latigid on

midiphy Eurorack Expander Modules

106 posts in this topic

What voltages do you read out of the output with 1/2/3/4/5 V... or min/mid/max?

Looking at the MS-20 manual, it seems to expect a linear 0-8V CV for the VCOs? So maybe normal V/Oct scaling is already good? If you really need to work within the positive range, remove the offset resistor on any channels you would like to use in unipolar mode. These resistors xR34,xR35,xR36x... etc. are listed in the BOM on the transmute8 page.

Share this post


Link to post
Share on other sites
19 hours ago, latigid on said:

What voltages do you read out of the output with 1/2/3/4/5 V... or min/mid/max?

Looking at the MS-20 manual, it seems to expect a linear 0-8V CV for the VCOs? So maybe normal V/Oct scaling is already good? If you really need to work within the positive range, remove the offset resistor on any channels you would like to use in unipolar mode. These resistors xR34,xR35,xR36x... etc. are listed in the BOM on the transmute8 page.

After calibration I got as close as 0.998V 1,997V....9,999V, so that seems working reallly good.

I didn't check the manual, but I've been using my ms-20 for over 20 years, and have been using it with different midi2CV converters and always needed to select V/Hz instead of V/Oct. I also used it on my 'old' SeqV4 with build in Aout board and that always worked good with the Hz/V setting in the CV configuration.  There is a workaround for using V/Oct on a Ms-20. You put the bi-polar CV signal into the Ext Freq input, and turn the corresponding pot to around 9 'o clock. But it's quite hard to get a good stable tuning over multiple octaves this way.

I'll try to remove the resistor and make a unipolar output and see what that does. If it's only offsetting the CV value I can't imagine that it will scale alright, but I'll try and let you know

Share this post


Link to post
Share on other sites

Let me check first if something can be done at the SW side (tomorrow...)

Best Regards, Thorsten.

Share this post


Link to post
Share on other sites
3 hours ago, TK. said:

Let me check first if something can be done at the SW side (tomorrow...)

Best Regards, Thorsten.

 

I will!, and I'm not on a hurry so if it isn't tomorrow, no problemo, thanks anyway!

Share this post


Link to post
Share on other sites

I added some code which should allow proper calibration of Hz/V - could you please check at your side?

-> http://www.ucapps.de/mios32/midibox_seq_v4_096_pre11.zip

Note also: by pushing the GP7 button in CV page we can now switch to bipolar voltage display, which is a bit more convenient :)

Best Regards, Thorsten.

Share this post


Link to post
Share on other sites

Evening all,

I noticed something strange with the midiphy connecting to the Euroceiver board. When i connect the cable to the Euroceiver i introduce a ground noise on my modular. When i remove the cable it is completely silent as it was. Do you guys also have this humm grounding issue ? It is extreme noticable when using headphone on a louder volume.

For now have a nice happy new year!!

Cheers,

Niels

 

 

Share this post


Link to post
Share on other sites

Hi Niels, 

We heard of another similar issue. Could you explain exactly the wiring setup you use? It is important to know the power sources of everything (linear or switching PSUs, power banks, hubs), how they are plugged into the wall, whether you use a computer or laptop, the various jumper settings on the Euroceiver and so on.

Is it particular modules or everywhere that you notice the hum?

Best regards and Happy New Year!
Andy

Share this post


Link to post
Share on other sites

Hi,

I removed all the unneeded power supplies and all is connected directly to the mac and via a grounded wallpower. I forgot the put one cable between the usb and expert sleepers es-9. When i remove the complete midiphy and use only eurorack i get no noise and a clear sound.

Thanks for checking out!

Cheers Niels 

8108BA4F-F5B3-4FBC-9A6F-C667873BC944.jpeg

Share this post


Link to post
Share on other sites

Thanks for the diagram! Just to preface, Eurorack is the Wild West of electronics. There are many ways you can put things together and some will give better results than others. The "fundamental flaw" in Eurorack is that the 0V line serves as both a reference and a current return path (i.e. an unbalanced system). If there are potential difference mismatches on the 0V lines, noise can result. So the idea is to have the 0V resistance as low as possible and to run power rails as much in a "star" configuration as possible.

Searching a bit on MuffWiggler, the Row Power seems to have some history of noise problems. It might be that this is not the ideal way to power your Eurorack though there are a few things to try/clarify.

  • Is the mains connector to the DC brick 2 or 3 pins on the brick end? 
    • With the power off and any voltages discharged (take care!), what is the resistance from the earth pin to the DC barrel?
  • Your MG rack has two Row Powers, correct?
    • Are things actually connected like this on separate rows or more like the hand-drawn diagram?
  • What happens when you disconnect the ES-9 USB?
  • Is it a desktop or laptop computer?
    • If a laptop, what happens when it's running on battery?
  • What is the power distribution in the case? Flying buss or bussboard?
  • Is the Euroceiver powered with the case +5V or the onboard Vreg?
    • (I suppose the Vreg)
  • What happens if you power the Euroceiver with USB (check J5 jumper!) derived from the Mac while also powering the SEQ with USB?
    • Better might be to use a powered hub to increase the available current.
  • What happens if you power the SEQ from e.g. a phone USB charger rather than the Mac?
  • What happens if you remove the Euroceiver J0 jumper? 
    • Generally you do want to leave this connected, but disconnecting it will cut the 0V connection between the SEQ and Eurorack.

Cheers,
Andy

 

 

Share this post


Link to post
Share on other sites

Hi Andy,

Happy New year! Must be a good year because the first thing i tried from you list was to connect the midiphy with usb not via the mac but via the power outlet and goodbye noise. It felt like the episode in breaking bad caching the fly in the meth lab

Thanks again for the help and now for me time to clean up the mess i made in my room finding the issue ;-)

Cheers,

Niels

 

1A96123A-7DCB-456D-AF45-92C1C1C80C41.jpeg

1 person likes this

Share this post


Link to post
Share on other sites

Good news, nice! 

:cheers:

This also seemed to eliminate the issue for the other user. I think it still remains to connect up the modular and the computer at the same time...

Have fun, that looks like a really cool rig to use :happy:.

Happy jams in 2020!

Best;
Andy

 

1 person likes this

Share this post


Link to post
Share on other sites

Posted (edited)

On 28-12-2019 at 4:10 PM, TK. said:

I added some code which should allow proper calibration of Hz/V - could you please check at your side?

-> http://www.ucapps.de/mios32/midibox_seq_v4_096_pre11.zip

Note also: by pushing the GP7 button in CV page we can now switch to bipolar voltage display, which is a bit more convenient :)

Best Regards, Thorsten.

Hi Thorsten, 

Thanx very much for the fast reply, I'm sorry I'm not that fast.... But I think the Hz/V calibration works like it should now, although you really need a unipolar CV output. And I didn't notice any difference in the calibration procedure but maybe you just changed the code and not the procedure?!.... The renaming (Aout > CV 1t/m4) and the GP7 uni/bi polardisplay button is indeed more convenient and looking good!

I didn't remove the resistors yet, but I measured the following after the Hz/V calibration:

C2 on the sequencer gave me -1V,  E2=0V,  C3=3V,  E3=5V .   If I add 5 volt to this (changing from bipolar to unipolar) C2=4V,  E2=5V and C3=8V and that's is exactly what it should be so that's why I think the calibration is correct. It is so confusing cause the CV input range of the ms20 is from 0 to 8 Volts, so the ms20 starts playing on E2/0V, and the highest reachable note in bipolar mode is only 1 octave higher E3/5V, and within this octave the ms20 jumps through 4 octaves. That's the difference between lineair and exponential, I was never good at this stuff, I hope you people understand me...

Thanks again for all the help and the new release, looks very promissing, 4 CV modules is superb and track labeling and all other stuff is just great.

Edited by Elektruck

Share this post


Link to post
Share on other sites

Thanks for the valuable feedback!

18 hours ago, Elektruck said:

And I didn't notice any difference in the calibration procedure but maybe you just changed the code and not the procedure?!....

yes, by intention I wanted to keep the procedure "as is".

Best Regards, Thorsten.

Share this post


Link to post
Share on other sites

@TK. cool!

 

21 hours ago, Elektruck said:

Hi Thorsten, 

Thanx very much for the fast reply, I'm sorry I'm not that fast.... But I think the Hz/V calibration works like it should now, although you really need a unipolar CV output. And I didn't notice any difference in the calibration procedure but maybe you just changed the code and not the procedure?!.... The renaming (Aout > CV 1t/m4) and the GP7 uni/bi polardisplay button is indeed more convenient and looking good!

I didn't remove the resistors yet, but I measured the following after the Hz/V calibration:

C2 on the sequencer gave me -1V,  E2=0V,  C3=3V,  E3=5V .   If I add 5 volt to this (changing from bipolar to unipolar) C2=4V,  E2=5V and C4=8V and that's is exactly what it should be so that's why I think the calibration is correct. It is so confusing cause the CV input range of the ms20 is from 0 to 8 Volts, so the ms20 starts playing on E2/0V, and the highest reachable note in bipolar mode is only 1 octave higher E3/5V, and within this octave the ms20 jumps through 4 octaves. That's the difference between lineair and exponential, I was never good at this stuff, I hope you people understand me...

Thanks again for all the help and the new release, looks very promissing, 4 CV modules is superb and track labeling and all other stuff is just great.

When removing the offset resistors, consider that the LEDs on the octal PCB will be driven by a higher voltage (current). So consider at a minimum doubling the LED resistor values or even swap the LEDs on unipolar channels to  different colour (with appropriate resistors).

Share this post


Link to post
Share on other sites
On 01/01/2020 at 2:26 AM, latigid on said:
  • What is the power distribution in the case? Flying buss or bussboard?
  • Is the Euroceiver powered with the case +5V or the onboard Vreg?
    • (I suppose the Vreg)
  • What happens if you power the Euroceiver with USB (check J5 jumper!) derived from the Mac while also powering the SEQ with USB?
    • Better might be to use a powered hub to increase the available current.
  • What happens if you power the SEQ from e.g. a phone USB charger rather than the Mac?
  • What happens if you remove the Euroceiver J0 jumper? 
    • Generally you do want to leave this connected, but disconnecting it will cut the 0V connection between the SEQ and Eurorack.

Cheers,
Andy

 

 

Hi @latigid on

I thought i was happy running the midibox from USB till i noticed that my productivity went down the drain like having to good midi clock for syncing to the daw etc. So i decided to dive deeper into this issue because i like to have the midibox connected to my computer via usb and have the eurorack connected as well.

The power is connected by flying bus boards but i have no issues with other modules and noise. Also moving the bus cables does not make any change in noise. I also powered the euroreiver by using an usb power from a phone adapter and jumpered j5 the right way. This has no impact on the noise. Then i cut the 0v by removing the J0 jumper and that cut the noise directly and the module is still working like it should. Since i am no technician at all, what is the inpact of cutting this 0v ? 

Cheers,

Niels

 

 

Share this post


Link to post
Share on other sites

Hi Niels,

 

The power situation is fairly complex and there could be more than one source of noise. Some setups have no issues, some do. With MIDI obviously we have optoisolation so things are different there.

The power distribution in the SEQ could be better e.g. by improving the return current path (certain Molex connectors on the lemec boards/USB PCB are meant for this). I think that the LED matrices introduce some digital noise as they switch. I'm not sure if this the source of your noise or whether it is the more commonly observed phenomenon encountered when connecting USB audio devices. Could be that your USB +5V from the computer is not designed for the current that you draw from it or is inherently noisy. Peter tested a laptop and USB power bank, which were both quieter than even a phone charger, but a PC tower had a much noisier +5V line. He could filter out some of that noise but it was still worse than the laptop. His Eurorack PSU is transformer based with linear regulators and uses a copper bussboard. I would consider this system superior to a switching PSU and a flying busboard.

I would tend towards the powered USB hub like TK. uses for the SEQ. Another user still had issues with their hub but then one might imagine that that PSU was also noisy. Very hard to guess. There are also solutions to galvanically isolate USB and could be worth a try? Some are quite cheap and some much more expensive. I can't guarantee anything though, so it might be throwing money at the problem without a satisfactory result.

For the Euroceiver ground lift, the RS-485 standard recommends to keep the 0V reference connected, even though the power supplies on either end could be mutually floating. Imagine an extreme case where you have a long signal cable and a device either end. You have a massive antenna and a real "ground loop" if the power supplies are connected. In our case, we can have two different power supplies and we keep the signalling within the common-mode range by referencing to a common potential (0V). Without J0 jumpered, the signals may become more susceptible to data corruption as the differential voltages drift apart. You might notice that the setup works in your studio but not at a gig for example.

Instead of leaving J0 unjumpered you could try to connect a low-value resistor (say 10R) of suitable power rating across it. This might attenuate the noise that you hear.

Best regards,
Andy

1 person likes this

Share this post


Link to post
Share on other sites

I noticed that isolating the earth on the PC / Mac power supply can sometimes help.
Bruno

Share this post


Link to post
Share on other sites

Thanks for the detailed answers! i did some testing with some other computers i have lying around and they all produce the same noise like my mac mini does. The noise gets louder when i play some notes on the midibox seq and when i stop playing it gets less. In my case the only thing i need to get the midiphy connected to my computer is for midi clock. Since my UAD interface does not have midi out i am trying to do this via the usb of the midiphy but maybe i step off this plan. Andy and Peter where so kind to provide me with a Multi midi thru pcb gift and maybe this is exactly what i have to build to tackle this noise problem. Since you already are saying Andy. Ground noise is a hard to find problem and since i am not really a technician in finding these kind of problems i think i am going for the midi interface road and provide usb power to the midiphy by an usb power supply and not via a computer. Also what i forgot to mention earlier is that i connect everything via an usb hub that is externally powered but this has no impact when i remove it and connect it directly to the mac mini. Also the mac mini has no ground power connection. let me build a midi thru interface from Midiphy and see if this works like i want to :) Thanks again for your help!

 

 

Share this post


Link to post
Share on other sites

Just an update. I went to a local store and bought myself a Roland UM-One mk2 . The cable cost me 39 euro and when connecting it like described above i get no noise. A very good midi clock to the midiphy from Ableton and Bitwig . Updating with Mios Studio works. Bad thing is that it requires drivers but hey for now a driver is better then noise :) Probably and midi interface would have done what i now describe but just for the info :)

Cheers Guys, i am happy again. Goodbye to ground noise and hello to white noise :) 

 

 

1 person likes this

Share this post


Link to post
Share on other sites

Hi Niels,

going MIDI (and using opto decoupling) is a cool, easy and pragmatic solution!

As Andy wrote, i did a few local measurements with all devices i had available out of curiosity, because i also sometimes experienced ground hums between some synths and my mixer and i was interested! :) Basically, i cannot reproduce your problem, i can see a minor bit of digital noise introduced to the eurorack system, when using my PC instead of my Laptop to power the SEQ/Eurorack modules over USB, but it is perfectly acceptable and there is no ground hum problem, scoped waveforms from multiple measurements are available, if requested :).

This post by Graham Hinton seems to be a really good reference regarding the ground hum problem in general, he is very nicely explaining what's going on and how you could address it:

https://www.muffwiggler.com/forum/viewtopic.php?t=102801#p1429534

In my own words (i may be wrong scientifically with some terms, as i am not an EE :)): when you experience ground hums, that is probably due to a different ground potential, e.g. between Eurorack PSU and PC PSU -or- one or both of those PSUs themselves being noisy, e.g. not being properly filtered. So, for example, if in one unit we have some unfiltered AC mains or switching regulator ripple being introduced, it will interfere with the other PSU over long and imperfect ground connections between those PSUs, over the PC<->SEQ<->Eurorack wiring, which is going over a long distance and many connection points, i.e. starting with the flying busboard wiring in your eurorack, the DB25-cable, the ribbon wires in the v4+, the USB cable and the wiring in your PC/PSU.

Graham also explains, how different potential can exist, even if all affected PSUs are of high quality and are well filtered. It occurs, if the two independent rails draw unequal currents (statically or dynamically).

Graham suggests to tackle the problem at its roots and run ground bonding wires between all affected PSU chassis/0V. Note, that also your mixer/powered amp/powered studio monitor elements might play a role in the audible ground hum noise and might benefit from proper external grounding too. Often, that step is not necessary, as the AC mains wire ground connection is completely sufficient, but sometimes it might be helpful.

Also note, that the AC mains cable ground connection should always stay in place for safety reasons, be very careful with ground lifters! But, if you create your own improved "studio ground", that will basically circumvent any suboptimal ground connections and should help a lot. This method could also be employed, when you experience ground hums with other audio gear, e.g. between a synth and your studio mixer. That is also why you will sometimes find external "ground" connection points on some studio gear, for example i've got an audio patchbay with an external "ground hook" - adding proper grounding there makes a lot of sense, as it will affect all connected sound sources/synthesizers.

Have a good upcoming weekend and best regards!
Peter

Share this post


Link to post
Share on other sites

Wow Peter, Thanks for the helpfull article. To be honest that when i read the Graham article i only understand 10% of it and what i understand from others is that it is a hard to understand or fixed subject. When looking how other People build their studio and big eurorack systems i probably started differently then the stuff i have bought right now but getting into eurorack is costly and selling and buying new stuff to avoid humm is not in my 2020 todo list ;-) for now fixing it by a workaround is good enough for me and saves money on buying new cables and psu's and new cases. I find it a really interesting topic like all things with electronic engineering and can only have deep respect for People knowing this stuff. Too bad in my life i probably never get this skill at a good level because of time ;-) i really like to solve this ground thing from the core so i can understand it but i am throwing the towel with a workaround because for now i like to create music with a nice sequencer. Thanks again for trying to let me understand these kind of things.
 

Cheers,
Niels

Share this post


Link to post
Share on other sites

What kind if db25 cable do you use? I see super shielded expensive ones and very cheap ones. The one i have right now  is around 1.5m and i like to get a shorter one.

 

Cheers, Niels 

Edited by sis.tm

Share this post


Link to post
Share on other sites

Hi Niels,

i'm using the same (cheap) DB-25 cable from Reichelt, about 1.5m long. It's also unshielded and you are right, it might pick up some noise, even a ground hum if you run it e.g. close to a transformer.

But: my expectation would be that while reducing the DB-25 cable length and upgrading to a better/shorter USB cable (e.g. with a ferrite ring) will help to battle the minor digital noise introduced by the PC, it will probably not help with the ground hum.

What you could do, if you are still unhappy with the MIDI-only solution is to try out one of these three options:

1) Try with a galvanic USB isolator, e.g. Reichelt has this one: https://www.reichelt.de/usb-isolator-with-5-kv-isolation-converter-delock-62588-p151307.html

It is affordable and if bus powering is not sufficient you can use e.g. your 5V USB phonecharger-type PSU (that was unproblematic when powering your v4+/Eurorack modules) to power the galvanic isolated device (the v4+) with it. In that case, you'd need to add a mini barrel connector to your USB phone charger PSU, or buy a USB <-> mini barrel connector cable, but that should be no problem! :)

2) If you are willing to upgrade your studio PC, some laptops that have e.g. "travel adapter" PSUs without an AC ground pin should help - that will technically "lift ground" on your computer side, but in a safe way: https://www.amazon.de/Lenovo-Travel-AC-Adapter-65W

3) Maybe the cheapest method: as Graham suggests, improve the grounding between your devices externally with a well-dimensioned ground bonding wire, it might be necessary to include your amp/studio monitors as well. You'd need to make proper cables with cable lugs for safe connection and find proper "ground points" to which you can screw them on every chassis for it to work.

---

All of these three solutions, at least in my thinking should eliminate the ground hum:

- Methods 1) and 2) "lift ground" on at least one PSU, making the interal internal ground connection (long wire from flying busboard -> db25 cable -> v4+ -> usb cable -> PC) the only connection path, there then can be no potential difference anymore, no "ground loop" however controversial that term is :).
- Method 3) will directly "shortcut" the potential difference while every PSU can stay connected to mains ground.

Many greets,
Peter

Share this post


Link to post
Share on other sites

Hi everybody

I have problems with my eurorack expander a1 cv module. It seems that I have a short somewhere on the transmute8 board. The tiptop uzeus (eurorack power supply) is blinking, which tells you that the module is shortening the power. This also happens if I connect the transmute8 without superdac and octal boards. I already checked for solderbridges and orientation of the powerplug. Is there a schematic of the board I can use to find the short? and whats the best approach to find it?

Best Michael

PS: I attached the cv module directly to J19 of my core module. this should work, right?

Edited by workspace

Share this post


Link to post
Share on other sites

J19 should work as connected!

What rail is shorted? I can show where the rail is on a layout, that's probably easier.

I would say to check the cable, but you're using a flying bussboard, so if another connector is fine then the issue must be on the transmute8.

Best,
Andy

 

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