Novation Circuit pad/ui glitch teardown and FIXed !

History / Motivation

I got a the Circuit by Novation back in Fall of 2018. I was blown away by how simple and flexible the UI was, both for sequencing internal sounds, and as a MIDI note-controller for external gear.

Then, suddenly, it started glitching, at the pads.

It seemed fine if I held it ant an angle, but placing it on flat surfaces made it spazz out Then, about a year, the unit’s pads started falling into glitch-fits, as if I was tickling all over the pads

… as seen in this other person’s video:

 

Initial “lemonade” ?

I tried to make the best of it at the time, by sampling the garbled noise of the over-stimulated synth-engine into a new noise-composition.

Then, wanting to salvage this clever-and-portable controller/synth, I contacted Novation.

They promptly responded that, since the unit was no longer under warrantee, that I’d have to ship it to them for diagnostics and repair.

Screen Shot 2019-05-23 at 3.53.39 PM.png
a polite, but discouraging response for my predicament.

SO…

  • $32 to look,
  • $65 an hour
  • 2+ weeks turnaround
  • alternative to buy a new model for $120…

…I decided to repair it myself:

To the BENCH !

Then, after a few months of deliberation, I decided to open it up and fix it myself. I followed this very-well-phographed teardown by BoodyHole (who was just cleaning out beer)…

So, I pulled apart, and found the problem was a very simple fix !

IMG_20190426_202004.jpg
first, take the caps off the Macro knob potentiometers.
IMG_20190426_202030.jpg
the rubber bottom is glued on. A hairdryer-on-low-heat softened the bond enough to get under it with my “spudger”
IMG_20190426_202040.jpg
peel slowly to preserve tack of the heat-activated glue.
IMG_20190427_112945.jpg
try to keep all the glue on the rubber side..
IMG_20190427_113501.jpg
use a spudger to CAREFULLY pry apart the top (UI) and bottom (motherboard) plastic housings. Note that the UI board and MotherBoard are connected by 3 ribbons. You WILL have to support this open-hood to unplug them before getting this further open.

 

 

one standoff for the power/battery lines.

IMG_20190427_113613.jpg
one standoff for the power/battery

 

IMG_20190427_113648.jpg
now we can get to Speaker and Headphone lines on the other side.

At this point, I noticed the two ribbon-cables between the two boards each had a magnetic shield, so I used some cellotape to keep each from being jerked around before proceeding…

 

IMG_20190427_113835.jpg
careful to remove the board with I/O jacks. I used some tape to hold down one to the under (bottom) side of the I/O PCB…
IMG_20190427_114317.jpg
then, where the other ribbon wanted it’s magnetic ferro-shield to rest on the other side, I taped the “front” (top) of the other board. This would make it MUCH easier to remove/replace these ribbon busses without being strained by the weight of these little shields.
IMG_20190427_114405.jpg
are clamped in by clip-hooks. I used a spudger to (carefully) open these from opposing sides..
IMG_20190427_114533.jpg
…so that each one left it’s weighty ferro-sheild attached to it’s local board. Analysis and prep like this makes it WAY easier to put back together.
IMG_20190427_114552.jpg
unscrewing the dedicated MacroKnob PCB.
IMG_20190427_114724.jpg
and pulling it out of the way to to get to the Pad PCB from the back…

it was right HERE that I found the culprit (revealed below).

IMG_20190427_114815.jpg
this should be where the problem is…
IMG_20190427_114840.jpg
Shown open here, we see the main pad UIs are work by having the the (backlit) silicon pushes an insert of (black) conductive rings onto the respective traces.
IMG_20190427_115010.jpg
the removable insert has rings of conductive material for each pad.

IMG_20190427_114925.jpg
a closeup of the fine trace blossoms shows how tight the tolerance would be for small motions, and how it might be vulnerable it might be to interference from (small) foreign metal bodies…

 

 

SideNote: SO HOW do these pads work ? (when they work).

At this point, I realized that the pads of the Circuit work quite differently than most other drum machines I’ve pulled apart. Most others have simple membrane switches for detective IF a button has been pushed, along with a piezo impact-sensor to catch how hard unit was hit. These are analyzed together to assign velocity to the last-hit-pad.

However, on the Novation circuit, there are no piezos.

I pulled a multimeter to check the rings.

IMG_20190427_115120.jpg
…relatively more-resistance resistance / less-conductive along the full length of the black trace.
IMG_20190427_115149.jpg
relatively more conductive along a shorter segment.

Each pad pushes it’s semiconductive ring on a the square bloom of copper tracing.

I couldn’t find a (traditional shaped) piezo in the unit, so I guess the unit might be detective not only IF/when these rings mate, but how forcefully.

The unit might be scanning the (relative) conductance or voltage-division off these pads continually, acting as continuous force-sensing resistor assemblies. This would allows the control we already see from these pads:

  • hit-events to be detected by looking for sudden changes per pad.
  • key-holding action, where you can hold a pad like a keyboard key for indefinite sustain (something NOT regularly possible on piezo-equipped drum machine interfaces).

With this, I wonder if such a pad-design could support after-touch…even poly-phonic.

anyhow…

 

…but WHAT WENT WRONG !

As mentioned above, when pulling up the PCB for the MacroKnobs, I heard a metal part fall to the desk.

IMG_20190427_115653.jpg
The culprit !

This little slug of metal had been rattling around between the knob board and the pad board, likely shoring the traces that handle scanning/status of the pads (hence the “storm” of input”).

That metal tab is the anchor for the Kensington Lock slot, a feature that allows you to bike-lock your music gear to the retail/studio environment of your choice at the hazard of forcfully ripping the body to remove it.

IMG_20190427_115415.jpg
The Kensington Anchor slot is seen top-right of this pic.

The metal anchor fit neatly int the little slot, but was not glued in, so it not only feel inward to the calamity hear, but it could get moved/lost lost whenever someone hazards to open their circuit.

To fix this, I decided to super-glue the metal tab into place before re-assembling. If you EVER consider doing this yourself, be VERY careful to glue the metal to the “top” (black, pad-sided) chassis, and not the “bottom” (blue, rubber-backed) chassis.

NOT here like this…

IMG_20190427_120246.jpg
NOT HERE… do NOT make this mistake, as will risk making getting glue directly between the black and blue chasis, and GLUING YOUR CIRCUIT SHUT ! (at one corner).

 

…but HERE, like THIS:

IMG_20190427_124148.jpg
if you glue the metal to the black/front chassis, it wont bind glue between the two chassis.

Man, removing my mistkenly-applied glue lead to considerable time wasted, but still avoiding the potential tragedy of sealing my Circuit shut… gotta think ahead, and consider what’ll happen when you put it back together.

Aftermath, afterthoughts…

After getting the unit re-assembled, it worked fine, and hasn’t glitched since (fingers crossed).

Having a second-life of functionality (and a new appreciation for the robust key-action) of my Novation Circuit has got me exploring this thing as a MIDI-keyboard-for-drummer… even looking into how I might store and play chords with and without the sequencer.

Bottom line…

Instead of paying Novation

  • $80+ to service my unit
  • OR
  • $120 for a new unit…

I spent 2 hours, 1 beer, and no money to fix it myself. I’d hope that, if Novation finds that “pad-glitch-ing” units consistently show up with their loose Kensington anchors loose, they should offer a special repair-rebate program… and spend the money to glue down such loose metal parts in future editions.

If you’ve got a Novation Circuit with glitching pads, let me know. I’ll be happy to fix it for or with you.

 

 

34 thoughts on “Novation Circuit pad/ui glitch teardown and FIXed !

  1. Shubhang says:

    Hey, I have a novation circuit. It had glitching pads just the same way you showed in your video two years back. It was in warranty so I got it fixed for free but a few time back this problem occurred again and on top of that it is dead now like it won’t even switch on so that I sample those glitching sounds lol. Please help me dude.
    Thanks

  2. Kevin M McCoy says:

    hey Patrick–Great job! I found your post after mine just started doing this today. Given the fix, I wondered if maybe just some shaking / light hitting of the unit would clear up the issue (temporarily at least!) and so far it has. I’m assuming whatever metal piece is now in a benign corner of the unit. Thanks again for the diagnosis; If I get the energy will have to open it up and fix more permanently. Kevin

    1. thepatrickrichardson says:

      Hey Kevin. I would NEVER settle for letting a piece of conductive metal rattle freely around inside, as it may move to short out something more serious (like around the power supply). Let me know if you want help opening it up.

      1. Tristan says:

        It feels more like the PCB flopping around inside, but in your teardown there are screw(s) that look like they would prevent that. I can take some video of the “rattle” later tonight. It’s deeper, more of a “shuffling” sound lol. Definitely not the Kensington tab thingy.

  3. Chromatest says:

    Hey! That’s my video up there! Glad it’s getting some use. I took it to send to Novation support. My unit was replaced under warranty, and at the time there were no tear-downs available for it. I’m a total fix-it guy, so if it hadn’t been under warranty, I would have opened it up.

    1. thepatrickrichardson says:

      Hey Chromatest ! Great to have you come by !
      Yes it is your video, and thanks for sharing/relating your problem for me to find and work on.
      Since my finding your video and blogging a tear-down, i’ve gotten TONS of inquiry about other people with the same problem. Might be my biggest source of blog-inquiry (so far).
      I’ve even done some repairs for people who sent their units to me (voiding their warrantee, of course).

      I’m surprised Novation have remained quiet on this matter.

  4. Stephan Möbius says:

    Thanks for this. Your post encouraged me to take it apart. Even though i did NOT have the luck of finding a loose metal piece as the culprit, the pads worked after cleaning them with alcohol. Hooray! Again, thanks.

  5. J4TUN says:

    Thank you so much for this walk-through. I was having the same issue with my circuit. Pads glitching. I let it sit for months and eventually tried to send it in but because I bought it used they were going to charge me for repairs. They said it could cost up to the price of a brand new one. So I followed your tutorial and it only took me about 45 mins. Same thing, the Kensington anchor was loose. Glued it back in and its good as new. I can’t thank you enough! Such a blessing. Thanks!

  6. mres says:

    Hello, thank you very much for your guide to open it, my circuit has been failing for some months and as the colleague says above with a small blow it was fixed temporarily … luckily that solution worked for me when the problem presented itself to me in full presentation about it scenario … I will encourage myself to open it to repair it definitively … I will comment on my results.

    greetings from Chile

  7. Hugh Chapman says:

    Many thanks for this blog post Patrick. I was having issues with an OG Circuit I recently bought second hand. I opened it up and there it was – the very same metal tab from the Kensington Lock slot rattling around inside. I just took it out – I don’t expect I’ll need to use the lock slot so didn’t see the need to glue it back in place. Fingers crossed that’s solved my issues – all seems well so far.

    I think this is a design flaw. I think maybe it’s just gravity holding the metal tab in its slot. So when the Circuit is turned upside down – as might well happen in the post say – the tab gets loose inside the Circuit and is entirely likely to short out all kinds of circuitry. Not sure if Novation have resolved this with newer Circuit models.

    Cheers!

    Hugh

  8. Hugh Chapman says:

    If anyone is having similar issues in the UK and would like me to take a look at their Circuit feel free to drop me a line by replying here

  9. poesiebrute says:

    Hi,

    My Circuit was glitching for two years and was sleeping somewhere in the attic. Thanks to your tutorial, I got it repaired in less than five minutes. You’re my hero! Thank you so much !

  10. felipe says:

    Hello, a potentiometer has broken. How do you change them and where do you buy the potentiometer?

  11. 0F.digital (@0F_digital) says:

    Hey, just thought I’d add another comment saying thanks for this detailed teardown and solution. I work at a library makerspace and our Circuit had been glitching out for months, with a few moments of working properly. I followed your guide and it was exactly the same issue you had! It was bothering me for ages and Novation had nothing to say about it when I pinged them on Instagram.

    1. thepatrickrichardson says:

      Great to hear.
      I’m a strong supporter of Right to Repair, as it’s part of
      “Reduce, Re-use, REPAIR, Re-purpose, Recycle”

      Where is your maker space ? I did a bit of MakerSpace work in Philadelphia.

      1. thepatrickrichardson says:

        Thanks.
        I meant the Comment on the MicroSampler for another thread.
        I checked it out, but it looks almost “too small”….and I’m not sure I’d dig the small screen.
        I’ll definitely check it out to see how “hands on, eyes off” it is.

  12. bc1009 says:

    Thanks for documenting the disassembly so well! Just thought I should mention that, on my Circuit, the problem has never been the Kensington slot. The metal shim has never been loose inside the case, and I’ve never had to glue it in place. Instead, the problem in my experience (and the experience of many on the official Circuit Facebook group) seems to be the ribbon cables connecting the pads to the processing board. Re-seating these often solves the problem for a while, but eventually I find they come loose again. I’ve actually drilled holes in the rubber foot for my Circuit to make this job easier, as it’s happened probably 5-6 times since I bought it used in 2019.

    Bottom line: if you follow this guide and the Kensington slot shim isn’t loose, and/or gluing it in place doesn’t work, try re-seating the ribbon cables.

    1. thepatrickrichardson says:

      Excellent observation, science, and engineering my friend BC1009.
      Awesome move to give your Novation Circuit an acess vent (“undercarriage sun-roof”) to fix the problem without having to pull panels.
      Baller customization. Check out my custom work on my Korg MicroSampler !

  13. CircuitGlitchOwner says:

    Just wanted to say thanks and yes my circuit had the exact same problem with the loose Kensington lock. Once removed, it worked perfectly. Looks like a design flaw indeed!

Leave a Reply to thepatrickrichardson Cancel reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s