Juno 106 problem
Ken Dean
this about 4 months ago (same problem, different person). If I remember
correctly they said the oscillator chip for the 6th voice should be
replaced. If anybody has more info on this problem and knows where I can get
this chip I would appreciate the help.
Thanks.
Stephen Hall
(Got a warranty though, but think I'll keep it and get it fixed - cost
me A$150 = US$90)
Mine also has stuck notes. With some patches, there is a sound all the
time before I hit a key.
My local synth repair shop said its a really common problem that seems
to be age related. He said the parts are still made by roland - its a
card for each voice. They are approx US$60 each. I did read somewhere
that the individual chips can be replaced if you are handy with a
soldering iron, but I'm not that confident.
--
*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*
Consciousness: That annoying time between naps.
----------------------------------------------------
Stephen Hall
----------------------------------------------------
Kurt Lochner
>
> In poly1 mode every sixth note doesn't sound. Somebody posted a problem like
> this about 4 months ago (same problem, different person). If I remember
> correctly they said the oscillator chip for the 6th voice should be
> replaced. If anybody has more info on this problem and knows where I can get
> this chip I would appreciate the help.
> Thanks.
I've fixed one of those, and it was an odd chip that I'd have to
go look up inside of my GR-707 to be certain...
Kurt Lochner
>
> I just bought a juno 106 last weekend with the same problem.
>
> (Got a warranty though, but think I'll keep it and get it fixed - cost
> me A$150 = US$90)
>
> Mine also has stuck notes. With some patches, there is a sound all the
> time before I hit a key.
>
> My local synth repair shop said its a really common problem that seems
> to be age related. He said the parts are still made by roland - its a
> card for each voice. They are approx US$60 each. I did read somewhere
> that the individual chips can be replaced if you are handy with a
> soldering iron, but I'm not that confident.
I have some experience with these, and yes, it's a bear to extract.
The chip in question is an Intel 8250, I *think* it was a programmable
counter used as an oscillator. I'll be checking up on that later..
--As in today..
Wes Taggart
Kurt Lochner
>
> Beware of this advice.
Why'se that?
--You think you know something about it?
We Are As1
<< Why'se that?
--You think you know something about it? >>
He does know something about it. I'm not sure why he didn't say more, but
here's the score:
Those 8253 counter chips don't go bad. Period. In almost 20 years of working
on synthesizers professionally, I have never seen one fail or had to replace
one, and they are used in a LOT of synths. Even if they did fail, you would
very likely lose three of the voices, not just one, because one 8253 chip
creates the basic oscillation frequency for three oscillators (there are two
8253's in a Juno 106)
The typical failure scenario in a Juno 106 has one of the custom Roland hybrid
IC's going bad. There are two types in there: One of them has the filter and
VCA and the other type is a "waveshaping" chip that converts the square waves
coming from the 8253's. This waveshaping IC, in conjunction with the 8253
counter chip, creates what we commonly refer to as the "DCO". The filter/VCA
IC is called a something-1007A (sorry, I don't have my service manuals handy),
and the Juno has six of them onboard. The waveshaping IC is a 5534-something
or-other, and the Juno has three of them (each IC handles two voices). The two
types fail about equally as often.
BTW, just because the waveshaper IC handles two voices does not necessarily
mean that a bad one will always take out two voices. Because of the hybrid
topology of the part, one half of it can fail and the other half can keep on
working just fine (sometimes both sides fail at once, though).
This does not necessarily mean that one of these two types of chips has
definitely failed in this particular Juno, but it happens so often, and the
symptoms decribed sound typical, so it's very likely that this is what's going
on.
Nevertheless, it's not something that an unqualified person should be trying to
repair. It's probably best to bite the bullet and spend a few dollars to have
a real synth tech fix it. Believe me, I have seen lots of damaged circuit
boards brought in by people who didn't know how to safely desolder IC's from
circuit boards - and it winds up costing them much more to have someone not
only repair the synth, but also repair *their* damage.
How do you know if you're unqualified? If you don't ALREADY know how to
determine which of the aforementioned nine hybrid IC's is the bad one, you're
unqualified. If you've never successfully desoldered a large-scale IC (24 pins
or larger) from a PC board, leaving ZERO damage to the circuit traces, you're
unqualified. If you don't ALREADY own a solder-sucker or vacuum desolderer,
you're unqualified.
Michael Bacich
sys...@webtv.net
Wes Taggart
Kurt Lochner
>
> << Why'se that?
>
> --You think you know something about it? >>
>
> He does know something about it. I'm not sure why he didn't say more, but
> here's the score:
>
> Those 8253 counter chips don't go bad. Period.
You are incorrect, as I've been doing this for +20 years also, and
that's
the chip I traced lack of signal to, unsoldered and replaced.
The unit still functions and is used for background effects at the
local planetarium.
> In almost 20 years of working
> on synthesizers professionally, I have never seen one fail or had to replace
> one, and they are used in a LOT of synths. Even if they did fail, you would
> very likely lose three of the voices, not just one, because one 8253 chip
> creates the basic oscillation frequency for three oscillators (there are two
> 8253's in a Juno 106)
Your sure of that? I didn't need the manuals..
> The typical failure scenario in a Juno 106 has one of the custom Roland hybrid
> IC's going bad. [...]
I'm currently trying to restore an MT-100 with a bad power supply.
Got any clues on that one?
> BTW, just because the waveshaper IC handles two voices does not necessarily
> mean that a bad one will always take out two voices. Because of the hybrid
> topology of the part, one half of it can fail and the other half can keep on
> working just fine (sometimes both sides fail at once, though).
What power suplly inputs go to this chip(s), if you don't mind...
> Nevertheless, it's not something that an unqualified person should be trying to
> repair. It's probably best to bite the bullet and spend a few dollars to have
> a real synth tech fix it. Believe me, I have seen lots of damaged circuit
> boards brought in by people who didn't know how to safely desolder IC's from
> circuit boards - and it winds up costing them much more to have someone not
> only repair the synth, but also repair *their* damage.
Been there, didn't like it.
> How do you know if you're unqualified? If you don't ALREADY know how to
> determine which of the aforementioned nine hybrid IC's is the bad one, you're
> unqualified.
Use a 'scope..
> If you've never successfully desoldered a large-scale IC (24 pins
> or larger) from a PC board, leaving ZERO damage to the circuit traces, you're
> unqualified.
Been doing it for years..
> If you don't ALREADY own a solder-sucker or vacuum desolderer,
> you're unqualified.
I've worn out three already.
--SO much for that steam-calliope music...
We Are As1
<< You are incorrect, as I've been doing this for +20 years also, and
that's the chip I traced lack of signal to, unsoldered and replaced. >>
I stand by my earlier comment that 8253's don't fail. Of course there's always
a possibility that *any* component can fail. It's just very unlikely (for the
reasons already elaborated) that the problem on this person's Juno 106 is due
to a bad 8253 chip. I'm sure that someday I'll run across one, just as you
did, but I'm not gonna hold my breath until that day.
Only an idiot would ignore the possibility that no signal was coming from the
8253, and keep insisting that the problem was due to a bad filter IC. But it
would take two seconds with a scope to answer that question....and the answer
is invariably: signal is coming from the 8253 just fine. It's good that you
were clever enough to find the bad 8253 at the planetarium, but don't let your
odd, isolated experience cloud the issue or confuse those readers who have much
less experience, and desperately need real, honest, no-BS information.
Speaking of experience, Wes Taggart is one of the most highly respected and
expert technicians for analog synthesizer equipment in North America, and has
been for many, many years. He is known worldwide. I have been a
factory-authorized technician for just about every major keyboard manufacturer
since 1982 (Moog, Roland, Yamaha, Korg, Oberheim, Sequential, Emu, Ensoniq,
Kurzweil, Kawai, etc.). My clients have included Frank Zappa, Qunicy Jones,
Yes, Earth Wind & Fire, Howard Jones, Danny Elfman, Madonna, Chick Corea, and
countless other demanding professionals. When you mention that you've "been
doing this for +20 years also", presumably to qualify your expertise, shall we
assume those are the kind of experience and credentials you are talking about?
Then you went on to ask:
<< I'm currently trying to restore an MT-100 with a bad power supply.
Got any clues on that one? >>
You're kidding, right?
Kurt Lochner
>
> Kurt wrote:
>
> > You are incorrect, as I've been doing this for +20 years also, and
> > that's the chip I traced lack of signal to, unsoldered and replaced.
>
> I stand by my earlier comment that 8253's don't fail.
Well, that *was* my take until I replaced one that was faulty...
> Of course there's always
> a possibility that *any* component can fail.
I chalked this one up to Murphy..
> It's just very unlikely (for the reasons already elaborated) that the
> problem on this person's Juno 106 is due to a bad 8253 chip.
Well, it sounded like the same problem to my, hence my synopsis,,
> I'm sure that someday I'll run across one, just as you
> did, but I'm not gonna hold my breath until that day.
Understood, I could quite figure out why *that* chip died, it didn't
make
'sense' but I copped one from a buddy that was spare, and he'd seen such
a failure in only a couple of instances..
> Only an idiot would ignore the possibility that no signal was coming from the
> 8253, and keep insisting that the problem was due to a bad filter IC. But it
> would take two seconds with a scope to answer that question....
Yup, that's about how long it took to find..
> and the answer
> is invariably: signal is coming from the 8253 just fine. It's good that you
> were clever enough to find the bad 8253 at the planetarium, but don't let your
> odd, isolated experience cloud the issue or confuse those readers who have much
> less experience, and desperately need real, honest, no-BS information.
Understood, I've been lurking here for a while, but infrequently...
> Speaking of experience, Wes Taggart is one of the most highly respected and
> expert technicians for analog synthesizer equipment in North America, and has
> been for many, many years. He is known worldwide.
I've never heard of him..
> I have been a
> factory-authorized technician for just about every major keyboard manufacturer
> since 1982 (Moog, Roland, Yamaha, Korg, Oberheim, Sequential, Emu, Ensoniq,
> Kurzweil, Kawai, etc.). [...]
I didn't get to go to the factory training, even though I would've paid
for
it myself. We don't get 'celebrity status' here in Oklahoma just 'cause
you
can fix something as ornery as Moog, Korg, Roland, etc...
> When you mention that you've "been
> doing this for +20 years also", presumably to qualify your expertise, shall we
> assume those are the kind of experience and credentials you are talking about?
I'm qualified in AF/RF/digital and hold a BcSci in physics, now working
on
a Masters in Medical physics. Recent research project is in medical
imaging,
particularly ultrasound imaging conversions to hard data..
> Then you went on to ask:
>
> << I'm currently trying to restore an MT-100 with a bad power supply.
> Got any clues on that one? >>
>
> You're kidding, right?
Nope. Looks like an 8-volt line is dead, bias to the pass xsistor is
pinched-off..
--It's for a friend, which is about all I'll mess with now..
Kurt Lochner
>
> This is more like it.
microprocessor grade foundry work or just plain-jane analog..
> We Are As1 wrote:
>
> > Regarding Wes Taggart's comments, Cap...@fakeaddress.com wrote:
The 'fake address' address should be rudimentary to de-munge..
--Otherwise, I likely don't need to hear about it..
MRAmolli
I thought you might be interested to know there is help for all of us out there
who experience this problem. Apparently, there is a VCF chip that will fix the
problem and according to this engineer from Motorola you can get it from Roland
in Swansea.
Good luck and let me know the results.