I regularly receive Marshall JMP-1s for service and customers often complain about the skipping or jumping of the data entry control. Cleaning (yeah, right) or even replacing the data encoder, doesn’t always resolve the problem. Hence, I designed the Eclipse Marshall JMP-1 skipping data encoder fix.
Unlike a potentiometer which comprises a wiper that's in constant contact with a resistive track, a data encoder is basically a series of switch contacts with the wiper moving from one contact to the other, as the encoder is turned.
As you may have guessed, this means that there’s a gap between each switch contact (correct) and as the wiper moves, there’s a region of zero contact between the wiper and any switch contact. When the wiper leaves or comes on to a switch contact, the electrical signal can be quite transient and messy, similar to an electrical mains spike that sometimes occurs when switching lights on and off but obviously not as big. This is called contact bounce and if the data from the encoder is going directly into a processor, then it should come as no surprise that the processor just wouldn’t know what to make of it. All it wants to see is noughts and ones at nice regular intervals (a pulse). Modern data encoders are optical and so there's no physical contact between a wiper and a track and hence, no bouncing.
To overcome the problem of bounce on mechanical data encoders, a bounce eliminator (or bounce filter, as it's sometimes referred to) is required to remove the spike, thereby making the signal look more like the regular pulse it's supposed to. What the processor wants to see is kind of ‘underneath’ the spike.
For some crazy reason, Marshall, didn’t incorporate a hardware debounce circuit in the design of the JMP-1. Debouncing can be done in software but generally, designers aren’t keen on this approach. I have no idea if the firmware of the JMP-1 includes debouncing suffice to say that if it does, then it doesn’t work! A hardware solution is always preferred.
One hardware approach is to use passive filters, a combination of resistors and capacitors but due to the unpredictable nature of bounce, this approach can also be unreliable and so Eclipse uses the preferred third option which is to delay the reading of the signals coming off the encoder, thereby missing the bounce. This is achieved using what are called Schmitt triggers.
Fundamentally, Eclipse Marshall JMP-1 skipping data encoder fix, is an active bounce eliminator which can be easily fitted between the JMP-1's data encoder and the main PCB and hence, the inputs to the 8031 processor. The data encoder in the JMP-1 is attached to the motherboard via a 3-way Molex connector. Unplugging this connector and plugging it into Eclipse and then plugging Eclipse into the socket where the encoder was originally connected, only leaves one wire to be soldered to a filtered +5V supply point and you're done! You don't have to remove the main board and as mentioned, there's only one solder point to make. Unless you intend to change the rotary data encoder at the same time, you don't even have to remove the front panel. Simple, eh?!? 🙂
I always seem to have a couple of JMP-1s in for service but just before Christmas 2020, I had one that was of particular interest. The customer mentioned that amongst other things, the data entry control on his JMP-1 was skipping, making it difficult to manually select patches and change parameters. I told him that finding a suitable replacement isn't easy and it sometimes doesn't fix the problem anyway, as the problem isn't actually the encoder but the lack of filtering between the encoder and the processor. Stuart volunteered his JMP-1 to be the guinea pig for Eclipse.
I wanted Eclipse to be easily installable by anyone with a little technical competence but I also wanted it to be as small as possible. In fact Eclipse was originally designed using surface mount devices (SMD) but the build-time was ridiculously long making it just too expensive. With a little ingenuity however, I was able to get full-size components on to the original board size!
So with Eclipse measuring only 37mm x 24mm and weighing just a few grams, I could use self-adhesive nylon PCB spacers, allowing for end-user easy mounting close to where the data encoder connects to the main PCB.
The prototype Eclipse was soak tested in Stuart's JMP-1 all over Christmas 2020 as I put it through its paces, suffice to say, that it all worked like a charm! I'm really happy with this little upgrade. "Simple but brilliant, Mr. Bond".
Although not as open as a potentiometer, the data encoder isn't hermetically sealed meaning that contaminants can get inside the encoder casing. The ingress of dust and dirt and more importantly, years of wear, won't do you any favours when it comes to skipping and like all electronic components, these data encoders have a limited shelf-life.
THE BAD NEWS
The data encoder that Marshall used in the JMP-1 was discontinued by Bourns, the manufacturer, in 2012 due to the lack of RoHS compliancy and is really difficult to find, now. Bourns did release a RoHS compliant version and I'm not sure why even this is hard to track down.
THE GOOD NEWS
Sometime ago, I did manage to find a bunch of original data encoders as used in the JMP-1 and while I try to source an equivalent part, I'm making Eclipse available with the option to buy a replacement encoder. I strongly suggest that you snap one up while I still have some.
I have pre-wired the data encoders with a 3-way Molex connector so that you don't have to fuff around taking the wires off the original encoder and attaching them to the new one. My wired connections are a little longer than the original, thereby giving you a bit more leeway when it comes to the placement of Eclipse.
If your data encoder isn't too bad, if it's just started skipping or jumping, please still consider buying Eclipse WITH a replacement data encoder! Just fit the Eclipse board and keep your data encoder in the supplied ESD bag, somewhere safe.
As I've already said, these things have a limited shelf-life so it will wear. Even if your JMP-1 is in a super clean environment, the contacts will degrade. Eclipse will help to prolong the life of your encoder but a point will be reached when it's just going to have to be replaced. These data encoders are getting increasingly more difficult to procure but hey, you've got a spare! 🙂
If you're getting really frustrated because your data entry knob is skipping and if you have a little technical competence, then the Eclipse Marshall JMP-1 skipping data encoder fix, should be of interest to you. Please don’t hesitate to contact me to learn more or alternatively, you can just...
There's not too much out there resource-wise about the JMP-1 but these links might be of interest:
I'm deeply concerned about the environment and the exploitation of labour and so I always use local manufacturers in preference to the Far East, with the following in mind:
- I can be confident that workers are treated fairly and earn a proper wage.
- I can be confident of the standard of quality of each item that is delivered to me.
- Communication is important and using local manufacturers, all correspondence is quick and understandable.
- I believe in supporting the local economy.
- I can be confident that the disposal of manufacturing waste is managed properly and in accordance with national and EU law.
Using local manufacturers isn’t the cheapest option but the above points are important to me. I hope that they’re important to you too.