Ed Nisley's Blog: Shop notes, electronics, firmware, machinery, 3D printing, laser cuttery, and curiosities. Contents: 100% human thinking, 0% AI slop.
After Eks set me straight on cleaning the contacts involved with the Ampeg’s Echo circuitry, we emerged from his shop into brilliant sunshine. He looked into the thing and shouted “Tin whiskers!”
It turns out the Hammond folks made the outer frame from tin- (or, shudder, cadmium-) plated steel that has grown a dense crop of whiskers on its interior surface. They glittered in the sunlight like a carpet of crushed glass, with the longest ones maybe 3 mm tall!
This view looks nearly parallel to the side of the channel (upward as it mounts on the speaker box), with the steel wall to the bottom of the image. I applied gruesome contrast stretching to make the whiskers more visible:
Ampeg Spring Echo Unit – Tin Whiskers
This is the first time I’d ever seen a tin whisker in person and there’s a bazillion of ’em in there!
If that Ampeg had transistorized components, it’d be dead as a doornail! Fortunately, a tin whisker doesn’t stand a chance in an analog vacuum-tube circuit. The power supply puts 400-ish V into 40 μF caps, providing plenty of energy to vaporize the errant whisker; all you’d hear is a pop.
Mad Phil asked me to fix up his trusty Ampeg B-12-XY (*) bass guitar amp, having recently fired it up and discovered that the power output tube plates glowed red-hot. I’d planned to replace the electrolytic caps, but Eks, who does this sort of thing all the time, suggested that leaky interstage coupling caps can also cause that problem; the leakage wrecks the phase splitter bias and thus kills the drivers.
While poking around in the amp I found that the Echo hardware circuitry doesn’t match the schematic for either the B-12-X or B-12-XY. Mad Phil says that’s probably because he had the factory upgrade his original B-12-X to a B-12-XY for the munificent sum of $25, back in the day. It’s unlikely you’ll ever need this, but here’s what I found:
Ampeg B-12-XY – as-found Echo circuit
The topology resembles the -XY schematic, but with different tube sections and part values.
The Echo unit over there on the left consists of two springs with magnetic transducers on each end, evidently made by the Hammond Organ folks, who should know something about reverb. This is the bottom view, with the unit attached to the board that supports the amp chassis:
Ampeg Spring Echo Unit
The input transducer, just in case you forget to label the ends before you take it apart:
Ampeg Spring Echo – input end
And the output transducer:
Ampeg Spring Echo – output end
Getting the thing off the speaker box posed a bit of a problem. Remove the four big screws holding the chassis to the board, tilt it carefully forward, hold it in place while you remove the six nuts-and-washers from the vibration isolators, then transport the whole disjointed affair to the workbench. Turns out you (well, I) can’t get the RCA plugs out of the Echo unit’s sockets from the top of the board, but the unit’s mounting screws are on the bottom of the board, where you can’t get to them before you remove the board. Of course, the cables leading to the aforementioned RCA plugs tether the chassis to the Echo unit with pretty nearly no slack at all.
With everything apart, I rounded the ends of the RCA plug cutouts enough to get them out from the top the next time around, with the board screwed in place atop the speaker box:
Ampeg Spring Echo unit – top view
After putting the whole thing together with new caps, the Echo circuit didn’t work. I had cleaned the contacts and connectors, but Eks showed me how it’s really done. Apart from the rotted caps, all the other problems came from minor corrosion in switches, connectors, and tube sockets. Now I know better.
The thermoplastic (who knew?) pads melted right off my long-reach clamps while calibrating those thermocouples, leaving the thermoset (who knew?) clamps behind. I tried a few of the obvious candidates for the job with no success, but (while fiddling around with something else) I came upon an unopened tube of Permatex Ultra Copper copper-loaded silicone gasket glop.
The cured silicone rubber is very flexy, which is sort of what you want in a pad, even if I’m not convinced they’ll stay in place. They seem securely mounted in the recesses of the pad tips; I worked the glop in with a screwdriver tip.
Using the pressure washer to blast the crud off the propane grill has become an early summer ritual around here. I’d reconfigured the extension pipes to reach up the side of the house, so I started by swapping the connectors around to put a shorter pipe at the handpiece. Surprise: those connectors were firmly affixed and a rubber strap wrench on the pipe lacked enough grip.
Rather than wreck that nice chrome plating with a pipe wrench, I clamped two pieces of scrap plywood in the drill press and poked a half-inch hole right down the midline. Add a dab of rosin to improve traction, crunch everything in the bench vise, and spin the connector off.
Well, that’s the way it went for the first connector, with the PTFE joint tape I remember adding last time around.
The connector on the other end was more recalcitrant, perhaps because it still had the manufacturer’s joint compound in place. It eventually yielded to the gentle persuasion of a propane torch, applying just enough heat to wreck the compound’s grip.
The good thing about a plywood clamp is that I don’t form a deep emotional attachment to it: make one when it’s needed to fit the pipe at hand, don’t worry about a precision fit, regard it as a consumable, and move on.
The SLA batteries in the MGE Ellipse 1200 UPS finally gave out. This picture shows how they’re arranged inside the box:
MGE Ellipse 1200 UPS – battery arrangement
They’re 12 V 5 Ah batteries that are about 12 mm thinner than the garden variety 7 Ah batteries you can get everywhere; they’re not the same size as the generic 5 Ah batteries you might think would work. Of course, there’s not enough room inside the stylin’ case for the larger ones, either. I’m thinking of using fatter batteries anyway and putting a belly band around the gap. Maybe an external battery box with a chunky cable burrowing through a hole in the UPS case?
For what it’s worth, APC absorbed MGE a while ago (so the MGE website redirects to APC), got Borged by Schneider, then spat out MGE’s consumer grade UPS units to Eaton. You won’t find any of that documented anywhere, but here’s the response from APC after I didn’t find this UPS on their list:
I do apologize; when APC was acquired by Schneider Electric, the single phase UPS line that MGE once offered was sold to Eaton. Eaton now provides support for the MGE single-phase products. We do not sell batteries for these models. You will actually need to contact Eaton for further assistance regarding the MGE Ellipse units. You may click on the link below to go to Eaton’s website:
The Eaton website does have a battery replacement for this one, but sporting the dreaded “Contact us for price” notation. Given that I got the UPS cheap-after-rebate, I’m thinking maybe this isn’t worth the effort.
I upgraded to ReplicatorG 25 and the Thing-O-Matic promptly got weird: the initialization code slowed to a crawl. The motors ran fine, the motion was properly coordinated, but the thing moved at a minute fraction of its normal 100 mm/s.
This was most obvious on the first move to the center of the stage after homing the axes. If you peer into the source code, that instruction looks like this:
G0 X0 Y0 Z10 (pause at center to build confidence)
The comment tells you exactly why I put that move in there when I first started tinkering with start.gcode: I long ago discovered that automation doesn’t always do what you want, so having a simple verification at the first opportunity sometimes pays off big.
Anyhow.
A bit of rummaging showed that RepG 25 has changed the semantics of G0, which is supposed to be a fast move to the programmed coordinates. Now G0 moves at the feed rate set by the most recent G1 and also accepts an F parameter, which it shouldn’t. I suspect somebody refactored the code and didn’t notice that G0 isn’t supposed to work exactly like G1.
Strictly speaking, I do not have a beard: I simply do not shave (*). There being no money in selling Trimmers for the Non-shaving, a while back I bought a battery operated Beard Trimmer. The NiCd cells lasted for the predictable few years and recently gave up the ghost entirely: an overnight charged produced a weak buzz with no cutting action to speak of.
The case uses one-time snap-together latches, which makes dismantling it a challenge. Start by removing all the gimcrackery on the business end, then pry out the two latches holding the it-was-white-once cutting length adjustment ring. With that out of the way, undo the two latches inside the top and work your way down, prying the case halves apart in the way the overlap flange doesn’t like, so as to force the latches loose.
This picture shows the six latches, three on each side. The ones just to the right of the blue impeller require the most cursing:
Beard trimmer case and innards
The circuit board snaps out, with the two PCB contact areas clamped down by springy contacts leading to the motor.
Beard trimmer – battery charger PCB
The two NiCd cells boast of their High Energy, but they’re only 600 mAh. That’s actually too much for this high-drain, short-run application, as they don’t completely discharge. They’re held in place on the right end with a blob of hot melt glue:
Beard trimmer – NiCd cells
I unsoldered the cells and gave ’em a brute-force overnight charge at C/10 = 60 mA, then ran a discharge test (clicky for more dots):
Beard Trimmer – NiCd Discharge Test
Lookee that! The cells still deliver their rated capacity, even though they no longer worked with the stock charger. I repeated the slow-charge and discharge trick, which produced a perfectly overlapping trace.
Flushed with success, I unleashed the built-in charger overnight, then produced a third overlapping trace.
So they suffered from voltage depression, most likely due to never being completely discharged and then being overcharged far too often. That’s cured by a complete discharge and recharge, which worked perfectly.
I hack back the overgrowth when it gets bushy and recharge the trimmer when it seems to be getting weak, which used to take a week or two. That’s a bad way to maintain a NiCd battery, particularly as the PCB applies a very low load to keep its computronium running, but I have better things to do than babysit a beard trimmer. Honest.
Anyhow, assembly is in the reverse order and it’s perfectly happy again.
I probably won’t change my evil ways, so the next time I’m sure the battery will be really and truly dead.
(*) Not shaving adds about ten minutes a day to my life, which I regard as a fair tradeoff over the course of several decades. It also added a decade to my apparent age, Back In The Day when that mattered. Now it seems to knock off a decade, which isn’t entirely a Bad Thing.
The Smell of Molten Projects in the Morning 