midibox cv - gates and cvs problem

[spippoli]

Hi,

I'm building a midi-cv using seppoman aout_ng module.

Here's where I'm up to:

-core has 18f4620 pic

-I'm using a datawheel

so I changed

PROCESSOR = 18f4620

in the Makefile.orig file and

#define DEFAULT_ENC_DATAWHEEL   0
...
#define ENABLE_J5 1
...
#define AOUT_INTERFACE_TYPE 3

inside setup_default.asm file. Recompiled with gpasm and uploaded the firmware. The application seems to work fine since LCD and DIN button/wheel behave as expected (menu navigation/settings work ok).

Sync tests are ok, j6 clock and enable pins works correctly.

To test gates I disconnected the aout_ng module and played some midi notes.

Here I get animations on the lcd but on the gate out (j5:a0) I only get short pulses (~uS).

Also with aout module connected playing midi notes I get no voltage across cv out.

Where could the problem be?

[stryd_one]

I should just mention that the hardware was also tested with the pre-compiled hex file in the distribution zip, and the J5 gates did not work. I'm of the opinion that this indicates a problem with the core, and my advice was to remove the aout module and just work on those gates first....

[TK.]

There was an incompatibility to PIC18F4685, caused by direct accesses to the GATES variable, which is located in a range which is reserved for CAN on this special derivative.

I fixed this in cv_map.inc

Best Regards, Thorsten.

[stryd_one]

You sure that's a 4620, spippoli?

[spippoli]

@stryd: yes it's a 4620.

@TK: thanks for the reply, I just tried uploading both mios and the recompiled midibox_cv with the modified cv_map.inc, but I get the same result.

[bugfight]

i vaguely remember having trouble with 4620 and cv, but can't remember the details...

[TK.]

Sorry - wrong movie ;)

MBCV uses it's own initialisation routine for J5, I still haven't changed it to j5_io.inc

Means: for proper initialisation, it's required to change

        ;; disable the ADC which allocates the analog pins
        movlw   0x07
        movwf   ADCON1
[/code]


to
[code]
        movlw   0x0f            ; all newer PIC18F derivatives (like PIC18F4685)
        movwf   ADCON1

in j5_dout.inc

(this change won't be required anymore once I'm using j5_io module in MBCV)

Best Regards, Thorsten.

[spippoli]

yup, that was it!

Now gates are working nicely, anyway gate's output are still always 0V.

It seems that there's no signal on the serial line to the aout_ng module.

[spippoli]

A short update after some tests:

application in compiled with this option

; define the AOUT interface which is used here:
;   1: one MBHP_AOUT module
;   2: up to 4 (chained) MBHP_AOUT_LC modules
;   3: one MBHP_AOUT_NG module
; all other values invalid!
#define AOUT_INTERFACE_TYPE 3

and aout_ng module is connected to core through j6 socket as stated here for MBSID: http://www.ucapps.de/midibox_sid/mbsid_aout_interconnections.pdf though j6:SI and j7:SO looks flat on my scope, while I see some data flowing through j10 pins as I tweak calibrations options.

I guess I have to use j10 to connect core to aout_ng module but couldn't find any info on it and I didn't want to fry my DAC. There's a link on the midibox_cv page but it's broken http://www.ucapps.de/midibox_cv/mbcv_aout_ng_interconnections.pdf

[stryd_one]

In the code the pins are assigned as such:

; Pin Configuration
; Can be overloaded from external
#ifndef AOUT_LAT_CS
#define AOUT_LAT_CS	LATC	; The chip select pin CS#
#endif
#ifndef AOUT_TRIS_CS
#define AOUT_TRIS_CS	TRISC	; is connected to Port C.5
#endif
#ifndef AOUT_PIN_CS
#define AOUT_PIN_CS	5	; (CANNOT be shared with other outputs!)
#endif
;
#ifndef AOUT_LAT_DIN
#define AOUT_LAT_DIN	LATC	; The data input pin DIN
#endif
#ifndef AOUT_TRIS_DIN
#define AOUT_TRIS_DIN	TRISC	; is connected to Port C.4
#endif
#ifndef AOUT_PIN_DIN
#define AOUT_PIN_DIN	4	; (can be shared with other outputs)
#endif
;
#ifndef AOUT_LAT_SCLK
#define AOUT_LAT_SCLK	LATD	; The shift clock input pin SCLK
#endif
#ifndef AOUT_TRIS_SCLK
#define AOUT_TRIS_SCLK	TRISD	; is connected to Port D.5
#endif
#ifndef AOUT_PIN_SCLK
#define AOUT_PIN_SCLK	5	; (can be shared with other outputs)
#endif
;

They do not appear to be overridden by the app. HTH!

Edit: See also the pin list on the wiki

[bugfight]

or just use this one:

http://www.ucapps.de/midibox_cv/mbcv_aout_interconnections.pdf

[spippoli]

yep, thanks to patient stryd_one I sorted that out 

now at least I get voltages at the cv outs, but I'm still fighting with the calibration procedure...

but I'll give me some more time before asking here  :D

[spippoli]

I still didn't get the aout_ng module to work correctly.

Yesterday I made some tests with smashtv and here's what I found:

first of all all gates behave the same; in unipolar mode when calibrating the outputs I get 0V for minimum and more or less 11V for middle and max values. Setting calibration to 1V no matter how I tweak the trimmers the closest value I get is 2V. When setting the jumpers to bipolar mode the output stays steadily at 11V no matter the calibration setting or the trimmers tweaking.

I tried supplying the core and the aout_ng module with two separate wallwarts (DC for the core and AC for the aout) or in parallel from a single AC wall wart to check for ground potential issues but it didn't solve.

I checked pin 16 on the DAC, which is the reference voltage and I got 2.03V. The datasheet says that typical value should be 2.043V so it should be ok.

While doing this measurments I noticed that c10 was not soldered properly since tin didn't get all the way through the holes and reflowed it.

After I did this I noticed something weird: using the same supply for both modules wouldn't turn the box on, while with different wall warts the unit turns on and behaves as before (steady 11V at the output in bipolar mode and out of range tuning in unipolar mode).

Any idea on what to check now?

[johnny5]

After I did this I noticed something weird: using the same supply for both modules wouldn't turn the box on, while with different wall warts the unit turns on and behaves as before (steady 11V at the output in bipolar mode and out of range tuning in unipolar mode).

Any idea on what to check now?

so im not as far as you but i did have j10 of the core hooked up to my ng and it didnt light up(now im thinking i miss wired somthing). so i also hooked up j6 via mbsid standards thinking maybe j10 was an obsolete connection? my ng light up! i was using a 9v dc wallwart. when i went and hooked up +15/-15v (from my linear psu)to the ng i get +17.7v at pin4 and -12.7v at pin11, but thats another story. im thinking maybe i needed to reference a ground from the ng board now. i was using a grnd from the core. not sure if that would make  a difference. i am currently away from home for a bit i kinda wish i knew what pic i used? ???

[stryd_one]

I wouldn't recommend trying non-standard connections unless you really know exactly what you're doing. There is a risk of frying components. Follow the schematic bugfight linked to and the pinouts I posted, and it will be correct.

After I did this I noticed something weird: using the same supply for both modules wouldn't turn the box on, while with different wall warts the unit turns on and behaves as before (steady 11V at the output in bipolar mode and out of range tuning in unipolar mode).

Huh? It's not supposed to work like that. Digital power (5V unipolar) goes into the core. The connection between core:j10 and aut_ng:j1 gets this digital power to the aout_ng board. Analog power is a different issue. Separate wallwarts is not recommended, ground shifts can cause damaged components.

Perhaps the power supply is where you guys are failing. What PSU did you plan to use?

[johnny5]

I wouldn't recommend trying non-standard connections unless you really know exactly what you're doing. There is a risk of frying components. Follow the schematic bugfight linked to and the pinouts I posted, and it will be correct.

Perhaps the power supply is where you guys are failing. What PSU did you plan to use?

sorry i wasnt trying to recomend the use of j6. I was just mentioning the fact that i made a mistake interconnecting j10, and the image i was going by afterwards was indeed a mbsid. witch is completely wrong.  :-[

as for the powersupply i am using a linear/analog psu. i have read here people using +15v/-15v for mbcv and thats what im using on my modular. so i shouldnt have problems there. anyway im sorry i feel like a thread jacker now! i have one going already acouple down from this one.

[spippoli]

Huh? It's not supposed to work like that. Digital power (5V unipolar) goes into the core. The connection between core:j10 and aut_ng:j1 gets this digital power to the aout_ng board. Analog power is a different issue. Separate wallwarts is not recommended, ground shifts can cause damaged components.

Perhaps the power supply is where you guys are failing. What PSU did you plan to use?

I'm using one of those AC wall wart with multiple voltages settings, set to 12V. The box says it can deliver up yo 1,8A. The plug is connected in parallel to j1 on the core and to this (http://www.musicfromouterspace.com/analogsynth/WALLWARTSUPPLY/WALLWARTSUPPLY.php) board to get bipolar +/-12V for the aout_ng module.

Now I got scared from your statement since using this supply configuration with the bipolar ground connected is still turning off the box (lcd and aout_ng led turn off). By disconecting the bipolar ground I find 50Hz 20V peak to peak signals at the gate outputs, when at least it should be the opposite, no?

So it could be that I damaged the DAC? Anyway notice that going back to separate wall warts I'm stuck in the out-of-range/steady-out situation

[stryd_one]

Now I got scared ......So it could be that I damaged the DAC?

Sorry mate I didn't mean to freak you out! I wouldn't be too concerned about it, it's a slim risk... it's there though. I think the odds of a problem are currently several thousand to one.

That's an unusual supply but hey, if it's good for MFOS...

It does say though: Make sure the wall wart outputs between 300mA and 1000mA.

....strange to specify an upper limit, as the supply should only supply whatever's drawn.... I dunno...

Mate if I were you I'd do a lot less experimenting, and a lot more waiting patiently for seppo. :D

[seppoman]

and a lot more waiting patiently for seppo. :D

Yes Master :P

First thing to say is: DON'T leave the box running for more than a few seconds while this problem exists - actually it would have been better not to switch it on again at all. Things like "turning off the box" usually mean that there's large currents flowing, like e.g. a short cirquit. If you're unlucky, you can fry everything!

I just looked at the schematics of that supply cirquit and I think I've got something here: Apart from the fact that half wave rectification in conjunction with "only" 12V AC doesn't seem the best way to get a very stable voltage, and that from my experience, most switchable wall warts output DC (you're sure that it's really AC?), I think there's a problem/danger with using a mixture of half wave (MFOS) and full wave (core) rectification. If you look at both schematics assuming that the "upper wire" on the input is the same transformer output on both. On the positive half wave, WW2 (MFOS) and P2 (core) are "GND", but there's already a difference in GND level of around 0.6...0.7 V (because the return path of the bridge rectifier introduces a voltage drop of around 0.6V). Now on the negative half wave, WW2 still is GND, but the bridge on the core makes P1 become the GND. I'm not quite sure if that theory might be utter nonsense but I think this leads to the effect that on the negative half wave, there's a short cirquit between both transformer terminals that's going from WW2 through "GND" and out again through the rectifier to P1.

Now, my first thought was maybe you could try to swap the wires going to the core, but I guess that only "mirrors the problem". But you could replace the bridge rectifier on the core with a wire connecting P2 to GND and a current-capable diode (like the 1N4004) connecting P1 to +. By this, you convert the core cirquit to half wave rectification and as long as the wiring is correct (i.e. WW2 and P2 are the same wire), this should fix your voltage levels.

Hope you didn't damage any parts yet, good luck :)

Seppoman

[spippoli]

thanks to seppoman I found I had swapped all the 1,5 and 15 resistors  :-[

after having them in the right places calibration goes soooo much better, one question though:

What voltage range should I expect for gates in bipolar mode?

I'm getting +/-3V and don't know if it is supposed to behave like this or the range should be wider...

[seppoman]

What voltage range should I expect for gates in bipolar mode?

I'm getting +/-3V and don't know if it is supposed to behave like this or the range should be wider...

gates? are you talking about the CV outputs on the NG or the gate outputs from core:j5? The gate outs should always be 0V or 5V, that's a logic signal. The CV outputs should be able to output either 0..11.7V in regular mode or around +/-5V in bipolar mode.

S