latigid on

midiphy Eurorack Expander Modules

114 posts in this topic

The cable is ok. I get the +12 / -12.

Layout would be great.

Thanks Michael

Share this post


Link to post
Share on other sites

Posted (edited)

What rail is shorted?

I always measure +12V-0V and -12V-0V or any rail to 0V before powering up.

Photos of the board would be useful. Do all op amps have the correct orientation? What about the electrolytic caps?

Check the pin headers J1 and J2 for shorts. 0V pins are in the middle, rails on the outside.

The schematic just has all of the 1206 caps lined up, so it would be easiest to tell you where to look. It is a pretty simple circuit: the +12V is on op amp pins 4 and the adjacent caps (C6/8/10/12), -12V on op amp pins 11 and adjacent caps (C7/9/11/13). 

Check all around the power section C2-C5 and the beads L1/2

R1 is the voltage reference, so check there too. 

 

Good luck!
Andy

Edited by latigid on
minus 12V

Share this post


Link to post
Share on other sites

Posted (edited)

I really don't know how this happened, I managed to miss-orientate all of the opamps! :disgusted: At least a learned now how to desolder smd parts with hot air....

After resoldering them in the correct way, the module shows the two top led in red, the 3rd in green the 4th dim red and 5-8 green again. In the CV configuration page 1st led reacts to cv1, 2nd to cv2, and 4th led to cv3. all the other leds show no sign.

Is it possible that i burnt the opamps? I used the hot air (300 C) for about 1-2 seconds on each opamp.

lgm-

Edited by workspace

Share this post


Link to post
Share on other sites

Posted (edited)

All good, happens to us all!

Without any input (i.e. DAC is not sending any voltage) all octal LEDs should be red, because 0/2048*gain - 5 = -5V. This also assumes all of the LEDs have the correct orientation. (As a sanity check, you can test the octal board with a binaire board to verify that all LEDs can be controlled.)

It's possible that the op amps didn't appreciate the reverse power or the heat or both. So I would be tempted to replace all TL074s. 

Measure on both sides of R1, do you get a proper +5V reference? If not, consider replacing the LM4040.

Viel Erfolg!

 

Edited by latigid on

Share this post


Link to post
Share on other sites

The two sides of R1 have +4.98V and +12.00V so I guess the LM4040 is bad as well.

Thanks for you support!

Share this post


Link to post
Share on other sites
10 minutes ago, workspace said:

The two sides of R1 have +4.98V and +12.00V so I guess the LM4040 is bad as well.

Thanks for you support!

Sounds about right to me (could be a voltage meter/probe thing)! But if you've already ordered the replacement it wouldn't hurt to swap it.

Share this post


Link to post
Share on other sites

Posted (edited)

Hi everybody,

I’ve finished my aout and everything seems to work as expected until I do calibration procedure:

 

i begin with the 5v step at 0v using the trimpot .

after that I do 1v, 2v, 3v, 4v steps and when I go back to the 5v step ( with zero calibration correction) the voltage isn’t a 0v anymore.... no matter how many I try to reproduce thoses step. I cannot have a précise calibration and my vcos aren’t tracking properly. ( I use bipolar calibration)

any help will be appreciated :)

Edited by Dimduj
Remove spaces

Share this post


Link to post
Share on other sites

Not sure I understand fully. You adjust the trim pot to get the scaling as linear as possible, i.e. the trimmer should be set so each octave step is as close as possible to the target voltage. Then you use the SEQ encoders to finely adjust for each volt setting. The DAC can't drive negative bits, so there is a limit to the lowest setting. But the remaining octaves should be very linear.

Did you already see @Hawkeye's video on the calibration procedure?

Share this post


Link to post
Share on other sites

Yep, thx for the explanation I thought the trimpot was a kind of offset not a coefficient.

my mistake was trying to set up the trimpot while I’ve values already defined for step 1v, 2v ,... once I setup all the value to zero again calibration was easy and now I got acceptable value for every channel.

is got a precision with 15-20 cent variation between octaves of a channel. Do you achieve more precise values on your side ?

thx for the support 

Share this post


Link to post
Share on other sites

@Dimduj can't exactly tell the absolute values, but we performed an "ear-based" tracking test together with TK. - while we've not got the "youngest ears :)", still to us there was no audible pitch difference between a MIDI controlled digital oscillator to a eurorack CV controlled analog oscillator over many octaves tested. So, i think 15-20 cent note variation sounds high and should be audible when listened to closely. We've not yet performed exact measurements (other than those with a DMM) and just can tell what we heard :).

There might be two pitfalls that cause imprecision:

a) Calibration with a DMM, that was not very precise.
b) The analog VCO used as a eurorack module is itself not tracking perfectly, even if supplied with the correct control voltage.

If you calibrated with a good digital multimeter, after calibration, the SEQ should reliably output the "CV" of every octave with a precision of about two millivolts, that's at least what i was able to reproduce when measuring with the DMM. In my super rough :) calculations, if an octave spans over 1000mV, assuming linearity between notes (which is not the case), each semitone of an octave would have about 83mV of "range". Now, if the DAC achieves an output voltage result that is within the range of say +/- 2 mV for every note, the "error" would be within the range of 2/83 or about 2 cent instead of 15-20. That is assuming the conditions a) (good DMM) and b) (good VCO) are met.

More questions regarding your calibration process:

a) did you have to enter high values to trim the actual DAC output voltages of every octave to get close to every octave's target voltage?
I remember not having used correction offsets of values > 20 per octave during the calibration process. If you needed high values there, this would indicate the DAC would loose a bit of its 12bit output precision, as the SEQ would have to do some more internal interpolation to reach the voltage targets.

b) are all of your CV channels behaving similarly? The DAC channels might have slight performance differences.

c) Can you configure a SEQ track to use a CV out channel, then play all of the C notes over many octaves (in SEQ live forwarding mode) and measure again the outputs with your DMM? Are these DMM readings close to the target -4.000V, -3.000V, ... +5.000V volts? This would be a double check outside the calibration menu. 

Have a nice weekend!
Best regards,
Peter

 

Share this post


Link to post
Share on other sites

Posted (edited)

On 03/10/2019 at 2:54 PM, TK. said:

A downstripped MBHP_CORE_STM32F4 module in Eurorack format would be nice-to-have as an alternative solution to the Euroreceiver - maybe with a SCS (6 buttons, encoder, display) - this would allow to run MBCV which gives us a lot of LFOs, Envelopes, sequencers, modulation matrices, etc.

Best Regards, Thorsten. 

On 03/10/2019 at 11:09 PM, TK. said:

Must-have requirements:

  • STM32F407VG (1MB flash)
  • SCS with 6 buttons, 1 encoder and 1 GLCD (so that we can also use the display as a scope)
  • 1 MIDI IN, 1 MIDI OUT
  • USB Device
  • SD Card
  • SRIO (J8/9)
  • J19 for AOUT
  • 50mm depth
  • HP doesn't matter as long as it still fits into the Pod40X case together with the Expander modules

Optional:

  • up to 3 additional MIDI IN/OUT (could also be provided as an optional module)
  • USB Host
  • 4 on-board LEDs
  • Audio-DAC

Potential extension modules:

  • SRIO based encoder/ledring modules
  • AINSER64 module (with at least 8, but maybe 3x8 INs?) - inputs should be buffered, protected and maybe also amplified and level-shifted for +/- 5V

Such a module could also cover MBNG for script based processing.

Best Regards, Thorsten.

 

On 03/10/2019 at 4:04 PM, Antichambre said:

I'm working on something like this, I try to do not exceed 50mm depth. I need it for the HAARP,
1905_mb-haarp_euro_v1b_core_02.png?raw=1

And adapt the SCS to the euro format is also a good idea, @latigid on ?

I plan to create another CS based on the work I made for @Phatline with a pot for each parameter and TFT screen, like I did for the HAARP.

Best regards,
Bruno

 

I did it... And it works.

  • STM32F405RG (1MB flash), dipCoreF4 based.
  • 2 MIDI IN, 2 MIDI OUT
  • USB Device/Host(OTG FS)
  • USB Host(OTG HS in FS mode)
  • SD Card(Micro)
  • SRIO (J8/9)
  • J19 for AOUT
  • J15A for serial LCD, 2 CS lines.
  • J18 CAN
  • J4 I2Cx2
  • J5 ADCx2
  • 2 Leds
  • User/Reset Button.
  • USB/EURO Power, auto switching with Euro Prior.
  • Power switch acts on both power source.
  • 44mm depth
  • 12HP

Voilà!

large.IMG_1682.JPG.33cfeceb85a6415c9a061large.IMG_1687.jpg.4782138f27a44a3e7f377large.IMG_1683.JPG.7335d1191d33bd64c61df

Best regards
Bruno
 

Edited by Antichambre

Share this post


Link to post
Share on other sites

I recently designed Eurorack versions of DIN and AIN, still need to test but hopefully in the next few weeks,

large.IMG_20200524_111714.jpg.ff3f48dc24large.IMG_20200524_111705.jpg.4b621acfb4large.IMG_20200524_111651.jpg.ff0197e029large.IMG_20200524_111642.jpg.57a1b23ec6

 

@Antichambre maybe you should start your own thread as to not mix up things, thanks.

Have fun!
Andy

Share this post


Link to post
Share on other sites

Sorry, I managed to upload one twice. The quell PCB functions as a terminator, so you clamp unused inputs to 0V.

large.IMG_20200524_112838.jpg.b0f0b5863a

 

Share this post


Link to post
Share on other sites
Just now, latigid on said:

@Antichambre maybe you should start your own thread as to not mix up things, thanks.

It was to complete the discussion about an Euro Core module, discussion which was started here. There's no mix, just a parenthesis. This is fully compatible with your Euro Modules and even made for.
Have a good Sunday my friend.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now