I'm developing a neww pcb for the OPL3 Module

MIDIbox FM
1

Hi, after a succesful build of a small midibox64 i’m trying with the FM synth.

i’ve downloaded the schematic and the pcb for building the OPl3 Module but i’ve experienced some troubles:

frist, eagle makes the inner hole of the pad too big, the result is an headache when trying to center the drill  ;D

second, some holes, expecially for the jumpers, are very small, drilling and soldering results difficult.

third, i “hate” eagle  ;D

so i’ve made another pcb using some cad a bit more complex than eagle, the attachment is the result.

i’ve take care expecially for ground plane and to make shorter the “hot” traces.

i’m going to etch the pcb, the hope is to get the frist prototype working in the next days.

please, leave your comments  8)

[opl3v2.pdf](< base_url >/applications/core/interface/file/attachment.php?id=247)

[opl3v2.pdf](< base_url >/applications/core/interface/file/attachment.php?id=3388)

2

Great addition to alternatives.

Not in the same case, but I’m going to build the Pic burner before… do you know if the PCB could be improved… ???

3

here is some photos of the pcb:

after etching, before drilling:

after drilling (with a dremel!), note the perfect alignment that such type of small hole in the copper can offer!:

the bottom side with the OPL chips:

the top side, some stuff is missing:

4

yeah, they look gorgeous.

Usually, I just nail the wholes a little bit, before drilling, so the pick can enter the whole without alignment problem.

How is the liquid called in english? (the one that eats copper? I haven’t found a good translation, I just call it iron perclorhide but I don’t think I got it right)

5

How is the liquid called in english? (the one that eats copper? I haven’t found a good translation, I just call it iron perclorhide but I don’t think I got it right)

Ferric chloride

6

Ferric chloride

Yeh that’d be closer to Latin than English :wink:

7

finished!

there is the power supply also, 2 sections, one for analog part e another for digital supply.

i hope to test this in the next days.

8

Great!  :smiley:

What about a pic of the yamaha chips side?

9

Wow, that’s really nice looking. So you just hammered out a PCB for yourself, including a PCB for the PSU? My God, I wonder sometimes where you guys learned all this stuff and got all the equipment, but then again I’m just jealous. Also I was going to say you shouldn’t solder the ICs, but you clearly know what you’re doing!

10

Wow, that’s really nice looking. So you just hammered out a PCB for yourself, including a PCB for the PSU? My God, I wonder sometimes where you guys learned all this stuff and got all the equipment, but then again I’m just jealous. Also I was going to say you shouldn’t solder the ICs, but you clearly know what you’re doing!

Soldering ICs gives a better connection than when socketing them, but if they get too hot during soldering your IC is damaged, and replacing them requires desoldering all pins which is a pain, so for us “simple” people socketing is the way to go  :wink:

11

Mr. Selfservice (a nice matching avatar name btw. ;-))

Could you please add a link to the overworked layout to this wiki page, so that it doesn’t get lost:

http://www.midibox.org/dokuwiki/doku.php?id=layout_improvements

Thanks for your help! :slight_smile:

Best Regards, Thorsten.

12

hi guys, many thanks for your interest in my work… of course i post all the files (schematic, copper side and assembly top) as soon as the prototype were working!

soldering ICs on the pcb reduces the noise, and give a better look to the assembly, and soldering ICs like the TL074 is just a joke, it’s nearly impossible to blast souch chip due to overheat. i had such results with a made in cina soldering station @ 300°C and 0,5mm Sn/Pb/Cu soldering wire.

13

wow, building complete, i’ve got some kind of prototype on my bench.

but the troubles is behind the edge…  ???

all connections doublecheked 16458 times,

lastest version of bootloader and mios burned in the pic

the application is testtone.

the lcd writes “OPL3 Testtone” and nothing else, nothing at the outputs.

the crystal works, the slave clock output of the '262 works, the '262 sends some kind of signal to the DAC

but nothing on the outputs. i’m going to check all signals on all pins, but i think my '262 was fried during desoldering.

any suggestion?

14

Sounds more like a problem with the YAC512, since the YMF sends the DAC signal.

First I would start to doublecheck the schematic vs. new layout, and the new layout against the old layout, just to ensure that this is no general problem with the interconnections. Check especially that the ICs are stuffed the right way (not mirrored), as this confuses most people

Best Regards, Thorsten.

15

yes, it’s a dac problem.

i’ve changed the YMF262 and the result is the same. i’ve noticed a thing may be useful for other debugs: the YMF262 without the core module attached sends random pulses on the dout pins, with the core module and the testtone firmware sends regular pulses. maybe useful for checking good connections between modules.

i’m going to quadruple check the layout and the values, but the voltages on the pins seems good… bah…

16

Yeah! this works!!

Finalli i had some spare time to put my hands on the new pcb and to doublecheck again and again

the result is: “I’m a moroon!!” because to two wire swapping on the core module.

but now everything works, the 1KHz tone at the output looks like this:

it’s normal to see on the sinusoid the small edges from the sample and hold circuits? this doesn’t need to be filtered out?

17

Great to hear, that you finally solved the issue :slight_smile:

it’s normal to see on the sinusoid the small edges from the sample and hold circuits? this doesn’t need to be filtered out?

Yes, thats normal. You could increase the capacity of C27/C28/C29/C30, but this would filter the “wanted” higher frequencies as well.

A better solution could be the usage of a 24 dB (or more) lowpass filter - but this requires some additional hardware (-> expensive solution)

Best Regards, Thorsten.

18

ok, nice to hear that it’s normal.

the datasheet says that the sampling frequency is 47.9 KHz.

it isn’t very clear, it’s the sampling frequency of each DAC??

however it isn’t difficult to filter out this, but maybe a such unclean signal give a stronger electronic sound idea!

how can i upload all the schematics and files to the wiki??

19

I gave you permission to upload files.

Best Regards, Thorsten.

20

Files uploaded and wiki updated. i’ve added my user page also:

http://www.midibox.org/dokuwiki/layout_improvements

http://www.midibox.org/dokuwiki/selfservice