A while back I bought a semi assembled Anycubic Kossel Linear Plus 3d printer from a Tesla engineer down in the bay area. He was selling it pretty cheap because it had just become a project he couldn’t finish while working 80 hour weeks (WTF, Tesla?!). I completed the build but could just never really get reliable prints from the thing – I’d often have issues with what looked like one of the stepper drivers losing a lot of steps and the effector drifting off to the side.
The stock configuration came with the “Trigorilla” motherboard which is a perfectly fine atmega based board, but doesn’t have support directly for TMC2209 stepper drivers from Trinamic. One of the things that bugged me about the hardware on this printer was that the screws for the endstop switches just barely rubbed on the belts, and, uh, no thanks. The endstops also basically took up about half an inch of possible travel – that could be more build area.
I ran into what was either a bug with Marlin or an issue with the SKR V1.4 Turbo board wherein the three tower motors (some people still call them X, Y, Z, but I’ve also seen Alpha, Beta, Gamma, which makes more sense) would run just fine but the extruder motor wouldn’t turn. The most I could get would be the motor being enabled which caused it to get a hold current, but then it just wouldn’t move. I kinda have in to cargo cult ways and just changed that driver to an A4988 from the old setup. A4988’s still a perfectly fine driver with microstepping, it just isn’t as quiet and doesn’t have the missed step detection, which didn’t matter to me for the extruder which doesn’t have to be an absolute position sort of system.
Now what to do with all this horror pasta?! As soon as I get it up and printing I’m gonna make a little box for the controller.
The bracket it’s on is reused from the stock setup which had the controller right under the heated bed! I didn’t like that. I think that was inherited from the early Anycubic Kossel delta printers which originally came out without the heated bed – that came out as a later option. Mine has the glass bed which I put the build plate sticker on (…why?) – I’ll probably strip that off later as I’m a believer in the magic of the glue stick. All hail the glue stick.
I can’t figure out if it’s actually possible to do Marlin’s delta calibration without a leveling probe. I tried using the delta calibration menus and just didn’t get anywhere. A probe is on order and I’m going to get some spiral wrap or split loom and Velcro for all that… aaaargh!
I was cleaning up and found an old Six-Letter usb charger that I remembered having removed from service but I couldn’t remember why. I opened it up and got a clue right away.
The unpopulated spots on the bottom are for the very important AC line noise filter to keep the high frequency switching pulses from the switch mode power supply from escaping into the house wiring and causing a myriad of interference to tv/radio reception and audio systems (among other things). In many cheap and nasty power supplies that have gone through quality fade, you will find this same issue. What should have been there is a common mode choke that looks like a small transformer, and a safety capacitor. The safety capacitor in this case is one rated to fail non-catastrophically if it suffers a short across its dielectric or a massive over current event from a surge.
Let’s have a look under the board. On a lot of the supplies I’ve seen cheapened like this, two wire jumpers bypass out the common mode choke. The way this works is it’s basically two inductors wound back to back. Any noise common to both sides is attenuated by being magnetically cancelled out. Neat trick, right? Well, it is if you don’t —-
Oopsie poopsie they made a fucky wucky! A real fucko boingo…. No wait, they didn’t, they just didn’t even want to have to change the build of the supply when they ripped out the inductor. The traces just cross over where it would have been inserted into the circuit. Cute. I’m guessing if they ever had to submit a sample of one of these to a potential reseller who was going to run it through EMC testing, they would have added the missing choke and capacitor and cut the shorting traces… then followed up with a production run missing all the parts when the actual order came in.
Of course, the fact this was directly sold via Amazon using a random, rotating six letter brand name, likely appended to a stolen ASIN (don’t even get me started) suggests the risk of anyone ever bothering to test one is approximately zero. I’m pretty sure if it didn’t get banished due to massive electrical noise issues, that unhappy looking melted diode at D1 probably got it hucked unceremoniously in what was supposed to be an e-waste box that I forgot to take for disposal.
When a power inductor overheats and shatters its ferrite core inside a powerfully cursed SIP intercom that was only available single-source from New Zealand with an 18 month lead time *before* the electronics supply chain got crushed….
Warning: Crude humor ahead as a coping mechanism. As I work in an industry that is completely dependent on the continued availability of electronic parts and devices, this shit makes me worry and I’m forced to make terrible jokes to get by. Or something. No wait, I always do this, carry on. Balls.
A few years back I started noticing that the state of the electronics industry was one in which, basically, you go to China for everything. Need parts? China. Need design work? China. Manufacturing? China… but following the Trans-Pacific Partnership, don’t forget to technology transfer all details of your design to the manufacturer, so they can cheerfully take off and ghost shift and clone your product! Need to order parts for one of the few remaining operations somewhere else in the world? China… and… good luck.
To reduce manufacturing costs, almost all work is outsourced offshore to where one can find the lowest possible labor costs for skilled work, and there is a tendency towards consolidating everything into a few large facilities where the cost of operation can be further reduced. To even further reduce costs, parts are not stocked, they are ordered for “Just In Time” supply, as the cost of maintaining warehouse facilities or winding up with leftover parts in stock are considered unnecessary waste.
Well, all of this basically put everyone’s nuts in a Chinese vise with particularly rough surfaced jaws that was being very slowly tightened.
During a certain urinal stain-colored person’s time as president, a ridiculous trade war dropped, which had no effect on reducing problems with Chinese led market manipulations, but put in place a lot of impediments to supply chains to almost every imaginable industry.
In short, the Chinese vise was replaced by a 10,000 ton hydraulic press that everyone’s nuts were secured into with Gorilla Glue, with the motor left running, and about 300 different corporations and foreign governments given little clicker remotes that will smash and explode whatever’s between the surfaces with a quick and easy press.
Several of the buttons have all been pushed now. SPPPLRRRRRRTTTTTTTTTT
From what I gather, we’ve reached a point where the industry is either going to have to sit on its hands waiting almost a year for parts, or actually, you know, do something about it. I’ve been seeing several people lament that they’ve had to suddenly redesign boards to work around parts that have gone unavailable with 40-80 week manufacturer lead times (!) or entirely redesign around different microcontroller architectures and stuff. Boards are being designed to have parallel connections to two or more different pad layouts to accommodate whatever part happens to be in stock when the device goes into production…. but that only gets you so far.
Common parts like normally 30 cent voltage regulators are anywhere between 60 week lead time and Fuck You NLA because some companies saw the light that there’s massive profit in inducing scarcity…. and bought up all the parts to just sit on them, selling what are either the real parts or complete fakes, for $15 and up. The same buy and scalp model that has applied to computer video cards is now being used on, well, everything…. and it fucking blows. Normally these brokers are only able to lock down supplies of obsolete parts that are no longer in production — this is the first time I’ve ever seen them able to leverage the lead times of active, in-production parts, because they’re able to create unpredictable demand spikes and kill the Just-In-Time supply.
Just like what morons did with toilet paper………
If we actually had manufacturing capacity outside of, you know, one or two big fab houses in China that have HUGE lead times because they have other orders to fulfill before they can tool up for a run of one particular part prior to manufacturing, then packaging and overseas shipping, we probably wouldn’t have this problem. Also, if there were actual stocks of parts kept by the manufacturers of devices or by suppliers, we wouldn’t be in a situation where a few guys who have some extra Shitcoin money can buy up all the STM32’s and completely lock down availability. Fuck you if you claim you didn’t see this shit coming, this was ENTIRELY AVOIDABLE, and at this point there is only one way out… and it’s a rough one… but we’re gonna have to take it.
The ONLY way out of this will be to re-establish manufacturing of electronic parts in the US and other places. The big problem with doing this? We’ve already sent everything to China…. But it still has to be done.
The best thing I can liken this to would be the electronics industry in the USSR after they effectively pissed off EVERYONE else and nobody would sell them electronics, so they had to scramble to get production set up locally. Some of it had to be done with the aid of industrial espionage because they just didn’t know what they were doing, and a lot of it was really weird looking or strange quality because… they just didn’t know what they were doing. This weird Soviet transistor could be used as almost a textbook example. Eventually, though, they got up and running making some really high quality stuff, particularly vacuum tubes which remained in production for quite a while to support older equipment.
Yeah— stuff’s gonna look weird at first, and there may be reliability issues, but we’ve gotta do it, or we simply will not have electronics anymore, and that just isn’t gonna work for anyone.
I don’t really know what we have left in the way of semiconductor fabrication, testing, or fabrication facilities in the US. The only one I’m really aware of is Skyworks in Newbury Park, California, who make the RF power amplifiers used in… well, damn near everything— if you’re reading this on a mobile or wifi equipped device, your packets went through one. It was also, amusingly, featured in a music video. The building shown in the aerial photo zoom is not the Skyworks plant, it’s some arena in Utah. I’ve always imagined it must make an interesting sound when you walk down that hallway with the air jets that dislodge dust from your clothing. Maybe I’ll find out firsthand when someone brings me in because I’m the only one who knows how to program the variable frequency drive in a conveyor belt in the cleanroom or something.
Also, what is with the weird half-video-line artifact at the bottom of the screen here? This is an older video and was edited in 4:3 standard definition. I’ve seen this half truncated line at the bottom of the screen in a few different videos and I have no idea what causes it. I got into television after it all got digital and hi-def and boring and stuff. My DVCAM gear doesn’t do it, I’m thinking it’s possibly attributed to a timing oddity on analog tape gear. Maybe we’ll be going back to tape soon if we don’t get flash memory manufacturing geared up in time…. or… film. I mean, someone at Eastman Kodak must still remember how to make that stuff, right?
A couple weeks ago I was at the tower making bad jokes about the liquid cooled EEV ESCIOT tube based Harris PowerCD transmitter being a space station toilet.
Really, it’s a three stall restroom, and today I got all three flushing again… and learned more about how freaking weird and scary *pure* deionized water can be.
First, here’s a questionable explanation of what’s in there. You’re looking at two separate liquid cooling loops. The external one which exits the cabinet at upper left circulates an ethylene glycol coolant solution (similar to automotive antifreeze, but nigh unobtainable outside of ordering it off Shamazon) between the heat sources and a set of fan cooled radiators outdoors. It’s circulated by an external pump station. I marked its flow with the orange arrow emojis. The internal one has a pump in the cabinet as it’s a closed loop within. The vertical accordion looking piece is a heat exchanger. Attached to the door on the left are two filters that keep the deionized (DI) water as pure as possible to keep its electrical resistivity high.
The supply manifold at the top sends the anode and collector water jacket water supplies to the tube cart around the front. The small line coming out the middle feeds the filters; you can set their flow rate with a valve up there. Everything finally returns to the pipe at the left that sends the DI water back to the reservoir on top. Now, have you noticed the middle finger emojis yet? Well.
In the DI water returns from the anode and collector are these Seametrics flow sensors. The pinwheel has magnets in two opposing vanes, and a Hall effect sensor screws into the recess seen at the bottom here. By measuring the interval between pulses, the transmitter controls can determine if there’s enough water flow… until the sensor breaks.
Now let me say this, I see absolutely nothing wrong with the design and build of the Seametrics sensor. It’s actually damn cool for what it is. No metal parts contact the working fluid, and it rides on a ceramic shaft and ruby bearings like a fine watch movement (and that wouldn’t even have ceramic shafts… Or would it?)
The Seametrics is even completely field rebuildable!
So, uh, time to be creeped out and amazed by mere water. In the picture of the cabinet you’ll see there’s one more sensor mounted horizontally in the glycol line. This never fails, as the glycol solution has some lubricity to it – that is to say it’s slick and forms a film that tends to isolate facing surfaces from direct contact, just like an oil would. The DI water, however…. No. When I got some on my hands, it felt really weird, almost more like I’d just rubbed them with a really cheap and nasty hand sanitizer that was stripping the oils and leaving behind sticky yackage. So let’s see what it does to those extremely hard, smooth bearings:
The bore of the bearing above has become egg shaped. This wasn’t even the worst one — that distinction goes to the one that was in the collector flow meter:
I wasn’t able to pull this one apart for further inspection but didn’t need to. You can see the axle right through the plastic — it chewed completely through the ruby bearing and started digging into the plastic. Funny thing was this one would work perfectly UNTIL the water temp rose to about 46 degrees C when I put the cabinet in Beam On (normal RF output state), at which point it’d abruptly start ticking down from 12.6 GPM to 10 and the controls would kick the beam supply off to avoid meltdown. After rebuilding both sensors on the DI water side, the flow readings come up the moment the pump starts and stay stable.
Want to read more about how damn weird pure water is? There’s a somewhat sensationalized (in their usual style and don’t even get me started on that Supermicro fiasco) article from Business Insider about the Super Kamiokande which is a massive subterranean neutrino detector tank lined with the stuff that physicists have had to enter on a rubber boat for maintenance. Just imagining what that’d feel like across a large area of skin makes me want to go rub an Aloe Vera leaf on my entire body.
This is several kilobucks worth of hardware that was misapplied to create a special filter that…. nothing would pass. See if you can spot the issue!
Our IT director dropped a non working Wyse serial terminal on my bench. I don’t think he expected me to tear it open and try to fix it, I think it has an open diode or bad cap in the +5v supply. It’s gonna be the serial console for a Linux machine if I can get it glowing.
Oh, I have such a love-hate relationship with some manufacturers.
Now… I had previously made a vague shitpost while working on a Harris Apex A2X exciter. This would have dated back to, well, when they were Harris Broadcast before the spinoff that left them independent as GatesAir, with another division becoming Imagine Communications
Imagine Communications…. Because they aren’t necessarily ever going to work outside of your vivid imagination
Anyway let’s just get right into YELLING IN BROADCAST ENGINEER. First stop: The Apex M2X oscillator board.
Frequency stability is vital to generating a good solid digital TV broadcast signal. The Apex M2X features a disciplined OCXO (Oven Controlled Crystal Oscillator) with several options as to how to ensure proper longterm calibration – it has a GPS receiver, 10 Mhz external reference, and 1 PPS external reference options provided. More on OCXO references here (this describing a more basic, free-running one, without sync inputs). However, the OCXO itself is, uhhhhhhhhhhhh
YOU GUYS JUST GLUED A BLOB OF PACKING FOAM OVER THE OSCILLATOR AND A LINEAR VOLTAGE REGULATOR WHY DID YOU DO THIS
LOOK AT THAT BULGING CAPACITOR OF HAPPINESS ALL UP IN THERE OH BABY
I wanted to peel this crap off but I was fairly confident that if this is anything like the antistatic foam that ICs used to come packed in, it may have broken down and corroded parts under it and I’d be faced with irreplaceable parts that went out of availability two years before they sold this exciter crumbling to dust. So I left it alone. It only has to last about another year, if even that. I hope. Did I mention HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH? Because HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH
So. Let’s move on to the power side.
The power supply is along the left wall of the exciter. On GatesAir’s admission, the power supply pretty much expires and becomes a ticking time bomb after 5 years – power it off and it will never ever start back up again.
But that’s unrelated to, uhhhhh, The Internal UPS Thing Of WEIRDNESS
For unknown reasons they felt the need to give the unit battery backup. It does not fully power it, like, the RF output disappears once it’s on battery. I think all it does is keep the controller with the RTAC data* in memory alive, and keep it from having to entirely run through the several minutes long boot process following a momentary power glitch.
It is, however, FUCKING RIDICULOUS.
On later versions it uses a lithium ion battery which consists of three 18650 cells, but rated at only 1.5 amp hour, which suggests… some 18650s of hilariously low quality. Behind the board is a charging / BMS circuit that steps the battery voltage up to 12V to keep the exciter powered. On earlier versions, they went through all the trouble of building this elaborate charger/step-up board, similarly….. And then, waaaait for it—–
Ok, take a good look at this, and prepare yourself for the description of what you’re actually looking at: someone… went through the trouble of getting a spot welder in house and welding tabs to unmatched dumpshit tier** Energizer retail packed NiMH cells.
Because, uh, only the finest with Brand HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH.
There is another memory backup battery, but I’m guessing that’s mostly just for the realtime clock.
* Sorry, I am not currently in a sufficiently coherent mental state to describe what RTAC does, other than that it is Fucking Magic and if the Fucking Magic doesn’t work, the signal comes out on air as distorted non-decodable dog turd
** only slightly better than Duracell