Hacking The Teenage Engineering PO-80
Moderators: piaptk, tragwag, Steve E., Aussie0zborn
Re: Hacking The Teenage Engineering PO-80
are there any models or instructions ? thanks
Re: Hacking The Teenage Engineering PO-80
Hi,
I'll try to put together a package. Give me a few days.
Mark
I'll try to put together a package. Give me a few days.
Mark
Re: Hacking The Teenage Engineering PO-80
I have one question for you Mark, would you have such a prepared analog signal with an extended stereo base and processed in DSP, without an elliptical equalizer because I have it, the signal of which I could introduce to my amplifiers powering the head? I would just download it to check the capabilities of my test heads Is it worth entering these digital systems at all because they are not cheap?
Re: Hacking The Teenage Engineering PO-80
Hi,
Actually, getting into DSP is very inexpensive these days. Check out this device from Sure Electronics.
https://store.sure-electronics.com/product/800
This allow programing via Analog Devices Sigma Studio. This software is free. The ADAU1701 is not super powerful by today's standards, but you would be surprised what you can do with it. The graphical environment let's you develop applications without code. Just drag and drop blocks. The M/S processing you want to do is trivial to build in Sigma. It will take some learning curve to get up and running, but if you dive in, you will have a great tool.
Mark
Actually, getting into DSP is very inexpensive these days. Check out this device from Sure Electronics.
https://store.sure-electronics.com/product/800
This allow programing via Analog Devices Sigma Studio. This software is free. The ADAU1701 is not super powerful by today's standards, but you would be surprised what you can do with it. The graphical environment let's you develop applications without code. Just drag and drop blocks. The M/S processing you want to do is trivial to build in Sigma. It will take some learning curve to get up and running, but if you dive in, you will have a great tool.
Mark
Re: Hacking The Teenage Engineering PO-80
thanks Mark, I will consider this device.
I'm listening to your authentic recording again, Mark, and I can't believe that such a simple PO80 device does such a great job!
I'm listening to your authentic recording again, Mark, and I can't believe that such a simple PO80 device does such a great job!
Re: Hacking The Teenage Engineering PO-80
hi Mark! how its going with motor ?
Re: Hacking The Teenage Engineering PO-80
Hi,
I haven't forgotten, just too many things going on right now. I will get to it soon.
Mark
I haven't forgotten, just too many things going on right now. I will get to it soon.
Mark
Re: Hacking The Teenage Engineering PO-80
Happy Holidays,
I finally had some time to package up the details of my NEMA 14 stepper retrofit for the PO-80/Gakken. This is not for the novice as there are several things that will need to be fabricated and purchased. I put everything into a zip file which is attached. My setup uses a custom designed PCB that replaces the on board electronics including the ceramic pickup preamp. I have included the full set of Gerber files if you want to have some boards made. I used JCLPCB to get mine made. It is not that hard to make a breadboard version of the driver rather than fab a PCB. I have included the full schematic so that you can do this. The design is based on an Arduino Nano and an off the shelf TMC2209 stepper motor driver. A bill of materials is included in the package. This BOM only lists the parts needed to drive the stepper. The preamp is optional and you can create a BOM from the schematic. The old motor and mount is removed and replaced by a shock mounted NEMA 14 motor. The original belt is re-used and a 3d resin printed pully is attached to the NEMA 14 motor shaft. The shock mount for the motor is cast using Smooth-On silicone rubber via an FDM 3D printed mold. I used the OOOMO 30 series silicone IIRC. This is super glued to an FDM printed motor mount and the completed assembly is fastened to the corner of the PO-80 where the old motor was located via 2 machine screws that have holes drilled into the top of the PO-80 chassis. The driver allows for 33,45,78 and half speed modes. However, I have found that the half speed modes are problematic due to stepper resonances that fall right into the rpm range required. The shock mounting works pretty well, but everything is in tight quarters and you would have to go to great lengths to externally mount the motor and isolate the vibration and/or move the resonance out of range. Open for experimentation. The software allows for control of the motor phase current without the need for pot adjustment on the 2209 driver as I have implemented the serial communications mode of the driver. I direct write to the timer hardware to send stepper pulses, so there is no time jitter to affect speed. I have also implemented an optional LED strobe in software (requires an external LED driver board to be fabricated) and there is a Lightburn file to make one. There is also a Lightburn file for a front panel. If you try to get this running, I will try to answer any questions, but I don't want to turn this into a full time job. If you are not good at fabricating, I would stay away. Here are some pictures of the motor mount.
I finally had some time to package up the details of my NEMA 14 stepper retrofit for the PO-80/Gakken. This is not for the novice as there are several things that will need to be fabricated and purchased. I put everything into a zip file which is attached. My setup uses a custom designed PCB that replaces the on board electronics including the ceramic pickup preamp. I have included the full set of Gerber files if you want to have some boards made. I used JCLPCB to get mine made. It is not that hard to make a breadboard version of the driver rather than fab a PCB. I have included the full schematic so that you can do this. The design is based on an Arduino Nano and an off the shelf TMC2209 stepper motor driver. A bill of materials is included in the package. This BOM only lists the parts needed to drive the stepper. The preamp is optional and you can create a BOM from the schematic. The old motor and mount is removed and replaced by a shock mounted NEMA 14 motor. The original belt is re-used and a 3d resin printed pully is attached to the NEMA 14 motor shaft. The shock mount for the motor is cast using Smooth-On silicone rubber via an FDM 3D printed mold. I used the OOOMO 30 series silicone IIRC. This is super glued to an FDM printed motor mount and the completed assembly is fastened to the corner of the PO-80 where the old motor was located via 2 machine screws that have holes drilled into the top of the PO-80 chassis. The driver allows for 33,45,78 and half speed modes. However, I have found that the half speed modes are problematic due to stepper resonances that fall right into the rpm range required. The shock mounting works pretty well, but everything is in tight quarters and you would have to go to great lengths to externally mount the motor and isolate the vibration and/or move the resonance out of range. Open for experimentation. The software allows for control of the motor phase current without the need for pot adjustment on the 2209 driver as I have implemented the serial communications mode of the driver. I direct write to the timer hardware to send stepper pulses, so there is no time jitter to affect speed. I have also implemented an optional LED strobe in software (requires an external LED driver board to be fabricated) and there is a Lightburn file to make one. There is also a Lightburn file for a front panel. If you try to get this running, I will try to answer any questions, but I don't want to turn this into a full time job. If you are not good at fabricating, I would stay away. Here are some pictures of the motor mount.
You do not have the required permissions to view the files attached to this post.
Re: Hacking The Teenage Engineering PO-80
If the minimum current goes to the motor, it should work quite quietly, the mechanical resistance is very small, and in the case of a compact disk, the drive is also a stepper motor and it works beautifully, the simplicity is brilliant. Thanks
Re: Hacking The Teenage Engineering PO-80
That’s the wow and flutter sorted…not the cheapest way but the fastest.
You do not have the required permissions to view the files attached to this post.
Re: Hacking The Teenage Engineering PO-80
Takes care of real-time cut monitoring and cutting 78 too
Thats one way great way to solve it thats for sure. Re: Hacking The Teenage Engineering PO-80
Yes the stock motor is rubbish for cutting. I found the best place for the supplied blanks, in the rubbish. Just swapping out the cutting stylus for a conically sharpened HSS 1.5mm drill bit and embossing on CDs with some WD40 instantly improved the sound too because the force is lower so wow and flutter reduced. I’m experimenting with a piezo attached to the head as a feedback loop to phase invert and re-combine with the signal being cut…interesting.
Re: Hacking The Teenage Engineering PO-80
Hi,
I'll be interested to see how that works out. A couple of things to look at.
The first task would be to prevent the EM field of the driver from coupling into the piezo. They are very high impedance devices and will need to be mounted close to the stylus to measure the movement. This could be an issue at the higher frequencies since the excursion of the stylus is small and the driver current is high. If you get too much EM coupling, you will not be measuring the stylus motion and if you close the loop, it won't really be controlling the head. One way to test this is to mount the piezo at the location you intend, but beak the physical coupling to the stylus. Now run some noise or do some sweeps and compare the open loop response of the uncoupled vs. coupled response and make sure that the coupled response it at least 10db above the uncoupled over the frequency range of interest. If you meet that spec, then you are probably good to go.
An other issue is that I believe the piezo element will be measuring excursion (once you get above the low frequency cutoff of the piezo due to its capacitance into your preamp input impedance), not velocity like the typical moving coil feedback element. This means you can't just sum the inverted feedback with the drive signal and expect the closed loop to be stable. That's because the open loop response of a typical moving coil head has flat excursion response up to the system resonance and then falls off at -12/db oct (180 degrees of phase shift). On the other hand, the velocity response of a moving coil head rises at 6db/oct and then falls at 6db/oct above resonance and so is inherently stable if you invert and sum (assuming no other response anomalies). So you would need some feed forward loop compensation or would need to differentiate the output of the piezo to convert it from a position sensor to a velocity sensor. You will also need to look at the open loop gain and phase response prior to closing the loop to see if there are other pitfalls like secondary resonances that will cause stability problems.
Finally, as was pointed out, the response of the piezo element needs to be taken into account. If it has too many wonky gain/phase issues, you will be fighting them.
Hope this all makes sense.
It should be fun to experiment and even if it fails, you will certainly learn something. Keep us posted on your progress.
Mark
I'll be interested to see how that works out. A couple of things to look at.
The first task would be to prevent the EM field of the driver from coupling into the piezo. They are very high impedance devices and will need to be mounted close to the stylus to measure the movement. This could be an issue at the higher frequencies since the excursion of the stylus is small and the driver current is high. If you get too much EM coupling, you will not be measuring the stylus motion and if you close the loop, it won't really be controlling the head. One way to test this is to mount the piezo at the location you intend, but beak the physical coupling to the stylus. Now run some noise or do some sweeps and compare the open loop response of the uncoupled vs. coupled response and make sure that the coupled response it at least 10db above the uncoupled over the frequency range of interest. If you meet that spec, then you are probably good to go.
An other issue is that I believe the piezo element will be measuring excursion (once you get above the low frequency cutoff of the piezo due to its capacitance into your preamp input impedance), not velocity like the typical moving coil feedback element. This means you can't just sum the inverted feedback with the drive signal and expect the closed loop to be stable. That's because the open loop response of a typical moving coil head has flat excursion response up to the system resonance and then falls off at -12/db oct (180 degrees of phase shift). On the other hand, the velocity response of a moving coil head rises at 6db/oct and then falls at 6db/oct above resonance and so is inherently stable if you invert and sum (assuming no other response anomalies). So you would need some feed forward loop compensation or would need to differentiate the output of the piezo to convert it from a position sensor to a velocity sensor. You will also need to look at the open loop gain and phase response prior to closing the loop to see if there are other pitfalls like secondary resonances that will cause stability problems.
Finally, as was pointed out, the response of the piezo element needs to be taken into account. If it has too many wonky gain/phase issues, you will be fighting them.
Hope this all makes sense.
It should be fun to experiment and even if it fails, you will certainly learn something. Keep us posted on your progress.
Mark
Re: Hacking The Teenage Engineering PO-80
Thanks Mark. Yes I can see some hair pulling ahead but even if I just use the piezo as a means of measuring the vibration level, it would be better than guessing every time I’m embossing. I can see when the needle is blunt because it starts to bounce off the surface rather than scratch and the waveform is totally different. I’m also able to plug the piezo out into my drum module and trigger some interesting rhythms along to the music in real-time. Quite fun
Re: Hacking The Teenage Engineering PO-80
Also thanks for your incredible contributions here. It’s the most amazing tech resource I’ve ever discovered. Thank you
Re: Hacking The Teenage Engineering PO-80
Not checking first, I made a 7” adaptor. I’ll share the files if anyone’s interested. I integrated the clip so you can quickly switch between 5” and 7” by sliding the head mount off. Recommend Markrob’s modified cutting head so you can change the cutter angle and length.
I’ve also found that cheap vinyl Tungsten cutting bits already have a conical end which can be sharpened to use as embossers. You’ll need an extension sleeve (use a long split pin for example). I just drilled a 1.5mm hole alongside the existing Gakken mount and installed the embosser there.
Also, I made a 3D printable 42 degree guide block to help with embosser sharpening. It allows you to use knife sharpening blocks (cheap on Amazon) and finally 3000 grit paper etc to get a nice 84 degree cone as per the original Presto spec. Just using a dremel with the stylus in the chuck mounted vertically and slide the guide block into place.
I’ve also found that cheap vinyl Tungsten cutting bits already have a conical end which can be sharpened to use as embossers. You’ll need an extension sleeve (use a long split pin for example). I just drilled a 1.5mm hole alongside the existing Gakken mount and installed the embosser there.
Also, I made a 3D printable 42 degree guide block to help with embosser sharpening. It allows you to use knife sharpening blocks (cheap on Amazon) and finally 3000 grit paper etc to get a nice 84 degree cone as per the original Presto spec. Just using a dremel with the stylus in the chuck mounted vertically and slide the guide block into place.
You do not have the required permissions to view the files attached to this post.
Re: Hacking The Teenage Engineering PO-80
7"/5" multi adaptor, CD to 7.3mm adaptor, 84 degree grinding guide, 7" drive adaptor for rack and pinion button
You do not have the required permissions to view the files attached to this post.
Re: Hacking The Teenage Engineering PO-80
Wootwoot finally my gakken is shipping next week!!
Re: Hacking The Teenage Engineering PO-80
Once mine arrives should I skip trying to record with the original parts and jump straight into modifying or should I try a few before modifying it?