Roland Juno 106 (#7) – this is New Jersey

Roland Juno 106
Roland Juno 106

Sold! Now FOR SALE in our Reverb shop!

I bought this Juno from a guy named Mike in South Jersey, with a shaved head and a gold chain. I met up with him at his garage studio behind a roller rink, where the Juno had been used for over 25 years by musicians providing live accompaniment for the roller skaters. Now the music is performed by two “old cowboys” (his words) on two Hammond console organs. I also met Mike’s overweight English bulldog, who he introduced to me as his girlfriend. I think her name was Bethany.

By the time we were ready to start preparing this to be sold, I had already pulled a couple of voice chips out of it to put in other Junos, so I took this opportunity to do a Juno with all AR clone voice chips, as I know that appeals to some customers. However, there were other issues that had to be fixed before the voice board was ready. Two of the oscillator signals (saw waves on voice 1 and voice 4) were getting lost before reaching the actual voice chips. People say that the next thing to suspect in this situation is the wave chips, but we traced the issue to the 4052 multiplexers that route the wave chip’s outputs to the voice chips. Two out of the three had bad switches on them, so we went ahead and replaced all three. Old CMOS 405x series multiplexers are notoriously unreliable in general and I’ve replaced a fair few of 4051s, 4052s and 4053s in other synths as well.

Other details on the rest of what we did can be found in our listing on Reverb!

5 thoughts on “Roland Juno 106 (#7) – this is New Jersey”

  1. Hi Alison. Firstly, I want say awesome blog and love your work!

    I’m currently trying to bring my Juno 106 back to life after sitting gathering dust for a couple of years and I’m having an issue with one voice (3) being really distorted and grungy only when the saw wave is engaged (no issue with square or sub). I’ve reconditioned the VCF/VCA chip but the issue persists. The thing is if I swap VCF/VCA chip 3 with another working VCF/VCA chip (e.g. 6) the distortion still occurs on voice 3.

    In your experience, does this sound like it could be the wave generator at fault? Reading this blogpost it’s got me thinking that it’s possibly the 4052 multiplexer…any advice on running diagnostics on the 4052?

    1. Hi Sunny,
      It could very well be either the wave generator or the multiplexer– I have seen the failure of both cause similar symptoms. Furthermore, when you have a bad multiplexer, even ‘scoping directly at the saw wave output pin of the wave chip you will see a deformed waveform. So there’s no easy way to test where exactly the problem is happening, though with the help of the 4052 datasheet you can check its logic gates to see if any are misbehaving. But since the CD4052 multiplexer is still in production and you can get one for under $1, my usual approach is to just replace it first and see if it fixes the problem. Sometimes this less-scientific approach is worth choosing to save time, especially because we keep these parts in stock at all times and can desolder a chip like this in just a couple minutes (of course, if you don’t have good desoldering equipment and skills this may not be so simple).
      So basically, I’d suggest try replacing the multiplexer first!

      1. Hi Alison!

        Thanks for taking time out to reply – if I wasn’t in Australia and local to you, I would have just brought it by your shop.

        The good news is, I’ve fixed it! One of the wave gen chips was the issue but in the process of removing and replacing it I lifted a trace…one jumper wire later, she’s back firing on all 6 voices. Tomorrow’s calibration time 😎

        Again, really enjoy your blogging of your repairs…I work at a library where we have a maker space and I’m getting in deeper and deeper with electronics. Going to be checking Bell Tone Synth Works regularly for new insights into the world of music tech repairs.


  2. Found this page because I was googling “Juno 106 bad multiplexer”. Can you elaborate on other instances where you had to replace a 4052?

    The issue with my Juno 106 is that the LFO speed affects the PWM of the square wave (even when set to manual). The circuit is pretty straightforward – CV voltage is divided by the sliders, passed on to the multiplexers, buffered, then directly to the CPU. So my hunch is that it’s a bad multiplexer, but honestly never had to deal with a bad MUX.


    1. A bad MUX can indeed allow a signal to “bleed” through when its supposed to be switched to be sent elsewhere– eg allowing the LFO to bleed through and be sent to PWM when only manual PWM level is supposed to pass through. Look at the schematic and trace the path of the digital messages that determine which PWM CV (manual or LFO) is sent to the wave chip. It’s a series of a couple of different logic chips. You should be able to trace a series of connections starting with the PWM CV mode switch, through ?? (don’t have time to look it up right now but it will be good for you to figure it out yourself!), to the CPU, back out of the CPU, to the CV MUX IC23, etc until it reaches pin 12 of the wave chips. Somewhere along there is your issue!

Leave a Reply

Your email address will not be published. Required fields are marked *