To build 3 synthesizers.To Build a Synthesizer...
The red faced unit with a green board is a Tsynth kit from ElectroTechnique. It is a "virtual analog" with the usual knobs for oscillators, amp, and filter but all the sound is generated digitally in a Teensy 4.1 module and output through a stereo DAC board. I built it about two years and it works good. I took it down from its perch in my modular mess cabinet to take the pictures and upgrade it's firmware. I noticed that two newer versions have been posted on their GitHub since I built it.
The black board comes with a black panel. It is called Progue from JSI and it a clone synth, part Moog Prodigy, part Moog Rogue. It took me several months to populate about 500 mostly SMD parts on the board. I started this build during the great semiconductor shortage and it has never worked. I got some of the chips from Ali Express and at least one of the LDO regulators is bad or counterfeit since the DC voltages are wrong. Since then I have acquired some genuine TI chips so I will simply swap them all out and hope for the best.
The shiny silver "rats nest" is the back side of the MFOS Ultimate panel. I installed the parts that I have yesterday and started soldering wires to them. More parts will be arriving in the next few days. The "jack field" at the top left must be populated one row at a time then wired up since it's too crowded for my shaky hands to solder the jacks without melting them. This will take a while to finish, then there is the Expander panel.
Attachments
You dont have an Xray machine, like at work? ;') Look inside, see a couple of smd's soldered onto the pins wireframe, that they encapsulated over.
We had an X-ray machine, as well as the ability to decap and analyze just about any chip made. At one time we had a fancy electron microscope that had the ability to "see" approximate voltage levels in the top few metal layers on a powered up die. The fancy equipment actually belonged to the semiconductor group, but needed to be installed in a place where there were no issues with vibration, earth tremors and anything that could interfere with putting an electron beam on an IC die with nanometer precision. The million dollar scope was first installed at a Motorola facility somewhere out west but got moved to South Florida in the early 1980's for reasons I never knew.
Sometime in 1983 or 1984 I got to fix that scope. Some big shots and some engineers from the Semiconductor Division out in Arizona or Texas (don't remember which) had come to Florida in search of answers to an unknown process issue with low Beta in some transistors that they had been reliably making for several years. When they fired it up it refused to finish the power up sequence. They came into the Calibration Lab looking for help but nobody there would rip into a million dollar scope and the closest service tech from the manufacturer was in Boston at least two days away. The lab boss told all the suits that "the evening shift lab tech can fix anything, and he gets here around 3:45 PM." I worked the evening shift in the Cal Lab for a reason, it was so I could be out in the Atlantic Ocean on my Hobie Cat all morning long, then slide into work by 4 PM. So, I showed up in my usual shorts, flip-flops and a T-shirt or tank top expecting my usual night of fixing factory stuff and playing with all the idle toys in the factory. Instead, I was met by half a dozen "suits" who had a justifiable concern with letting the "surfer dude" play with some seriously expensive stuff. They gave me the 45 minute lecture about high vacuum equipment and the ease of damaging a diffusion pump, but the night shift crew worked a little differently than the structured day time environment. There were several different "factories" each with their own maintenance teams, but we all knew each other and often learned what others did. We had high vacuum stuff in three different "factories" since we made quartz crystals and thin film microelectronics. I convinced the suits that I wouldn't destroy their equipment and went on to make my rounds through the two way radio factory. I had to do that before anything else.
Once I got the covers off the back of the scope's rack there was an obvious blown electrolytic cap. I dug up a suitable cap, swapped it in and proceeded to go through the power up sequence. I got a friend from the microelectronics group to go through the hi-vac stuff. Within about an hour I had the "suit's" transistor die up on the CRT screen so I paged them (it was the 80's, no cell phones yet). They arrived in a few minutes since the Mirage Disco Lounge was right across the street from the plant (not a coincidence). They were totally shocked when they walked into the room and saw the CRT screen. I got free silicon for all of my home projects for several years after that, even stuff that wasn't out yet (MC68HC11 chips).
The "bad transistors" were traced to a guy who had been dipping metal belt buckles into one of the gold plating solutions somewhere along the process flow for those parts. He did this every week or so with a new buckle which he was then selling at a local flea market. Exactly what happened to him is unknown, but it cost Motorola a lot of $$$$.
Little by little all of that fancy stuff went away. By the time I left my 41 year career in 2014 all of the neat playthings were gone. The X-ray machine was one of the last things to go and I think it was just scrapped. It went when they took out the full in house PC board shop we had that could do multi layer and HDI boards. The last time I visited the plant was before Covid, 2019 or 2020 maybe. The SMD assembly line was the only "toy shop" left in the building and I think it's gone now too.
I soldered three of the suspect transistors into the PC board and I am about 1/3 of the way through the "wiring it up" process. This has taught me just how NOT to design an electronics kit, especially one this complex. I would use a PC board behind the panel, and plug the actual synth board into it, or put the synth on the back side of the front panel board.
Sometime in 1983 or 1984 I got to fix that scope. Some big shots and some engineers from the Semiconductor Division out in Arizona or Texas (don't remember which) had come to Florida in search of answers to an unknown process issue with low Beta in some transistors that they had been reliably making for several years. When they fired it up it refused to finish the power up sequence. They came into the Calibration Lab looking for help but nobody there would rip into a million dollar scope and the closest service tech from the manufacturer was in Boston at least two days away. The lab boss told all the suits that "the evening shift lab tech can fix anything, and he gets here around 3:45 PM." I worked the evening shift in the Cal Lab for a reason, it was so I could be out in the Atlantic Ocean on my Hobie Cat all morning long, then slide into work by 4 PM. So, I showed up in my usual shorts, flip-flops and a T-shirt or tank top expecting my usual night of fixing factory stuff and playing with all the idle toys in the factory. Instead, I was met by half a dozen "suits" who had a justifiable concern with letting the "surfer dude" play with some seriously expensive stuff. They gave me the 45 minute lecture about high vacuum equipment and the ease of damaging a diffusion pump, but the night shift crew worked a little differently than the structured day time environment. There were several different "factories" each with their own maintenance teams, but we all knew each other and often learned what others did. We had high vacuum stuff in three different "factories" since we made quartz crystals and thin film microelectronics. I convinced the suits that I wouldn't destroy their equipment and went on to make my rounds through the two way radio factory. I had to do that before anything else.
Once I got the covers off the back of the scope's rack there was an obvious blown electrolytic cap. I dug up a suitable cap, swapped it in and proceeded to go through the power up sequence. I got a friend from the microelectronics group to go through the hi-vac stuff. Within about an hour I had the "suit's" transistor die up on the CRT screen so I paged them (it was the 80's, no cell phones yet). They arrived in a few minutes since the Mirage Disco Lounge was right across the street from the plant (not a coincidence). They were totally shocked when they walked into the room and saw the CRT screen. I got free silicon for all of my home projects for several years after that, even stuff that wasn't out yet (MC68HC11 chips).
The "bad transistors" were traced to a guy who had been dipping metal belt buckles into one of the gold plating solutions somewhere along the process flow for those parts. He did this every week or so with a new buckle which he was then selling at a local flea market. Exactly what happened to him is unknown, but it cost Motorola a lot of $$$$.
Little by little all of that fancy stuff went away. By the time I left my 41 year career in 2014 all of the neat playthings were gone. The X-ray machine was one of the last things to go and I think it was just scrapped. It went when they took out the full in house PC board shop we had that could do multi layer and HDI boards. The last time I visited the plant was before Covid, 2019 or 2020 maybe. The SMD assembly line was the only "toy shop" left in the building and I think it's gone now too.
I soldered three of the suspect transistors into the PC board and I am about 1/3 of the way through the "wiring it up" process. This has taught me just how NOT to design an electronics kit, especially one this complex. I would use a PC board behind the panel, and plug the actual synth board into it, or put the synth on the back side of the front panel board.
Great story! Good luck with the remainder of your wiring up process.
I heard a similar one from a component engineer, I think working at Raytheon at the time of the story. More than the usual # of dies failing off a wafer. Exam turned up an oil based contaminant. "How the H*** could oil be getting onto these wafers?" Turned out a worker, noticing the hot, dry environment the wafer containers had inside, was keeping their french fries left over from lunch in there along with, so they could be enjoyed still hot and crispy later in the afternoon.
Not quite as nefarious as the belt buckle guy, but makes me glad I was never a people manager.
Well after fighting the flu and slinging solder for over a week the MFOS Ultimate got fed power for the first time about an hour ago. Nothing blew up or smoked. All of the front panel controls seem to work, all three VCO's work, but there is no audio output. There is likely something miswired, but I haven't got the patience to start digging into it right now.
Attachments
I got an email from Perfect Circuit, a seller of everything synth related from component parts to complete synths, with an emphasis on Eurorack. Their "newsletter" is just an advert for stuff they sell, but most of it is unique stuff that I haven't seen yet. It seems that Serge Tcherepnin is back in the game:
"Serge Modular is more visible than ever thanks to the work of Random*Source. Not only due to their work in producing authentic Serge panels in the classic 4U format with refined circuits for improved performance but also from adapting the designs to the Eurorack modular synthesizer format. And through their partnership with Serge Tcherepnin himself, Random*Source has become the source for official Serge modules and synthesizers straight from the man who started it all."
Now, the all-new GTO kicks off the latest generation of Serge Modular designs, drawing its functionality from the famous Smooth & Stepped Generator, or SSG for short."
I read the whole article just to see where the name came from. My guess was correct:
"Even the name of the module, which is an homage to Pontiac's iconic GTO, further drives (pun intended) the point home that it is truly a revved-up version of the original SSG."
In my last post, I had completed the MFOS Ultimate synth but could not get it to work. The next morning I busted out the scope and followed the signal to the VCA where it went in but did not come out. It helps to apply some "V" to one of the Control inputs, so the Amplification can happen. DUMM BLONDE at the controls, the synth works good, but I haven't taken the time to go through the calibration procedure yet.