Dating GK3 pickup, engineering information.

Started by Eight_Stringer, October 21, 2011, 12:26:36 AM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

Eight_Stringer

After disassembling a GK3 and noting the manufacturer dated parts on the PCB assembly and comparing to the product code on the bottom of the control module, best estimate is

ZS04253 

   Would be the 04,   after the "ZS" is the year of manufacture,  the year starting from 2000 ie 2004 for this module, the next digit could be the month perhaps or the unit is 253rd made in 2004.
Most of the date stamping on the internal parts ( of ZS04253 ) was very late 2003, including the PCB substrate.

   The GK3 uses NJM2068 dual low noise op amps for signal conditioning,  a voltage regulator on top of the DC output volume control pot. Some output rail clamping and output current limiting,  HF gain reduction in the negative feedback loop of the individual pickup signal conditioners, end to end gain at about 12,  rolling off at higher frequencies.

   Going to produce a variant of the signal conditioning pcb for use in the myriad of basses i have here. Need just 4 individual pickups on some of the basses,  the clever bit will be how to sort the magnetic pickup assembly,  canvass ideas here and share information as required.

HTH

gumtown

I have always wondered if compact cassette tape heads would work as seperate pickups.
Free "GR-55 FloorBoard" editor software from https://sourceforge.net/projects/grfloorboard/

Eight_Stringer

Quote from: gumtown on October 21, 2011, 12:38:52 AM
I have always wondered if compact cassette tape heads would work as seperate pickups.

  Chance of course, easy enough to test out with a lash up.  Supply quantity may be an issue.

Have a range of Neodymium magets to hand here, smallest is 10m/m long by 4m/m diameter ( round ).
Centre locate on the magnet body, a dual winding,  and small "end" pole plates towards the string. Mean the pole faces would be 10-12m/m along the length of the string. Not a problem in my installation as i will not be installing usual pickups ie just the GK style pickups, i have this already on two guitars and one bass.  Hope i make sense. Thank you for the reply gumtown, Regards.

gumbo

...as on the other (GK pickup) thread, I would also like to see where you're going on this one..

I have a separate project going alongside the other weird stuff that could perhaps benefit from this kind of technology...up 'til now I've been planning to just wreck another GK3 system to get what I want...  hmm...

No (real) panic, but perhaps talk to you further about this in the future.    :)

Kind regards, and +1 for all your work!
Read slower!!!   ....I'm typing as fast as I can...

Eight_Stringer

Quote from: gumbo on October 21, 2011, 04:16:18 AM
...as on the other (GK pickup) thread, I would also like to see where you're going on this one..

I have a separate project going alongside the other weird stuff that could perhaps benefit from this kind of technology...up 'til now I've been planning to just wreck another GK3 system to get what I want...  hmm...

No (real) panic, but perhaps talk to you further about this in the future.    :)

Kind regards, and +1 for all your work!

  Ok Gumbo,  appreciate the positive comments. There are some clever people lurking here, know the information provided has been very helpful from them.  See what i can contribute, most important will it be useful? Hence canvassing of opinions and input.  Just do a partial schematic of the GK3,  being a guide for repair rather than a factory drawing 100% complete.  The plan here is to design and fabricate a generic signal processing PCB assembly. Something that can be implemented on purpose basses and guitars ie no other pickup system/controls installed.  Be a straight forward process for me, have many years in design and prototype fabrication in electronics.  Tempting to sling a micro controller onto the pcb for trick control, am able to code in assembler,  yes i know archaic, yet the most efficient executing code ( read fast ).  Will be an Surface Mount Technology board as am geared for that these days,  many reasons, not least component availability is better for SMD's.

   The GK series pickups are good value even at $200,  just i have 20 basses and new GK's are a noticeable cost in converting even part of the collection. Purchasing used GK's has been ok mostly,  yet paying $150 for a damaged pickup gives great impetus to "roll my own".  The only real section that requires much thought, time and effort will be the actual magnetic pickup assembly under the strings.  Though i feel confident given 40 years in electromechanical and electronic fields. Have full workshop here, lathe, tig welding, etc etc.

  Would like ( and encourage ) others to join in,  to produce a GK compatible diy item that can be implemented on custom instruments or even the more usual retro fit to production instruments. The later is better met by original GK series pickups, fabricating housing and mounting systems is a lot of work in mechanical and aesthetic successful outcomes.  Have the concept of the signal conditioning pcb sub assembly in my mind already. Say about 6 weeks for a working prototype to be produced,  give or take what is not time fatigued chemicals and photo sensitive laminates on hand.

  Thank you for reading this post. Regards to all. 

Elantric

#5
Thanks for posting the pics

Here is the GK-2A schematics
More on this thread
https://www.vguitarforums.com/smf/index.php?topic=73.0



I understand the GK-3 is very close to the GK-2A circuit above, but uses lower noise components, and voltage over-swing power rail diodes on the audio signals. 

The GK-3 adds a new circuit which sends a DC Control voltage on Pin#9 to indicate the status of the 3 position "Guitar / Mix / Synth (GK) " switch using a resistor voltage divider.


GK-3  pin no. 9 assignment:
Name: "Sel"

Function:select signal
GK-3 Pin  9 Specifications:


GK-3 (Pin#9) Voltage Range: (GK  = 1.56V / Mix =  1.04V / Guitar =  0.52V
GK-3B (Pin#9) Voltage Range: (GK  = 4.48V / Mix = 3.96V / Bass = 3.44V

other conditions are prohibited



Prior to 2011, anybody with a credit card could call Roland US and order a Service Manual with full schematics.

But effective January, 2011, they no longer offer that service.

And this thread is worth reading too:
https://www.vguitarforums.com/smf/index.php?topic=1082.msg5501#msg5501

Eight_Stringer

#6
Once agin thank you Elantric,  links viewed already. Be very useful if anyone that obtained the factory schemtic drawing for the GK3 prior to the unavailability, could post if for ready access.  Save me some time,  though have mostly completed the hand drawing already.  Measuring the smd capies in circuit is proving elusive so far,  though will persevere and obtain same for reference. Power up the GK3 and measure the LDO reg output voltage on top of the volume pot today. Done the measurement for the LDO, confident it also drives the resnet to provide the GMS pin 9 control voltages

  Extra picture of the rig here for background on what i use in reference to this forum.

Regards.

Elantric

#7
QuoteMeasuring the smd capies in circuit is proving elusive so far,

I use an Agilent LCR meter to measure SMD components in circuit

Here is a refurb one for $159

http://www.newark.com/jsp/search/productdetail.jsp?SKU=63T4339&CMP=AFC-GB100000001



If you need help, I can use this LCR meter to measure my own GK-3 for the capacitor values you have a problem measuring.

Eight_Stringer

 Thank you Elantric for the information about LCR meters,  very interesting.  Being old school ( mortifying as that is :-) when HP was not Agilent  ), with a few  digital readout meters here ( that measure capies ),  also various Japanese and English Wheatstone Bridges,  nobody knew how to drive the old gear so i won them.  How i measured the pickup inductance value,  using the old wheatstone bridge gear.  Try the older gear or some of the old techniques ( reactance ), if i have no luck will seek your kind offer, thank you.   The only capy that is important for value is the negative feedback capy around the pickup op amp. Can do a frequency roll off measurement and set my design to roll of the same slope, worst case. Would only take 5 minutes on the prototype.

   The onboard LDO 5volt reg does indeed drive the three step voltage divider resnet,  which is output on pin 9 of the din13 connector. Complete that section of the schematic today.

   Regards.

Elantric

#9
Good work

I too "cringe" at what became of HP, and often forget Agilent is the new name for the old company.


Might contact BillBax for a GK-3 schematic
https://www.vguitarforums.com/smf/index.php?action=profile;u=86

https://www.vguitarforums.com/smf/index.php?topic=2915.msg26416#msg26416

He has a web site here and he is a member here too.
http://billbax.110mb.com/index.html

http://billbax.110mb.com/gk%20studio%20extreme.html

Eight_Stringer

#10
Ok Elantric,  all noted in your comprehensive reply above.  Pending any official schematic forthcoming,  have completed the hand drawing for the GK3 here.  Dependent on what schematic capture program i use ( how easy to generate www friendly drawing ),  in a week or two will have a formalised schematic for posting here.  All circuit components are accounted for on the GK3 pcb assembly, and the hand drawing looks usable to/for me in regard to further prototype work.  The drawing from me will not not be annotated in regard to exact component identifiers as per silk screen,  it will be generic in listing major component id only, which are not repeated ie wont list every component for every pickup amplifier section. Similar to the GK2 drawing in this regard.

  Like to keep the prototype similar to the OEM GK 2 and 3 series pickups,  my finger memory is pretty used to the current location and spacing compared to the volume control and mode switch.   Last thing i need is to memorise a new control map, :-).

  Just quickly,  have done a mockup of string pickup assembly concept. The magnet is 10m/m long and 4m/m diameter. Either make a former for 32SWG winding wire and slip the magnet inside with flat plates addressing the string direction. This wire is too thick, however does indicate the concept of contra or self cancelling winding,  been a while on exact humbucker mechanical details,  correct me if i am mistaken at this quick stab at a design idea and concept. 

   Regards.

billbax

#11
Hi Eight_Stringer,

Just throwing in a few ideas.

1. A metal case for your preamp is an essential design principal, and will help keep noise to a minimum.

2.  OPAMPS.  Start off with OPA2134.  I've used it in all of my GK Studio Extreme designs and found it to give a better performance than any other tried. Believe me, I've been all round the world with op-amp choices.  Here's a few I've tried:  NE5532, OPA2604, OP275, LM4562, AD8066, AD825.  None of them of them will give the same pickup sensor noise/EMI immunity that OPA2134 offers.

3. GK3B and GK3 are very good hex pickups and a excellent place to start.  Once you've got your pre-amp going and roughly settled on component values, you can then experiment with pickup choices. Also ground the casing housing the pickup/sensor.  That way you'll be adding 1-2 dB to your signal-to-noise ratio.

4.  BATTERY POWER SUPPLY.  Using a battery supply will remove AC mains ripple from your precious audio.  You will increase your signal-to-noise by about 2-3dB.

5. PREAMP GAIN. Don't be afraid to experiment with gain on each of your opamps.  For instance, the output of top E on a guitar is a good deal less than the output on string G.  Roland chose 10k across all opamp feedback resistors, therefore not taking into account top E and its lower output.  (If you intend to reverse the orientation of your pickup, then consider this in your resistor value choice).

OPA2134.  Excellent opamp with very good sensor noise immunity.





GK STUDIO EXTREME 2010.  The latest GK preamp with separate string outputs.  Dynamic-range -105dB. THD 0.00459%.





GK STUDIO EXTREME 2009.  Dynamic-Range -100dB.  THD 0.00459%





GK3B EXTREME BASS.  A hard-wired OFC Breakout Cable and connectors of choice.  Dynamic-Range -95dB.  THD 0.005%.  This design is under review.





Almost forgot, you can easily get the GK3/GK3B service manual notes from Roland spares. They've always been pretty cool about this sort of thing. Failing that, drop me an email.

All the best,

Bill

www.separate-strings.co.uk

http://www.youtube.com/user/BaxBreakoutBox


Elantric

#12
+1 on the OPA2134 opamp choice. Very nice part.

QuoteAlmost forgot, you can easily get the GK3/GK3B service manual notes from Roland spares.

This might be true in the UK, but here in the USA, Roland US clamped down on all such activity, starting in Jan. 2011 -
Here in the states, we are now in the dark ages. Today there is a more free flowing culture of sharing electronic ideas in Hungary. 

http://elektrotanya.com


Bill - if you have a source to order a Roland GR-55 Service Manual, share it, or buy one for me and I will pay you!


Eight_Stringer

#13
Hello Billbax, Elantric and other helpful folk,

   Well what a surprise! Nice work Billbax,  know only too well,  just how much effort is required to produce your extensive designs and impressive products.  Had an uneasy feeling working away these last few days, i was doing what had already been done or trail blazed before me.  Must say you beat me to nearly every idea and notion in the back of my mind,  big laugh.   Your very correct in the concept of, the overall performance of any equipment is the sum of the individual specifications and item performance, every little bit helps, or detracts.

   One aspect i am acutely aware off, is not repeating was already been done.  Had i known about the availability ( and i should have done my homework better ) for your ultimate devices, then the impetus for what i am doing would be redirected to new fields.   Being low volume production your able to utilise prime specification parts,  unlike mass manufacture that dictates price consideration in the final design.  Have worked in the sound reinforcement industry many years ago,  including high end studio equipment. Installation and repair allowed me to see first hand the performance of included parts in many designs.  Burr Brown were an engima in the audio industry,  producing silicon of unusual and superior performance when engineered into a design correctly.  Tell you nothing new there. Some of the best performing "op amps" have seen were discrete transistors at the front end,  sure you have seen the same.  No rule that a purpose op amp ic will be the ultimate performer,  matter of best performance over package and price.  Your ( and myself ) do not have that cost/ease of production imperative for mass production.  Can well understand the designs you produce,  matter of performance, looks second and price last. Now there is the real world of the true fringe dweller, big smile.

  One of my better skills concerns shielding and the mis quoted "earthing" systems,  pre EMI days like we have now.  You could well imagine the problems in a TV broadcast station, and the implementation of new "digital" products into the analog world of broadcast TV,  many techs fell by the wayside.  Nothing like an Eddystone diecast box is there for a real mans approach to EMI reduction and susceptibility,  if not outright mass.

  Agree,  if i understand you correctly,  that the GK hex pickup assembly is very good.  Be hard to better the price and  performance for this sub assembly. Say even at $50 "workshop" rates you would be hard pressed to make one in 4 hours,  at $200 for the whole internal or external kit the pickup sub-assembly is worth that alone in my view.

Well i have much to think about now,  given that there are already choices like your products Billbax.

  Attach a picture of the development bass for this project.  Became quickly obvious that magnetic hex pickup output signal was governed in the main, by string height to pole face of the sensor.  Lateral movement along the string length much less influence, though possibly "tone" may be a factor there.  That elusive non technical "sound factor" is never far away from laying waste to even the best paper designs.

  Do appreciate the people efforts in making this forum available and for the contributors making known hard won knowledge for nothing more than goodwill, cheers to all.

warmingtone

Quote from: gumtown on October 21, 2011, 12:38:52 AM
I have always wondered if compact cassette tape heads would work as seperate pickups.

Hi there, first post here...my fellow antipodean Gumbo alerted me to this thread and interesting proposition.

As it happens, I have been doing quiet a bit of research into this lately for my own project and unique guitar and wanted to explore some of the possibilities of hex systems (though not within my skill set to go all midi with it, nor my needs)...

I have written a separate thread bogging my ideas and findings and how things progress as things move along if people here are interested....

http://guitarnuts2.proboards.com/index.cgi?board=wiring&action=display&thread=6014

It is just a part of a bigger project that I am working on which may also be of interest though I dare say once you go midi, multi-tuning bridge and such might not be necessary  ;)

I'm working as a kind of 'open source' on the project, so input is appreciated and acknowledgement if taken further for my input would be nice.

...

As far as tape heads go...a cheap supply of such things is available these days off of ebay...Car Cassette Adaptors...$1 per tape head  ;D

The aperture to read the string on the ones I got are only 2mm...it needs to be very close to the bridge like any roland and similar..and very close to the strings. Magnetics are needed to magnetize the string over the head and be separate from other strings...but all doable. On a bass, one does not have the restrictions of string spread, nor height that you usually find on a guitar, perhaps one could source bigger and better heads. I noticed the adaptor in my car has two read heads that could be wired as one and the reading aperture is wider...so there may be better options for a bass (though, I apid about $10 for that).

The things only came in today, which you can read about in that thread that goes through the whole proposal...and though busy, could not resist some very crude tests.

The impedance is slightly under 300 ohms...so low impedance and would need a preamp...but enough that it should produce a sound and conveniently pre wired to a 1/8" jack plug. Using a adapter to 1/4" I plugged the directly into a conventional guitar amp. Holding teh tape head directly over the string and very close...plus over the magnet of any of the pickup coils...I was surprised at how good it sounded (though it might be a bit harmonically rich closer to the bridge).

You would want, or at least for midi conversion, I imagine to filter things down to the fundamental, probably limit and compress things and all that kind of thing, probaly a large part of what is in the roland circuitry and I understand needed even for something like the mono sonus audio/midi thing and similar audio to midi stuff (like software solutions that are improving all the time).

I'm not sure how this instrument falls into this forum, but I thought I would drop by and contribute where I can. The guitar I am building is a true acoustic. electric hybrid with piezo and mag pups, this hex thing, on board tuner and eq and a multi-tuning system...so quite a radical guitar, but disguised and working like a fairly traditional sort of "jazz-Box"....everything is fairly stealthy.

The HEx thing though has tremendous possibilities as I am sure you guys would all appreciate. For me, it is vastly simpler...basically a hex system that will have a separate out and allow through simple switching and a mix control, to run the signal of those strings selected in mono to a digital effect/modelling box or other processing. There will of course be no latency issues of ADC and such, just some simple pre-amping. The intention is that since this instrument is designed to excel at solo guitar arrangements, one could say double selected lower strings by pitch shifting down an octave and doubling those strings subtly to give more range down in the bass (should I choose). Alternatively, one might want to double the higher strings as selected an octave above for say a mandolin/12 string effect...or a sitar sim...ir delay only the high strings...etc...

Obviously the potential is limitless even in this incarnation.

Personally, I don't have the skills to go all midi with the concept, but that is an obvious potential. I am disinclined to make six outputs with all that this entails (six preamps, etc) though perhaps I might be tempted with a future version if all goes well and I make a lot of use of the thing. I would love to have a midi guitar to do things like 'transcriptins' or even to inut midi data (better at guitar than keyboards) but would need a heap of guidance in that department.

At least for this project, I am going the cheaper more reliable and low tech solution more in keeping to the needs of this thing...I don't want the thing to be swamped in technology but appear to be an "acoustic' guitar would you believe LOL  ;) But I would be intrigued to see and hear how people might take this simple and cheap solution and go in any direction that suits their needs...

...

The idea is not completely original btw...The first roland guitar synths GR500 used this technology...



But they did not have the digital power we have now and have made their own devices, very similar really, from these early stages. That does not mean that there is not a heap of potential in resurrecting a good idea to todays specs though!

Here are some car adaptor pics...



A lot more and continuing if interested wiill be posted on the link above to the GN2 forum and you would be welcome there if you'd like to comment, have ideas or questions...specs and such are already there as well as another thread on my current project if interested that this is being developed for....

nice to meet you all, and high Gumbo  ;) you get around LOL!

Eight_Stringer

Hello All,   So much information here from Bill and recently,  warmingtone, the later with some novel approaches to effective pickup assemblies,  there is dedication for you :-).  Been carefully going over the information and communications with other contributors here. Also making progress on the GK3 schematic,  few more days work there.

  Apart from the good contributions here,  the GK3 pickup and signal conditioner kit is very good value for money. Hard to better the well proven part with own design of same outcome. However if a custom design is required then experimentation and information from the process is invaluable,  thank you Bill and warmingtone.

  More as time permits,  regards.

warmingtone

Thanks 8-S

I've started to follow these threads with interest. Did some more experimentation with the tape head idea and coming with problems, though for a midi signal, perhaps these could produce a decent enough thing...but for my application of a decent audio signal it is coming up short.

Synth devices are of interest to me though for what I want to do, and the midi input is a big bonus for doing things like transcriptions and playing in parts as a midi controller perhaps. I agree, that in general one can get a roland GK pickup unit cheaply enough to make the DIY route less of an attractive proposition and may well end up doing this for the added benefits. However, there is a lot of on costs in cabling and the interface or synth itself that is required, a bit that can go wrong I expect...still...

The treads showing the guts of these things, the pin outs and such are of great benefit to many I am sure who are exploring this avenue and alternatives that might exist. Not sure the 'tape heads' are teh way to go, but there are other options yet to be explored...has anyone tried using hall effect devices for instance?

mapperboy

So hey Billbax, Eight Stringer and other GK/VG experimenters!
I recently acquired a GR-55 (with GK-3a) and being both a guitar, bass and synth player as well as DYI and electronic experimenter for about 50 years now got me interested in rolling my own break-out boxes and miscellaneous interface circuits for the GK magnetic and RMC piezo hex pickup systems.
Does anyone (perhaps from across the Pond or EUR) know of posts, links or PDFs showing a Roland Serv Man or schematic for the GK-3a?
If so could they please post!! I'd rather not (yet) open up my example of the GK-3a to rev eng. but will if I have to.
Also very interested in any info about what the variances are (if any) between the GK-3a and GK-3b (bass model) systems.
-Is it just the pickup itself or does is there a circuit board in the clamp-on interface.

Regards and Many Thanks in Advance...
-m


"No such thing as spare time.
No such thing as free time.
No such thing as down time.
All you got is life time. Go!"
- Henry Rollins

gumtown

As far as i can tell the GK-3 and GK-3B circuit board inside the 'wart' are identical, only the pickup is different (both use the same circuit board mini plug on the P.U.).
There is most likely not much difference in the circuit diagram of the GK-2A (as shown a few posts below) either, apart from the spare pin/cable core being used, and some components being SMD. They all work the same at a basic level, six preamp buffers, a voltage divider for the GK volume, two push buttons, the GK-3 adds a voltage divider to the GK selector switch so the conneted GR/VG unit knows which selection is made.
Free "GR-55 FloorBoard" editor software from https://sourceforge.net/projects/grfloorboard/

Moja

Just noticed this GK-2A schematic. It shows a 8.2K resistor on the input. I am confused. My GK-2A boards have 820 ohm resistors.
   Can anyone verify this ?


Quote from: Elantric on October 21, 2011, 02:55:28 PM
Thanks for posting the pics

Here is the GK-2A schematics
More on this thread
https://www.vguitarforums.com/smf/index.php?topic=73.0



I understand the GK-3 is very close to the GK-2A circuit above, but uses lower noise components, and voltage over-swing power rail diodes on the audio signals. 

The GK-3 adds a new circuit which sends a DC Control voltage on Pin#9 to indicate the status of the 3 position "Guitar / Mix / Synth (GK) " switch using a resistor voltage divider.


GK-3  pin no. 9 assignment:
Name: "Sel"

Function:select signal
GK-3 Pin  9 Specifications:


GK-3 (Pin#9) Voltage Range: (GK  = 1.56V / Mix =  1.04V / Guitar =  0.52V
GK-3B (Pin#9) Voltage Range: (GK  = 4.48V / Mix = 3.96V / Bass = 3.44V

other conditions are prohibited



Prior to 2011, anybody with a credit card could call Roland US and order a Service Manual with full schematics.

But effective January, 2011, they no longer offer that service.

And this thread is worth reading too:
https://www.vguitarforums.com/smf/index.php?topic=1082.msg5501#msg5501

Elantric

#20
8.2K is confirmed on GK-2A




You might be misinterpreting the method of labeling 1% SMT resistors?
http://en.wikibooks.org/wiki/Electronics/Component_Identification

jpdesroc

Quote from: gumtown on October 21, 2011, 12:38:52 AM
I have always wondered if compact cassette tape heads would work as seperate pickups.

This is what the Roland GR500 guitar used as hex pickup (6 tape heads alligned and potted in a pickup housing)

admin

#22
Quote from: Moja on December 07, 2014, 01:27:58 PM
Just noticed this GK-2A schematic. It shows a 8.2K resistor on the input. I am confused. My GK-2A boards have 820 ohm resistors.
   Can anyone verify this ?

You have a GC-1

it uses OEM version of the internal GK kit , built for Fender  - they resemble the older GK-2A internal kit boards - but are loaded with the same preamp gain structure as the GK-3


https://www.vguitarforums.com/smf/index.php?topic=5173.msg107659#msg107659