Roland BK-7M and FR-8X MIDI Frame Error and MIDI Disconnected

[Thomas N]

Hello all -

I know a lot of us use this combination. It's been working fine for me for probably close to 10 years. About a month ago I started seeing "MIDI Frame Error" pop up on the BK-7M screen and the chords wouldn't change and the pattern would become unstable. It would clear itself after maybe 5-10 seconds. Sometimes it wouldn't appear for a few days. It started becoming increasingly more common. Eventually it would say "MIDI Disconnected" and put itself into the Wizard Setup screen where the only solution was to select Accordion---V-Accordion----Write.

This eventually turned into constant MIDI Disconnects and now I cannot connect the units. I did a complete Factory Reset on both the FR-8X and the BK-7M, and verified that they are both on the latest version of firmware (2.52 and 1.07 respectively). Still doesn't work.

A quick Google search of this error yields a single question from one user years ago who had the same issue with this combination. In this user's case he was able to use another MIDI device and connect it to the BK-7M and it still provided a disconnect error which meant the issue was with the BK-7M and not the FR-8X.

I sent in an email to Roland support but I don't expect much from it. The BK-7M is discontinued and my only real hope would be that this is a common known problem.

It's very odd because nothing had changed in either device. I tried different MIDI cables, blowing air through the MIDI ports gently in case there was some debris. It almost seems like a software problem which is really strange to manifest after 10 years.

If anybody has seen this please let me know! Thanks!

 

[JerryPH]

Change your MIDI cable for a new one… about $15. If it continues there may be something more serious happening With the plugs wearing out or the circuit board. The BK-7m is discontinued… so an email to Roland won’t do a thing. Try reaching out to an authorized Roland service centre.

 

[Thomas N]

Change your MIDI cable for a new one… about $15. If it continues there may be something more serious happening With the plugs wearing out or the circuit board. The BK-7m is discontinued… so an email to Roland won’t do a thing. Try reaching out to an authorized Roland service centre.

Thanks Jerry. I bought a Roland 10-ft Gold MIDI cable just to eliminate any cheap MIDI cable I had laying around. It didn't change anything so to the Service Center it will go!

 

[Airy]

On older type electronic devices intermittent and increasing failures are often a sign of a dead memory-backup-battery. And 10 years is a good lifespan for this type of batteries. Is there such a thing in the BK-7M? These batteries are mostly not meant to be user-servicable but any electronic savvy friend or service shop should be able to replace that one.

 

[Ventura]

that is an excellent suggestion

if this happens to anyone in the DC region with a BK you can come over
and we will pop it open and have a look..

 

[JerryPH]

This is what is inside a BK:

 

[Ventura]

i couldn't spot a battery in the video

don't see anywhere it would be hidden from view

don't see a trickle capacitor circuit either
(some backup systems used a high farad low voltage capacitor
that would store a good chuck of energy that has the same
function as the usual coin cell.. such devices were usually
of a type that were used frequently as the cap needs to recharge
before it is depleted)

so to pursue the backup trickle circuit idea we would need to
see the schematics and find iF it exists there, then
follow that roadmap to the component

for the device in general, as you can see the chips are not field
replaceable (if a chip is the failure) and you must replace the
board sub-assembly (where you find the fault to exist)

the only other things to look closely for are any signs of discoloration
(chloride residue/moisture affecting the traces) or a crack due to
a flex from incorrect mounting

after that you can always touch (re)solder all potentiometers, jacks,
power plug connections, etc to assure there is no hidden cold solder joint

by contrast, all 3 i5M's, the i40m. and the i300 (Korgs) i own still are fully functional
all these decades later, except for the floppy drives physically dying.. and i probably
didn't spend $1000 for the whole batch of them

 

[Thomas N]

Trust me, I was really hoping for some kind of backup battery when this first started happening. I went through the manual and couldn't find any reference of one, then scoured the Internet and came up empty there as well. I know from working with computers that the CMOS batteries can eventually wear out and cause all types of intermittent issues. No such luck with the BK-7M.

I do think it's extraordinarily odd for it to fail after 10 years without any changes whatsoever. And the fact that it was a slow degradation over several weeks had me on the backup battery trail as well. It's not unprecedented, so hopefully this post will be able to be referenced should it happen to another one.

Thanks for the input all.

 

[Ventura]

i spent a little more time in a slow review of the take-apart video
and it seems like the RAM loads on every boot up from the flash
memory, which is used apparently instead of a battery held type of
memory. Then if that fails, it sounds like there is a section of
permanent code that defaults instead, but then takes you
back to that wizard thing were you make a choice

this flash memory also must be written to before the device shuts down
apparently, so there is maybe some caching involved too that can
get lost

the thing that strikes me is using Flash memory as a critical
component of the start process. A super nerd friend long ago
explained 2 me how that type of memory works how it caches and pages
and uses blocks differently than RAM / memory we have become accustomed to.
(and as opposed to being a miniature SSD built in)

so Flash can be notoriously unreliable over time.. some lasts but much of it
develops errors.. those bad spots get automatically isolated but over time
can also cause critical errors as things get complicated for the "housekeeper"
control software.. i guess there are some advantages to using the Flash instead
of the battery type that we have in pretty much all computers
(if have you ever booted into setup, the settings are saved but preserved by
that little battery, then thats what the computer boots from the next time)

and as you can see from the video, the device is essentially simply
a specialized computer.. one that shares components and even assemblies with
many other Roland devices, the main difference being what the Software
configures and uses for the specific product.

i was suprised to see them STILL using the Rogers Organ board all these
years later.. their entire line of Atalier Organs piggybacked off the Rogers too..
and they had floppy drives !
Roland sold/spun Rogers off quite awhile ago when it started to divest itself of
those properties (like Rhodes and Sequential Circiuits) which they had
acquired and absorbed tech from.

well i don't know how this is going to help or what kind of issues you
guys are gonna have as some of these Roland modules age out.. it
doesn't look like the flash memory can be R&R'd..

a different scheme but similar that KORG uses sometimes is to load the
OS and saved settings from an internal Hard drive. My Mp-10 Pro does that,
but even losing the Hard drive doesn't kill the beast, as it is easy to replace
and even reload. For mine, i just cloned the OEM drive with an SSD
and swapped it out (faster load time too) so i even have the original
as a spare and a dual drive cloning dock in my shop if needed.

 

[Rosie C]

so Flash can be notoriously unreliable over time.. some lasts but much of it
develops errors.. those bad spots get automatically isolated but over time
can also cause critical errors

My car's built-in satnav has this sort of arrangement. When I bought it I tried to update the maps but more than half of the flash drive is bad sectors now, so it has no maps now. The only option VW have is to replace the complete unit

 

[Airy]

Sorry for getting off-topic but some of us are the generation who has experienced the transit from acoustic instruments (which are mostly servicable a life long and you did invest in "for once in a lifetime") to electonic and nowadays digital instruments (and other daily-life gear) that is just not planned that way. I don't feel that this is neccesarily the rumored "planned obsolency" but at least accepted -- or not beeing aware of -- by us consumers.

 

[JerryPH]

I was sitting here thinking... did you try to spray a little good quality contact cleaner in to the MIDI sockets? Just reaching out further now to be sure there is no other solution other than sending it in. I don't think it would help, but its a low cost thing to try that if it does work, would be amazing.

 

[Chris Laarman]

Sorry for getting off-topic but some of us are the generation who has experienced the transit from acoustic instruments (which are mostly servicable a life long and you did invest in "for once in a lifetime") to electonic and nowadays digital instruments (and other daily-life gear) that is just not planned that way. I don't feel that this is neccesarily the rumored "planned obsolency" but at least accepted -- or not beeing aware of -- by us consumers.

One step more general, I think. Some of us are used to "everything contains no electronics", some more to "things may contain some discrete electronic parts", but now we're in the "everything contains highly integrated electronics". With matching servicability.

 

[Thomas N]

I was sitting here thinking... did you try to spray a little good quality contact cleaner in to the MIDI sockets? Just reaching out further now to be sure there is no other solution other than sending it in. I don't think it would help, but its a low cost thing to try that if it does work, would be amazing.

Hey Jerry, I didn't do that, and I do have contact cleaner for PC builds my son and I have completed. That's an excellent idea. I took some canned air and gave a couple squirts earlier but it does appear to only be an issue with MIDI since I can still use it as a drum machine and the styles play fine. I'll give it a try. I don't hold out much hope for it but it's an excellent suggestion. Thank you.

 

[davidplaysaccordion]

4x to bk7m. I used to get frame errors when using widi jacks to connect them. Once I turned off the clock sync and unnecessary stuff being sent from the 4x things stabilized. I assume it was dropping a byte in the stream from time to time. It never happens when using a cable though.

 

[Chris Laarman]

4x to bk7m. I used to get frame errors when using widi jacks to connect them. Once I turned off the clock sync and unnecessary stuff being sent from the 4x things stabilized. I assume it was dropping a byte in the stream from time to time. It never happens when using a cable though.

You reminded me of the Running Status in MIDI. (That link seemed attractive when I searched the Web.) So the receiving end would perceive the data transmitted as some incomplete shorthand.

(My thoughts are elsewhere, but this may be a clue. I hope so.)

 

[JerryPH]

4x to bk7m. I used to get frame errors when using widi jacks to connect them. Once I turned off the clock sync and unnecessary stuff being sent from the 4x things stabilized. I assume it was dropping a byte in the stream from time to time. It never happens when using a cable though.

I had that happen using a WIDIJACK but never using a WIDI master. Interesting... I may try that.
Want to tell me what other things you specifically disabled as well on the BK and 4X?

 

[davidplaysaccordion]

In the configuration that gave me the problems I had a widi bud pro in the accordion's usb input on the front and a Widi Master plugged into both midi ports on the BK7m. Had to do the bk like that because even though I only needed midi input, that is provided by the unpowered smaller connector of the WIDI master.
The meant that I TWO way midi communications going between the accordion and the BK7m. IE the accordion was sending it's midi to the bk7m which is what I wanted but the bk7m was also sending midi to the accordion.
From what I can recall I turned off active sensing tx in the accordion which helped but the main thing that helped was turning off the midi clock data coming from the bk7m. There is just too much info being transmitted when time code is included and I really don't think the WIDI devices can keep up.
Since then I have switched to using WIDI jacks. thr 4x only has one real midi connector that can be in or out so I set it to out and just have that plugged into the widi jack. On the BK7m I just have the midi input to the bk connected.
The asccordion powers the widi jack on that end via the midi out and on the bk side I power the widi jack there via a usb powerbank. I also power the bk7m from the powerbank using a usb step-up converter which works great so no mains power needed anywhere.

 

[Thomas N]

Following up on this one in case somebody searches for this problem.

The issue was not with the BK-7M. Sadly, it's with the FR-8X. I connected the BK-7M MIDI IN and MIDI OUT to two separate keyboards I have recently purchased and it works perfectly with both of them. All of the chord changes are recognized and all of the sounds and styles can be changed. There are no MIDI Errors whatsoever and the connection is 100% stable.

I guess I have to look into the FR-8X now.

 

[JerryPH]

I know that the 8X can be set up to send a ton of info related to the bellows, I cannot recall the settings and I am on a short break at work, but check in to that direction... it may help.