not responding. I'll check any slodering bridge for this.

Two other units share a same strange behavior on Ch1 and Ch2 (on Ch3 for a third unit): the rythmic patterns look OK on the LEDs but the trigs/gates OUTs don't always fire accordingly. The result sounds strangly regular making longer patterns but adds a bit of randomness. hmmm.....

On one unit Ch1 and Ch3 don't output anything with the pot set before 12 o'clock... The rythm doesn't fade in, it rather enters at the speed set by the knob. seriously, i just don't get it

In general I would consider soledering bridges in between ICs pins for most of these cases. But my concern is the following:

As I'm not sure I did set the Fuses the same way on each load, could some or other strange behavior described above come from this? I don't know what the fuses settings do.
fg
hello, i builded grids not ?ugrids? and for grids, fuses flashing are important

i did it with a nano board 16mhz but it's working very well


avrdude -c usbtiny -p m328p -U flash:w:grids.hex -U lfuse:w:0xff:m -U hfuse:w:0xd8:m -U efuse:w:0xff:m -U lock:w:0x2F:m

voilà
henrry
Thank you we're not worthy
I'll try to check with these values.
henrry
I checked the soledring: no unwanted solder bridge.

Tested the programming with the new fuses values and for sure some of the described bugs change totally:
Now all modules seem to work the same, all channels generate triggers with Pots set before 12 o'clock and the Tap button works normally.

But unfortunately I still have these missing somhow random triggs on the outputs that do not correspond with the LEDs blinking...

this is the message I get for the fuses: avrdude: safemode: Fuses OK (E:FF, Hgrin8, L:FF)
But acutally I can't enter this code on AVRFuses:
Quote:
-U lock:w:0x2F:m
I can just enter Low, High and Extended values for the fuses.

I'm not really wanting to enter random fuses values...
Did I mention: AVRFuses automatically recongnizes an ATmega328P. It outputs this message:
Quote:
avrdude: Device signature = 0x1e950f (probably m328p)


What to do?
help
henrry
I don't know much about coding in Terminal, but I just opened a Terminal window from directory containing grids.hex file and ran the comand line you gave me.
It loaded perfectly. I guess it did also write the lock value:
Quote:
lock:w:0x2F:m
according to feedback message. But still do I get this bugg... Dead Banana
fg
i look in the grids makefile and the fuses recommandation is

LFUSE = ff
HFUSE = d8
EFUSE = 05
LOCK = 2f

avrdude -c usbtiny -p m328p -U flash:w:grids.hex -U lfuse:w:0xff:m -U hfuse:w:0xd8:m -U efuse:w:0x05:m -U lock:w:0x2F:m

with my nano the efuse makes an error but may be ?
fg
http://eleccelerator.com/fusecalc/fusecalc.php?chip=atmega328p

that could help to understand
juicepatrol
I just finished a build and all the firmware seems to work fine. The tap tempo, leds, knobs, inputs work fine but i am getting nothing out of the outputs. I see in the schematic that the triggers are connected to 5v. Anyone have any idea where in the circuit i should troubleshoot. Or has anyone else had a similar problem?
fg
check this

juicepatrol
Thanks a lot. Will check and post back with results applause
henrry
fg wrote:
i look in the grids makefile and the fuses recommandation is

LFUSE = ff
HFUSE = d8
EFUSE = 05
LOCK = 2f

avrdude -c usbtiny -p m328p -U flash:w:grids.hex -U lfuse:w:0xff:m -U hfuse:w:0xd8:m -U efuse:w:0x05:m -U lock:w:0x2F:m

with my nano the efuse makes an error but may be ?


I encountered an error using EFUSE = 05 with AVRFuses.
In this post from MI forum Olivier suggests that:
Quote:
0x05 or 0xfd will give the same results

Since I don't know whether the error comes from the software or the hardware side, I'll try both with your provided code lines and will give feedback soon.

thanx thumbs up
henrry
Used this procedure with the three units:

First I used Erase Device with AVRFuses for safety then,
Quote:
Copy and paste both grids_bootloader.hex and grids.hex files in the same directory

Then open a Terminal window from this directory using right click Service>New Terminal at Folder.
Type and enter the following lines:

avrdude -V -p atmega328p -c usbtiny -P usb -B 10 -e -u -U efuse:w:0xfd:m -U hfuse:w:0xd8:m -U lfuse:w:0xff:m -U lock:w:0x2f:m

avrdude -V -p atmega328p -c usbtiny -P usb -B 10 -U flash:w:grids_bootloader.hex

avrdude -V -p atmega328p -c usbtiny -P usb -B 10 -U flash:w:grids.hex

Everything loaded right according to the feedback message but still, my three units keep on missing random trigs...
I'll check my BOM to see if I didn't source a part uncorrectly.
henrry
Very strange behavior indeed: after feeding a clock, everything seemed to work normaly. At first, euclidian mode also missed random trigs, then I switched off and on the system and euclidian en drumming duties mode where inverted (according to the LEDs feedback and also for the strange behaviour). Missing notes on drumming and euclidian normal with no random missing notes. Except that CH1 length is controlled by Y knob (should be X), X chontrolling CH2 and chaos for CH3. Turning clock pot while in the settings mode generates strange behaviors woah
I give up for a while ... seriously, i just don't get it
npseaver
Hope this isn't too much of a threadjack, but I'm building a µGrids now, and can't figure out the orientation of the LEDs from the PCB markings. This may just be me being an idiot, but how did you figure it out? (You can see them here, just a set of concentric circles...)
fg
hello, just search the ground (-) with a multimeter

npseaver
Thanks! I'm in over my head with this build, but that was just enough specificity to bring me up a notch.
henrry
Just got an Atmel AVRISPMKII programmer and Atmel Studio 7 on a PC and tested the programming using the method described here (link from magpie modular): https://drive.google.com/file/d/1ij3SMkusbL2pcQ3qNHOqQblQwNoDTMC_/view

After a short test on a single uGrid it looks like the strange missing gates keep on ocuring at low BPM...

The only hardware problem I can possibly imagine beeing the cause of this is that I may have used LEDs with wrong values. As I didn't have left any of the ones specified on the BOM I just picted up the ones I had without checking their values... could this be the cause of these missing gates? sad banana

Or did I brick the Atmega using wrong fuses values in my previous programming attempts? hmmm.....

help
cnicht
The LEDs should cause a problem.

Have you tried reading the status of the fuses in the microcontroller via the programmer?
Ron_H
npseaver wrote:
Hope this isn't too much of a threadjack, but I'm building a µGrids now, and can't figure out the orientation of the LEDs from the PCB markings. This may just be me being an idiot, but how did you figure it out? (You can see them here, just a set of concentric circles...)

LED orientation is clear from the open source documentation.
https://github.com/jakplugg/uGrids/blob/master/control_layout_diptrace _export.pdf
henrry
Hello everybody.
This will hopefully be my last report regarding the uGrids.
In advance I appologies If all of this may have resulted in confusion and loss of time by reading my stuff...

batchas pointed this in another thread:

Quote:
Just some ideas which came to mind while reading what your're describing.
Sounds a bit like the ACC outs and not the main TRIG/GATE outs. You would then see LED blinking but only accents if you know what I mean.
As for the signal only outputing after 12 o'clock: if you are in euclidean mode this is normal depending how the CH3 pot is set (the pot on the left. the one for random in DRUM mode).
Easy way to be sure is set in DRUM mode and also check the other OUTs (in case TRIG and ACC are inverted).



Part of my confusion was due to my ignorance regarding this lack of Trigs before 12 o'clock when in Euclidian mode...

And then I did further testing comparing two units: one that was flashed using the AVR pocket programmer and the other one flashed using the Atmel AVRISPMKII programmer which I did yesterday. Both units seem now to work the same and fine.

1/ regarding the missing trigs/gates not coresponding to LEDs flashing, the problem may be due to my bad testing. Maybe did I once test the ACC outputs instead of the real outputs as suggested.
Moreover the fast retrigs of uGrids are sometimes too fast for some digital modules. The clock input fed with high speed pulse with variable pulswidth may have increased this problem with some digital drum modules used during my previous testings.
So in order to test the outputs properly I used simple passive low pass gates pinged by the pulse outputs of the uGrids and fed with a simple oscillator to hear the signal
Everything is now aparently working fine....meh

2/ Some trouble testing and confusion came from the Advanced Settings mapping when you tweak the X and Y pots. Their mapping and position does not correspond to those of the original Grids (at least in my case as i did flash the uGrids Amega with a standard Grids firmware):

- When you tweak the X pot you actually change the Trig/Gate output setting.
- When you tweak the Y pot you actually switch between Drum duties and Euclidian mode. And when in Euclidian mode Y is linked to Ch 1, X is linked to Ch2 and Chaos to Ch3!!!!

Can anyone tell if this is also the case on their respective DIY unit if you just loaded the standard Grids firmware to a uGrid?

3/ If what I previously said is correct (and not a random good functionning of my deffective units, combined with poor testing skills oops ), the good news is that you can successfully programm the uGrids using an AVR Pocket programmer and a Macbook pro following the procedure described above.

Thanx for all of your contributions ant patience helping and advising DIY builders!!
we're not worthy
terumi
Hey people, today I intent to program my diy grids.
I dont have an avr programmer but I own an arduino uno.
On the software side I have the mutable enviroment installed.
Do I have to have grids powered when programming it?
I saw that one of the programming pins of grids reads 5V. This comes to addition of the power connected or can I ommit this when programming?
Oes the mutable envirome t has avrdude installed?
Do arduino need a capacitor between gnd and reset?

Thank you.
looprication
Hey, we're Grids buddies! I just got done today with the capacitors and resistors, planning to get it completed by the end of the weekend.

terumi wrote:
Hey people, today I intent to program my diy grids.

Do I have to have grids powered when programming it?



Yes. I used a Sparkfun Pocket AVR Programmer to do Branches and what I found was I needed to power the Sparkfun with my Mac's USB first, then power on the module and upload the code.

terumi wrote:

Does the mutable enviroment has avrdude installed?


I *think* so? Not 100% sure but by following the process in the blog article I wrote recently I was able to flash Branches, no problem.

Cheers,

-loops
MUFF WIGGLER Forum Index -> Music Tech DIY  
Page 1 of 1
Powered by phpBB © phpBB Group