Korg Poly800/EX800 Users group photo

Yahoo Groups archive

Korg Poly800/EX800 Users

Archive for korgpolyex.

Index last updated: 2026-03-30 01:10 UTC

Thread

The latest news on HAWK

The latest news on HAWK

2016-05-25 by Michael Hawkins

I bought an STM32 evaluation board today. And a bunch of level shifters (needed to convert the 3.3 volt signals to 5 volts).

This is the small beginning of the new HAWK. I've talked about this before. But let me recap for those who missed it or are new to the group.

The goal is to replace both the 80C85 CPU and the MSM5232 tone generator in the Poly/EX-800 with an STM32 processor.

Then rewrite the code entirely so that we get:

1) Eight independent voices each with...
2) Independent low pass and resonant filters as close to NJM2069 as we can program.
3) Independent frequency modulators.
4) Multiple waveforms (not sure how many or even how they will be generated but square, saw, triangle, PCM, phase shifting, cross modulation etc are all under consideration).
5) MSM5232 emulation (exactly the same operation and characteristics that Poly already has in its TG).
6) LCD display option 16x2 back lit (while also keeping original display and keypanel).
7) keep all other Poly/Hawk functions.
8) add multiple input/output options for expansion.
9) An effects stage (and reduce noise).

As I've said before, I have so much other work going on that I really have no idea how long this is going to take. But I am at least in the early stages of actually doing it. Which is still better than not at all.

And the big question still remains, do I create a brand new drop in replacement main board (and move the NJM2069 over to it)?

/Mike

Re: The latest news on HAWK

2016-05-26 by ultragalore@...

I for one am very curious about this new setup Mike, I also have the PreenFM and setups like this make me very happy (open source project, maybe you want to look into that and get ideas; http://ixox.fr/preenfm2/

 

 ). 

I do *need* a piece of hardware which does the NJM2069 however ;-)

Re: The latest news on HAWK

2016-05-26 by bperkins211@...

Sounds like a cool project!

Seems it would take over so much of the Korg that it may just be best to make it's own module and transplant a NJM2069 into it.  And if you already are simulating VCF's on the STM, then would just one NJM be used at the final output as an extra filter to add in the mix?
Just seems the old non velocity keys on a Poly800 would be dead weight, as would the old chorus/amp board if you design FX in the code or use another uC for...  not much meat left on the bone when the new uC takes over.

Still it would be a cool project, I would love to help with the coding if I could..  To get the waveforms I suppose you would use premapped tables to reference, then convert to PWM output with a filter.

Tom the Electric Druid tried to explain how it all works with his little PIC vco chip.  His code would be something to look at to see how the tables work and how to "morph" waveforms from one to the next.

I am very green with C++/Arduino coding, but did have alot of fun making an arduino VC LFO for my NJM2069 external module project.
Help coding a project like this would be a nice challenge to sharpen coding skills.

/Blaine

Re: [korgpolyex] Re: The latest news on HAWK

2016-05-27 by Michael Hawkins

I would include the NJM2069 so that the entire originality of the Poly-800 would also be retained.

I don't mind using some pre-mapped tables in a sampler mode. But I would also prefer discreet oscillator wave forms that can be modulated, PWM'd etc. Just how much can be done will depend upon the CPU performance capacity. And coding skills and techniques.

To begin with, the most important goals are to simply replace the 80C85 CPU and MSM5232 with the STM32 and recreated all of the Poly functions. This in itself doesn't sound like much etc when you consider just how much more can be done with additional CPU performance and storage capacity. For a start, the envelope generators and all performance controls (except the keyboard itself) will be 16 or even 32 bit instead of 8. The DEGs will be able to be made much faster. And we can create many more modulation routings than are currently possible where the only restriction on doing more of them has been the CPU. Just making the existing TG with its locked clock into 8 fully independent oscillators will be a huge improvement too. The richness of tone where the detuning is fully independent should help alot. And, initially we can make the TG into an 8 stage instead of four stage harmonic block. That should also improve the character of the basic TG before we even consider using any wave samples.

I don't think the keyboard is something that bothers me that much because it's just too easy to grab a controller or other MIDI source to get velocity. But it might be nice to include a variable peddle input. Do people even use such a thing anymore? The challenge is to create a board that can be dropped into Poly or can be used to convert over to a 1RU unit. MK2 owners might end up being left out for now.

The Mk1/EX chorus board is definitely going away. The cables in and out of that board will just have to be integrated onto the new board.

I also want to include stereo panning effects as well.

Should be fun!

/Mike

From: "bperkins211@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: korgpolyex@yahoogroups.com
Sent: Thursday, May 26, 2016 7:14 PM
Subject: [korgpolyex] Re: The latest news on HAWK

 
Sounds like a cool project!

Seems it would take over so much of the Korg that it may just be best to make it's own module and transplant a NJM2069 into it.  And if you already are simulating VCF's on the STM, then would just one NJM be used at the final output as an extra filter to add in the mix?
Just seems the old non velocity keys on a Poly800 would be dead weight, as would the old chorus/amp board if you design FX in the code or use another uC for...  not much meat left on the bone when the new uC takes over.

Still it would be a cool project, I would love to help with the coding if I could..  To get the waveforms I suppose you would use premapped tables to reference, then convert to PWM output with a filter.

Tom the Electric Druid tried to explain how it all works with his little PIC vco chip.  His code would be something to look at to see how the tables work and how to "morph" waveforms from one to the next.

I am very green with C++/Arduino coding, but did have alot of fun making an arduino VC LFO for my NJM2069 external module project.
Help coding a project like this would be a nice challenge to sharpen coding skills.

/Blaine


Re: [korgpolyex] Re: The latest news on HAWK

2016-05-28 by imanjl@...

>>NJM2069 so that the entire originality of the Poly-800 would also be retained.

NJM2069 and MSM5232 both have influence in character of poly-800. It is better to do the emulation of 2069 in a form of standalone software that runs on PC, no need for a hardware platform here.

A good emulation of M5253 which is an ASIC chip is tricky and maybe impossible, because inside the chip all oscillators/functions are working in parallel while STM32 code runs sequentially. Also 5253 is not 100% digital, it seems there are some analog VCAs, EGs inside this chip. I'm not sure even there is 8 DACs on die for building waveforms (regarding the size and cost of the chip) maybe oki/korg used only SQR and Mixed SQRs in analog domain for building saws... 

The curve of the AMP envelopes, the color of VCA and also the phase management of oscillators are the points that must be considered.

Re: [korgpolyex] Re: The latest news on HAWK

2016-06-02 by domgoold@...

thought i'd mention: i'm messing around with MFB Synth Lite II and PolyLite
synths at the moment. much lamenting online that the oscillators are not
VCO, and not quite DCO - they are to be considered 'digital oscillators'...

point being: still an excellent (tiny) synth that does a great job of sounding
like a large analog poly. it's filter and VCA are analog.

the 1st edition elicited complaints, and he eventually revised the waveforms,
which are generated by a cpu (these were made in 2003 and again in 2006,
if that means much re: what was available then - and now). saw sounds
pretty cool now.

just saying that digital oscillator doesn't have to mean 'VA', and the thing can
still sound juicy and analog. korg poly saw isn't exactly (isn't exactly a saw)
a showstopper, so pretty much anything will sound good there :)

i read yesterday that the practical difference between a DCO and a 'digital osc'
is that the digital will always start with the same phase, and that analog,
including DCO, are free-running. don't know to what extent this is true.

the other issue with a digital osc used in unadulterated form is aliasing at
high pitches, and weird digi-artifacts (quite good when you find them on the
MFB) - you can generate a saw that looks good in the midrange but if you
want to avoid aliasing, you have to start limiting bandwidth as you go up
in pitch (nyquist etc) -

if you are thinking of using wavetables:are wavetables 'honest'(?)
*in a DCO synthesizer?* 8-)  (hey, who cares, more waveforms to mess
around with is good....could have squares and saws from a multitude of
machines.

the MFB has ringmodulation and sync - apparently these are-necessarily-
digital,  with whatever sound differences that implies (in practice, it's a
machine that makes loads of good sounds,  has loads of variation, so it
isn't an issue, for me,anyway) - so if you want to do crossmodulations of
oscs, will that also be 'digital'?(worth worrying about?)

and for the filter,..so you are suggesting you would reproduce the existing
filter, which is single across 6 voices, and offer a filter for each voice. like
a 'proper' poly. question: is there a big difference between 'global' and
individual filters?(if all are calibrated the same)
(what will be left of the original Poly ? ...!!!)

Re: [korgpolyex] Re: The latest news on HAWK

2016-06-02 by Michael Hawkins

i read yesterday that the practical difference between a DCO and a 'digital osc'
is that the digital will always start with the same phase, and that analog,
including DCO, are free-running. don't know to what extent this is true.

Well, that can be avoided by various methods that I can immediately think of.

the other issue with a digital osc used in unadulterated form is aliasing at
high pitches,

Well, that depends upon how fast the digital oscillators are. Too slow, not good.

and for the filter,..so you are suggesting you would reproduce the existing
filter, which is single across 8 voices, and offer a filter for each voice. like
a 'proper' poly. question: is there a big difference between 'global' and
individual filters?(if all are calibrated the same)
(what will be left of the original Poly ? ...!!!)

Yes, I think it's a big difference being that you get to hear consecutively played notes being affected by their own EG with their own filter. One filter limits the sound experience.I consider the single filter and the lack of independently variable waveforms to be Poly's major shortcomings on the sound generation side of things. HAWK gave Poly all of the MIDI control and extra modulations and stuff. But HAWK didn't do diddly squat for enhancing the sound experience of the Poly.

I figure, any new waveforms, modulations and filters are going to be a huge benefit to Poly. Yes, there isn't much left of Poly once we do this upgrade. But as with HAWK, if the external Poly looks exactly the same and the cost of doing of an upgrade is kept low enough, what's not to like?

/Mike


From: "domgoold@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: korgpolyex@yahoogroups.com
Sent: Thursday, June 2, 2016 1:32 PM
Subject: Re: [korgpolyex] Re: The latest news on HAWK

 
thought i'd mention: i'm messing around with MFB Synth Lite II and PolyLite
synths at the moment. much lamenting online that the oscillators are not
VCO, and not quite DCO - they are to be considered 'digital oscillators'...

point being: still an excellent (tiny) synth that does a great job of sounding
like a large analog poly. it's filter and VCA are analog.

the 1st edition elicited complaints, and he eventually revised the waveforms,
which are generated by a cpu (these were made in 2003 and again in 2006,
if that means much re: what was available then - and now). saw sounds
pretty cool now.

just saying that digital oscillator doesn't have to mean 'VA', and the thing can
still sound juicy and analog. korg poly saw isn't exactly (isn't exactly a saw)
a showstopper, so pretty much anything will sound good there :)

i read yesterday that the practical difference between a DCO and a 'digital osc'
is that the digital will always start with the same phase, and that analog,
including DCO, are free-running. don't know to what extent this is true.

the other issue with a digital osc used in unadulterated form is aliasing at
high pitches, and weird digi-artifacts (quite good when you find them on the
MFB) - you can generate a saw that looks good in the midrange but if you
want to avoid aliasing, you have to start limiting bandwidth as you go up
in pitch (nyquist etc) -

if you are thinking of using wavetables:are wavetables 'honest'(?)
*in a DCO synthesizer?* 8-)  (hey, who cares, more waveforms to mess
around with is good....could have squares and saws from a multitude of
machines.

the MFB has ringmodulation and sync - apparently these are-necessarily-
digital,  with whatever sound differences that implies (in practice, it's a
machine that makes loads of good sounds,  has loads of variation, so it
isn't an issue, for me,anyway) - so if you want to do crossmodulations of
oscs, will that also be 'digital'?(worth worrying about?)

and for the filter,..so you are suggesting you would reproduce the existing
filter, which is single across 6 voices, and offer a filter for each voice. like
a 'proper' poly. question: is there a big difference between 'global' and
individual filters?(if all are calibrated the same)
(what will be left of the original Poly ? ...!!!)


Re: [korgpolyex] Re: The latest news on HAWK

2016-06-02 by LARRY HAWKE

I like it!

From: korgpolyex@yahoogroups.com <korgpolyex@yahoogroups.com> on behalf of Michael Hawkins korgpolyex800@... [korgpolyex] <korgpolyex@yahoogroups.com>
Sent: Thursday, June 2, 2016 3:47:24 PM
To: korgpolyex@yahoogroups.com
Subject: Re: [korgpolyex] Re: The latest news on HAWK
 
 

i read yesterday that the practical difference between a DCO and a 'digital osc'
is that the digital will always start with the same phase, and that analog,
including DCO, are free-running. don't know to what extent this is true.

Well, that can be avoided by various methods that I can immediately think of.

the other issue with a digital osc used in unadulterated form is aliasing at
high pitches,

Well, that depends upon how fast the digital oscillators are. Too slow, not good.

and for the filter,..so you are suggesting you would reproduce the existing
filter, which is single across 8 voices, and offer a filter for each voice. like
a 'proper' poly. question: is there a big difference between 'global' and
individual filters?(if all are calibrated the same)
(what will be left of the original Poly ? ...!!!)

Yes, I think it's a big difference being that you get to hear consecutively played notes being affected by their own EG with their own filter. One filter limits the sound experience.I consider the single filter and the lack of independently variable waveforms to be Poly's major shortcomings on the sound generation side of things. HAWK gave Poly all of the MIDI control and extra modulations and stuff. But HAWK didn't do diddly squat for enhancing the sound experience of the Poly.

I figure, any new waveforms, modulations and filters are going to be a huge benefit to Poly. Yes, there isn't much left of Poly once we do this upgrade. But as with HAWK, if the external Poly looks exactly the same and the cost of doing of an upgrade is kept low enough, what's not to like?

/Mike


From: "domgoold@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: korgpolyex@yahoogroups.com
Sent: Thursday, June 2, 2016 1:32 PM
Subject: Re: [korgpolyex] Re: The latest news on HAWK

 
thought i'd mention: i'm messing around with MFB Synth Lite II and PolyLite
synths at the moment. much lamenting online that the oscillators are not
VCO, and not quite DCO - they are to be considered 'digital oscillators'...

point being: still an excellent (tiny) synth that does a great job of sounding
like a large analog poly. it's filter and VCA are analog.

the 1st edition elicited complaints, and he eventually revised the waveforms,
which are generated by a cpu (these were made in 2003 and again in 2006,
if that means much re: what was available then - and now). saw sounds
pretty cool now.

just saying that digital oscillator doesn't have to mean 'VA', and the thing can
still sound juicy and analog. korg poly saw isn't exactly (isn't exactly a saw)
a showstopper, so pretty much anything will sound good there :)

i read yesterday that the practical difference between a DCO and a 'digital osc'
is that the digital will always start with the same phase, and that analog,
including DCO, are free-running. don't know to what extent this is true.

the other issue with a digital osc used in unadulterated form is aliasing at
high pitches, and weird digi-artifacts (quite good when you find them on the
MFB) - you can generate a saw that looks good in the midrange but if you
want to avoid aliasing, you have to start limiting bandwidth as you go up
in pitch (nyquist etc) -

if you are thinking of using wavetables:are wavetables 'honest'(?)
*in a DCO synthesizer?* 8-)  (hey, who cares, more waveforms to mess
around with is good....could have squares and saws from a multitude of
machines.

the MFB has ringmodulation and sync - apparently these are-necessarily-
digital,  with whatever sound differences that implies (in practice, it's a
machine that makes loads of good sounds,  has loads of variation, so it
isn't an issue, for me,anyway) - so if you want to do crossmodulations of
oscs, will that also be 'digital'?(worth worrying about?)

and for the filter,..so you are suggesting you would reproduce the existing
filter, which is single across 6 voices, and offer a filter for each voice. like
a 'proper' poly. question: is there a big difference between 'global' and
individual filters?(if all are calibrated the same)
(what will be left of the original Poly ? ...!!!)


AtomaHawk - new hardware version in final testing

2016-06-05 by Michael Hawkins

The original AtomaHawk board used a Maxim DS1666+ 10K quasi audio taper digital potentiometer chip that Maxim stopped making and then the after market source dried up and they got way too expensive.

So, I went looking for a new through hole chip that could be used but alas I was completely unable to find one. What I did find was a far better device but it is only available in surface mount. I finally had to go through the process of figuring out how hard (or not) it is to create surface mount boards in small quantities.

The AtomaHawk version 1.4 is now in final testing. Here's a photo of the first board made.

Major improvement here is that the original DS1666 was only capable of handling 7 bits and also used a non linear range. The new device is 8 bits - the Analog Devices AS5290. Datasheet is attached for anyone that's interested.

I expect to be shipping the new version out in a few weeks.

/Mike

Re: [korgpolyex] AtomaHawk - new hardware version in final testing [2 Attachments]

2016-06-05 by Gordonjcp

On Sun, Jun 05, 2016 at 03:45:14PM +0000, Michael Hawkins korgpolyex800@... [korgpolyex] wrote:
> So, I went looking for a new through hole chip that could be used but alas I was completely unable to find one. What I did find was a far better device but it is only available in surface mount. I finally had to go through the process of figuring out how hard (or not) it is to create surface mount boards in small quantities.

... and you found it's actually way, way easier than through-hole?

--
Gordonjcp MM0YEQ

Re: [korgpolyex] AtomaHawk - new hardware version in final testing

2016-06-06 by Michael Hawkins

Ha ha! Yes, I was very pleasantly surprised at how easy it is. I am actually kicking myself for not having tried surface mount ten years ago. Stupid really. But I guess you get used to what you're doing and sometimes you don't want to change what works.

But in this case, going to surface mount has some challenges but they can all be overcome. As long as the board isn't too big or uses the really small SMD components.

The AD5290 is a TSSOp which is very small to handle with tweezers. It can be positioned but it is really difficult. So I am making a jig that will help me to place it. I shall let you know how that goes.

But as always Gordon, your questions are always a strong indication of your insight. So why the heck didn't you tell me to try SMD earlier? Eh?

/Mike


From: "Gordonjcp gordon@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: "Michael Hawkins korgpolyex800@... [korgpolyex]" <korgpolyex@yahoogroups.com>
Sent: Sunday, June 5, 2016 6:07 PM
Subject: Re: [korgpolyex] AtomaHawk - new hardware version in final testing

 
On Sun, Jun 05, 2016 at 03:45:14PM +0000, Michael Hawkins korgpolyex800@... [korgpolyex] wrote:
> So, I went looking for a new through hole chip that could be used but alas I was completely unable to find one. What I did find was a far better device but it is only available in surface mount. I finally had to go through the process of figuring out how hard (or not) it is to create surface mount boards in small quantities.

... and you found it's actually way, way easier than through-hole?

--
Gordonjcp MM0YEQ



Re: [korgpolyex] AtomaHawk - new hardware version in final testing

2016-06-06 by Wayne

The places I worked at that had surface mount components we didn't use tweezers most of the time. We had little suction cups with bulbs on them. You squeezed the bulb then put the cup on the component. Releasing the bulb then cause a vacuum that held the component. It was like attaching a handle to the chip. You can buy kits that have several sizes for different size chips. Much better than tweezers and less work than a jig.

Wayne


From: "Michael Hawkins korgpolyex800@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: "korgpolyex@yahoogroups.com" <korgpolyex@yahoogroups.com>
Sent: Sunday, June 5, 2016 8:50 PM
Subject: Re: [korgpolyex] AtomaHawk - new hardware version in final testing

 
Ha ha! Yes, I was very pleasantly surprised at how easy it is. I am actually kicking myself for not having tried surface mount ten years ago. Stupid really. But I guess you get used to what you're doing and sometimes you don't want to change what works.

But in this case, going to surface mount has some challenges but they can all be overcome. As long as the board isn't too big or uses the really small SMD components.

The AD5290 is a TSSOp which is very small to handle with tweezers. It can be positioned but it is really difficult. So I am making a jig that will help me to place it. I shall let you know how that goes.

But as always Gordon, your questions are always a strong indication of your insight. So why the heck didn't you tell me to try SMD earlier? Eh?

/Mike


From: "Gordonjcp gordon@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: "Michael Hawkins korgpolyex800@... [korgpolyex]" <korgpolyex@yahoogroups.com>
Sent: Sunday, June 5, 2016 6:07 PM
Subject: Re: [korgpolyex] AtomaHawk - new hardware version in final testing

 
On Sun, Jun 05, 2016 at 03:45:14PM +0000, Michael Hawkins korgpolyex800@... [korgpolyex] wrote:
> So, I went looking for a new through hole chip that could be used but alas I was completely unable to find one. What I did find was a far better device but it is only available in surface mount. I finally had to go through the process of figuring out how hard (or not) it is to create surface mount boards in small quantities.

... and you found it's actually way, way easier than through-hole?

--
Gordonjcp MM0YEQ





Re: [korgpolyex] AtomaHawk - new hardware version in final testing

2016-06-06 by Michael Hawkins

Thanks Wayne, coincidentally, I ordered a couple of those last night. I use a fixture jig for placing the board and orienting it to a kaptan stencil overlaid for applying the solder paste. I then use tweezers which is by far the hardest part (especially with the tiny AD5290). I then place on a hot plate, using a laser temperature sensor and visually watching for the solder to run. I take the board off using tweezers, I found that applying the solder paste in the right quantities and consistently across all solder pads is tricky. But that is something that gets better with practice. And most importantly, if it isn't perfect, wipe off the paste, clean the board and stencil, and do it over until you get it perfect.

And then the really hard part is placing really small SMD footprints such as MSOP. Tweezers make the job quite awkward and I realized that is because they don't help you to line up the device radially/rotational at all and they also amplify your hand natural shaking movements. So a suction tool and a arm/wrist bench/rest are critically important.

It's been a lot of fun going through the learning experience and it has now completely revolutionized my view of designing and building surface mounted designs. Especially with the STM32 upgrade in mind.

/Mike


From: "Wayne atlas1301962@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: "korgpolyex@yahoogroups.com" <korgpolyex@yahoogroups.com>
Sent: Sunday, June 5, 2016 10:09 PM
Subject: Re: [korgpolyex] AtomaHawk - new hardware version in final testing

 
The places I worked at that had surface mount components we didn't use tweezers most of the time. We had little suction cups with bulbs on them. You squeezed the bulb then put the cup on the component. Releasing the bulb then cause a vacuum that held the component. It was like attaching a handle to the chip. You can buy kits that have several sizes for different size chips. Much better than tweezers and less work than a jig.

Wayne


From: "Michael Hawkins korgpolyex800@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: "korgpolyex@yahoogroups.com" <korgpolyex@yahoogroups.com>
Sent: Sunday, June 5, 2016 8:50 PM
Subject: Re: [korgpolyex] AtomaHawk - new hardware version in final testing

 
Ha ha! Yes, I was very pleasantly surprised at how easy it is. I am actually kicking myself for not having tried surface mount ten years ago. Stupid really. But I guess you get used to what you're doing and sometimes you don't want to change what works.

But in this case, going to surface mount has some challenges but they can all be overcome. As long as the board isn't too big or uses the really small SMD components.

The AD5290 is a TSSOp which is very small to handle with tweezers. It can be positioned but it is really difficult. So I am making a jig that will help me to place it. I shall let you know how that goes.

But as always Gordon, your questions are always a strong indication of your insight. So why the heck didn't you tell me to try SMD earlier? Eh?

/Mike


From: "Gordonjcp gordon@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: "Michael Hawkins korgpolyex800@... [korgpolyex]" <korgpolyex@yahoogroups.com>
Sent: Sunday, June 5, 2016 6:07 PM
Subject: Re: [korgpolyex] AtomaHawk - new hardware version in final testing

 
On Sun, Jun 05, 2016 at 03:45:14PM +0000, Michael Hawkins korgpolyex800@... [korgpolyex] wrote:
> So, I went looking for a new through hole chip that could be used but alas I was completely unable to find one. What I did find was a far better device but it is only available in surface mount. I finally had to go through the process of figuring out how hard (or not) it is to create surface mount boards in small quantities.

... and you found it's actually way, way easier than through-hole?

--
Gordonjcp MM0YEQ







Re: [korgpolyex] AtomaHawk - new hardware version in final testing

2016-06-08 by Wayne

Amplifying you hand movement is the basic function of a lever. In this case it isn't desired. When you pick up something with tweezers you don't even notice the hand shaking. Until you look under a microscope at the pcb you are trying to populate. Then you look like you have Parkinson's or another muscle problem. Indecently the first place I worked with surface mount we made a implantable device that would stop the tremors from Parkinson's. With the suction device it is much better because the lever is shorter and you don't need as much finger flexion. This gives you less shaking transfer. Plus you can roll the bulb between your finger and thumb to rotate the chip. Not and easy thing to do with tweezers. Still you need to be calm and take your time. That is why the company I used to work for shelled out the one million dollars for a pick-place machine. It was amazing how fast it could populate boards. Even with the smallest resistors. Then the boards went right an oven on a conveyor.
When you put your boards on the hot plate do you have something overtop to reflect the heat back down? I think this would be better because the heat would surround the paste making melt faster. That way the chips aren't heated for as long.
Wayne


From: "Michael Hawkins korgpolyex800@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: "korgpolyex@yahoogroups.com" <korgpolyex@yahoogroups.com>
Sent: Monday, June 6, 2016 9:12 AM
Subject: Re: [korgpolyex] AtomaHawk - new hardware version in final testing

 
Thanks Wayne, coincidentally, I ordered a couple of those last night. I use a fixture jig for placing the board and orienting it to a kaptan stencil overlaid for applying the solder paste. I then use tweezers which is by far the hardest part (especially with the tiny AD5290). I then place on a hot plate, using a laser temperature sensor and visually watching for the solder to run. I take the board off using tweezers, I found that applying the solder paste in the right quantities and consistently across all solder pads is tricky. But that is something that gets better with practice. And most importantly, if it isn't perfect, wipe off the paste, clean the board and stencil, and do it over until you get it perfect.

And then the really hard part is placing really small SMD footprints such as MSOP. Tweezers make the job quite awkward and I realized that is because they don't help you to line up the device radially/rotational at all and they also amplify your hand natural shaking movements. So a suction tool and a arm/wrist bench/rest are critically important.

It's been a lot of fun going through the learning experience and it has now completely revolutionized my view of designing and building surface mounted designs. Especially with the STM32 upgrade in mind.

/Mike


From: "Wayne atlas1301962@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: "korgpolyex@yahoogroups.com" <korgpolyex@yahoogroups.com>
Sent: Sunday, June 5, 2016 10:09 PM
Subject: Re: [korgpolyex] AtomaHawk - new hardware version in final testing

 
The places I worked at that had surface mount components we didn't use tweezers most of the time. We had little suction cups with bulbs on them. You squeezed the bulb then put the cup on the component. Releasing the bulb then cause a vacuum that held the component. It was like attaching a handle to the chip. You can buy kits that have several sizes for different size chips. Much better than tweezers and less work than a jig.

Wayne


From: "Michael Hawkins korgpolyex800@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: "korgpolyex@yahoogroups.com" <korgpolyex@yahoogroups.com>
Sent: Sunday, June 5, 2016 8:50 PM
Subject: Re: [korgpolyex] AtomaHawk - new hardware version in final testing

 
Ha ha! Yes, I was very pleasantly surprised at how easy it is. I am actually kicking myself for not having tried surface mount ten years ago. Stupid really. But I guess you get used to what you're doing and sometimes you don't want to change what works.

But in this case, going to surface mount has some challenges but they can all be overcome. As long as the board isn't too big or uses the really small SMD components.

The AD5290 is a TSSOp which is very small to handle with tweezers. It can be positioned but it is really difficult. So I am making a jig that will help me to place it. I shall let you know how that goes.

But as always Gordon, your questions are always a strong indication of your insight. So why the heck didn't you tell me to try SMD earlier? Eh?

/Mike


From: "Gordonjcp gordon@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: "Michael Hawkins korgpolyex800@... [korgpolyex]" <korgpolyex@yahoogroups.com>
Sent: Sunday, June 5, 2016 6:07 PM
Subject: Re: [korgpolyex] AtomaHawk - new hardware version in final testing

 
On Sun, Jun 05, 2016 at 03:45:14PM +0000, Michael Hawkins korgpolyex800@... [korgpolyex] wrote:
> So, I went looking for a new through hole chip that could be used but alas I was completely unable to find one. What I did find was a far better device but it is only available in surface mount. I finally had to go through the process of figuring out how hard (or not) it is to create surface mount boards in small quantities.

... and you found it's actually way, way easier than through-hole?

--
Gordonjcp MM0YEQ









Re: [korgpolyex] AtomaHawk - new hardware version in final testing

2016-06-08 by Michael Hawkins

No, I don't use a reflector. The conduction from the hot plate to the board seems to me to be dramatically more than anything a reflector might do. Still, maybe I should look into and do some testing.
/Mike

From: "Wayne atlas1301962@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: "korgpolyex@yahoogroups.com" <korgpolyex@yahoogroups.com>
Sent: Tuesday, June 7, 2016 8:30 PM
Subject: Re: [korgpolyex] AtomaHawk - new hardware version in final testing

 
Amplifying you hand movement is the basic function of a lever. In this case it isn't desired. When you pick up something with tweezers you don't even notice the hand shaking. Until you look under a microscope at the pcb you are trying to populate. Then you look like you have Parkinson's or another muscle problem. Indecently the first place I worked with surface mount we made a implantable device that would stop the tremors from Parkinson's. With the suction device it is much better because the lever is shorter and you don't need as much finger flexion. This gives you less shaking transfer. Plus you can roll the bulb between your finger and thumb to rotate the chip. Not and easy thing to do with tweezers. Still you need to be calm and take your time. That is why the company I used to work for shelled out the one million dollars for a pick-place machine. It was amazing how fast it could populate boards. Even with the smallest resistors. Then the boards went right an oven on a conveyor.
When you put your boards on the hot plate do you have something overtop to reflect the heat back down? I think this would be better because the heat would surround the paste making melt faster. That way the chips aren't heated for as long.
Wayne


From: "Michael Hawkins korgpolyex800@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: "korgpolyex@yahoogroups.com" <korgpolyex@yahoogroups.com>
Sent: Monday, June 6, 2016 9:12 AM
Subject: Re: [korgpolyex] AtomaHawk - new hardware version in final testing

 
Thanks Wayne, coincidentally, I ordered a couple of those last night. I use a fixture jig for placing the board and orienting it to a kaptan stencil overlaid for applying the solder paste. I then use tweezers which is by far the hardest part (especially with the tiny AD5290). I then place on a hot plate, using a laser temperature sensor and visually watching for the solder to run. I take the board off using tweezers, I found that applying the solder paste in the right quantities and consistently across all solder pads is tricky. But that is something that gets better with practice. And most importantly, if it isn't perfect, wipe off the paste, clean the board and stencil, and do it over until you get it perfect.

And then the really hard part is placing really small SMD footprints such as MSOP. Tweezers make the job quite awkward and I realized that is because they don't help you to line up the device radially/rotational at all and they also amplify your hand natural shaking movements. So a suction tool and a arm/wrist bench/rest are critically important.

It's been a lot of fun going through the learning experience and it has now completely revolutionized my view of designing and building surface mounted designs. Especially with the STM32 upgrade in mind.

/Mike


From: "Wayne atlas1301962@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: "korgpolyex@yahoogroups.com" <korgpolyex@yahoogroups.com>
Sent: Sunday, June 5, 2016 10:09 PM
Subject: Re: [korgpolyex] AtomaHawk - new hardware version in final testing

 
The places I worked at that had surface mount components we didn't use tweezers most of the time. We had little suction cups with bulbs on them. You squeezed the bulb then put the cup on the component. Releasing the bulb then cause a vacuum that held the component. It was like attaching a handle to the chip. You can buy kits that have several sizes for different size chips. Much better than tweezers and less work than a jig.

Wayne


From: "Michael Hawkins korgpolyex800@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: "korgpolyex@yahoogroups.com" <korgpolyex@yahoogroups.com>
Sent: Sunday, June 5, 2016 8:50 PM
Subject: Re: [korgpolyex] AtomaHawk - new hardware version in final testing

 
Ha ha! Yes, I was very pleasantly surprised at how easy it is. I am actually kicking myself for not having tried surface mount ten years ago. Stupid really. But I guess you get used to what you're doing and sometimes you don't want to change what works.

But in this case, going to surface mount has some challenges but they can all be overcome. As long as the board isn't too big or uses the really small SMD components.

The AD5290 is a TSSOp which is very small to handle with tweezers. It can be positioned but it is really difficult. So I am making a jig that will help me to place it. I shall let you know how that goes.

But as always Gordon, your questions are always a strong indication of your insight. So why the heck didn't you tell me to try SMD earlier? Eh?

/Mike


From: "Gordonjcp gordon@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: "Michael Hawkins korgpolyex800@... [korgpolyex]" <korgpolyex@yahoogroups.com>
Sent: Sunday, June 5, 2016 6:07 PM
Subject: Re: [korgpolyex] AtomaHawk - new hardware version in final testing

 
On Sun, Jun 05, 2016 at 03:45:14PM +0000, Michael Hawkins korgpolyex800@... [korgpolyex] wrote:
> So, I went looking for a new through hole chip that could be used but alas I was completely unable to find one. What I did find was a far better device but it is only available in surface mount. I finally had to go through the process of figuring out how hard (or not) it is to create surface mount boards in small quantities.

... and you found it's actually way, way easier than through-hole?

--
Gordonjcp MM0YEQ











Re: [korgpolyex] AtomaHawk - new hardware version in final testing

2016-06-08 by Louis Botterill

SMD for the win. When I started working I started out designing circuits and a lot was becoming SMT. Tiny 0402 components and several hundred leg QFPs were all around. I used to really like it, although routing by hand was tricky for really dense boards. We had a pick and place machine the size of a house that used to rattle through placing parts on boards like it was possessed. Funny thing to watch when that went wrong, literally thousands of tiny components all stacked up in a big mess in a split second, all destined for the bin once off the feed rolls.

One thing I did note, when designing high freq rf circuits the SMT really helped with the circuit characteristics.

All good fun, now I don't do any of that but still have fond memories. Glad you gave it a go. Slightly different techniques for designing, building and re-working SMT boards but it's all largely the same. I was able to swap out a several hundred pin QFP sometimes just with a basic iron and not even a hot air gun. Super fiddly and have to be careful removing off the board without destroying, but possible all the same! 

Glad to hear the Ploy 800 and mods are not dead, would upgrades be compatible with the hawk/atomahawk already fitted?


On 6 June 2016 at 01:50, Michael Hawkins korgpolyex800@... [korgpolyex] <korgpolyex@yahoogroups.com> wrote:
 

Ha ha! Yes, I was very pleasantly surprised at how easy it is. I am actually kicking myself for not having tried surface mount ten years ago. Stupid really. But I guess you get used to what you're doing and sometimes you don't want to change what works.

But in this case, going to surface mount has some challenges but they can all be overcome. As long as the board isn't too big or uses the really small SMD components.

The AD5290 is a TSSOp which is very small to handle with tweezers. It can be positioned but it is really difficult. So I am making a jig that will help me to place it. I shall let you know how that goes.

But as always Gordon, your questions are always a strong indication of your insight. So why the heck didn't you tell me to try SMD earlier? Eh?

/Mike


From: "Gordonjcp gordon@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: "Michael Hawkins korgpolyex800@... [korgpolyex]" <korgpolyex@yahoogroups.com>
Sent: Sunday, June 5, 2016 6:07 PM
Subject: Re: [korgpolyex] AtomaHawk - new hardware version in final testing

 
On Sun, Jun 05, 2016 at 03:45:14PM +0000, Michael Hawkins korgpolyex800@... [korgpolyex] wrote:
> So, I went looking for a new through hole chip that could be used but alas I was completely unable to find one. What I did find was a far better device but it is only available in surface mount. I finally had to go through the process of figuring out how hard (or not) it is to create surface mount boards in small quantities.

... and you found it's actually way, way easier than through-hole?

--
Gordonjcp MM0YEQ






--
Web: www.chillipower.com
Blog: http://louisbotterill.blogspot.com/
Twitter: http://twitter.com/BinaryJunkie
LinkedIn: http://uk.linkedin.com/pub/louis-botterill/10/3b2/265

“The trouble with the world is that the stupid are cocksure and the intelligent are full of doubt.” –Bertrand Russell

Please consider your environmental responsibility before printing this e-mail

Re: [korgpolyex] AtomaHawk - new hardware version in final testing

2016-06-10 by Michael Hawkins

Q: Glad to hear the Ploy 800 and mods are not dead, would upgrades be compatible with the hawk/atomahawk already fitted?

A: Not sure about the AtomaHawk because I may redesign it entirely. Definitely not the HAWK because we would be replacing it entirely.

/Mike


From: "Louis Botterill chillipower.uk@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: korgpolyex@yahoogroups.com
Sent: Wednesday, June 8, 2016 2:42 PM
Subject: Re: [korgpolyex] AtomaHawk - new hardware version in final testing

 
SMD for the win. When I started working I started out designing circuits and a lot was becoming SMT. Tiny 0402 components and several hundred leg QFPs were all around. I used to really like it, although routing by hand was tricky for really dense boards. We had a pick and place machine the size of a house that used to rattle through placing parts on boards like it was possessed. Funny thing to watch when that went wrong, literally thousands of tiny components all stacked up in a big mess in a split second, all destined for the bin once off the feed rolls.

One thing I did note, when designing high freq rf circuits the SMT really helped with the circuit characteristics.

All good fun, now I don't do any of that but still have fond memories. Glad you gave it a go. Slightly different techniques for designing, building and re-working SMT boards but it's all largely the same. I was able to swap out a several hundred pin QFP sometimes just with a basic iron and not even a hot air gun. Super fiddly and have to be careful removing off the board without destroying, but possible all the same! 

Glad to hear the Ploy 800 and mods are not dead, would upgrades be compatible with the hawk/atomahawk already fitted?


On 6 June 2016 at 01:50, Michael Hawkins korgpolyex800@... [korgpolyex] <korgpolyex@yahoogroups.com> wrote:
 
Ha ha! Yes, I was very pleasantly surprised at how easy it is. I am actually kicking myself for not having tried surface mount ten years ago. Stupid really. But I guess you get used to what you're doing and sometimes you don't want to change what works.

But in this case, going to surface mount has some challenges but they can all be overcome. As long as the board isn't too big or uses the really small SMD components.

The AD5290 is a TSSOp which is very small to handle with tweezers. It can be positioned but it is really difficult. So I am making a jig that will help me to place it. I shall let you know how that goes.

But as always Gordon, your questions are always a strong indication of your insight. So why the heck didn't you tell me to try SMD earlier? Eh?

/Mike


From: "Gordonjcp gordon@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: "Michael Hawkins korgpolyex800@... [korgpolyex]" <korgpolyex@yahoogroups.com>
Sent: Sunday, June 5, 2016 6:07 PM
Subject: Re: [korgpolyex] AtomaHawk - new hardware version in final testing

 
On Sun, Jun 05, 2016 at 03:45:14PM +0000, Michael Hawkins korgpolyex800@... [korgpolyex] wrote:
> So, I went looking for a new through hole chip that could be used but alas I was completely unable to find one. What I did find was a far better device but it is only available in surface mount. I finally had to go through the process of figuring out how hard (or not) it is to create surface mount boards in small quantities.

... and you found it's actually way, way easier than through-hole?

--
Gordonjcp MM0YEQ






--
Web: www.chillipower.com
Blog: http://louisbotterill.blogspot.com/
Twitter: http://twitter.com/BinaryJunkie
LinkedIn: http://uk.linkedin.com/pub/louis-botterill/10/3b2/265

“The trouble with the world is that the stupid are cocksure and the intelligent are full of doubt.” –Bertrand Russell

Please consider your environmental responsibility before printing this e-mail


Re: The latest news on HAWK

2016-11-02 by bperkins211@...

Any developments?

I've been getting back into messing with arduino and looking to get a Teensy after seeing how easy it is to make a Teensy synth using the Teensy Audio Library.

Re: [korgpolyex] Re: The latest news on HAWK

2016-11-02 by Michael Hawkins

Well, I've been learning how to populate SMD's on the new AtomaHawk. And I'll be shipping out the first batch to customers very soon.

Other than that, not a lot going on right now because of 'real' work. One has to provide for the family.

But I did buy a bunch of STM32 development boards.

/Mike


From: "bperkins211@... [korgpolyex]" <korgpolyex@yahoogroups.com>
To: korgpolyex@yahoogroups.com
Sent: Tuesday, November 1, 2016 10:24 PM
Subject: [korgpolyex] Re: The latest news on HAWK

 
Any developments?

I've been getting back into messing with arduino and looking to get a Teensy after seeing how easy it is to make a Teensy synth using the Teensy Audio Library.


Re: [korgpolyex] Re: The latest news on HAWK

2016-11-03 by bperkins211@...

well having 'real' work is always good ;)  it pays the bills.


I thought I might share with you what I found recently with the Teensy project.  I will most likely invest into this little guy soon...

Teensy Audio Library, high quality sound processing in Arduino sketches on Teensy 3.1


Teensy Audio Library, high quality sound processing in ...
PJRC Store Teensy 3.6, $29.25 Teensy 3.5, $24.25 Teensy 3.2, $19.80 Teensy LC, $11.65 Teensy ...
View on www.pjr...
Preview by Yahoo

 


They have developed a very nice setup you may want to check into.. helps most the hard work is already done.

Here someone built a nice little synth with it.

Teensy Polyphonic Synth

Teensy Polyphonic Synth
My first venture into the teensy audio library! This is a polyphonic synthesizer with various controls. ...
View on www.yo...
Preview by Yahoo

 



The teensy is not a STM32.  Think it is a Cortex M4



as for the STM32, did you ever run across this board?


ST STM32VL-Discovery ARM Development Tool; Value Line USB Demo Board Kit MCU USA | eBay
ST STM32VL-Discovery ARM Development Tool; Value Li...
The STM32 Value line Discovery is the cheapest and quickest way to discover the STM32. It includes eve...
View on www.eba...
Preview by Yahoo

 



evidently it has a debug feature built into it.. not sure but I'm guessing it will allow you to peek inside it and run code line by line to see what it is or is not doing..



/Blaine