So apparently in 1993, Avon (the MLM (not that MLM (or that kind either))) commissioned a Soft Foam Telephone.
I got one for @gewt but sadly, it arrived broken.
So apparently in 1993, Avon (the MLM (not that MLM (or that kind either))) commissioned a Soft Foam Telephone.
I got one for @gewt but sadly, it arrived broken.
I opened it up (it's glued together! AHH!) and it turns out it's a Recyco Inc SoftPhone
Inside the plastic, we've got a single-sided PCB with a single IC and a bunch of discrete (and discreet!) components.
The IC is an HM9102D. That's a Hualon Microelectronics TONE/PULSE SWITCHABLE DIALER WITH REDIAL
So the way this thing is broken is odd: You plug it in, and you can turn it on (yeah this landline has an on/off switch, which is itself weird) and the LED comes on, but that's it. You CAN hear it clicking when you pulse-dial, but nothing otherwise
This thing was definitely hand-soldered. With one possible exception they look fine, though.
Nope, there's no bridge here. I checked.
Annoyingly there's two version of the datasheets: a 1-pager that just gives the specs, and a 5-page one that talks about circuits and such.
the 5-page one is only available in chinese and when I attempted to translate it, google just gave me punctuation.
no really. it only included the punctuation.
Testing in circuit, I'm seeing 2.1v across VSS/VDD on that chip.
The nominal voltage seems to be 2.5v? so that might be a problem
hmm. no, if this translated electrical characteristics table is at all correct, it should be fine down to 2v. So we're close, but not dangerously low.
@skotchygut yeah, that's the one I specifically tested! but no, it's fine.
@foone that one trace on the far left side about half way up looks pretty close to a bridge of three pins instead of just two
this thing has a bunch of little caps and diodes and transistors. I've visually inspected it but I don't see any issues with any of them.
I'm getting the expected -51V on the input, so there's no problem with the cord or the PBX powering this
and the fuse is fine, I did test that.
Here's the translated(ish) schematic from the HM9102D datasheet.
I'm gonna assume this device is basically this design with with minimal changes.
I'm suspicious of this goop. It looks like they hotglued the caps down, so... why would there be brown goop? Maybe these caps leaked out the bottom and it pooled under them.
C2 & c9 are 47µF 16v caps
C6 & C8 are 100µF 16v caps
C7 is 1µF 50v cap
@xyhhx FOONE B TURING: GOOP INSPECTOR
@foone "im suspicious of this goop" is definitely a sentence
Desoldered them all. Let's test.
My cap meter is saying C7 is 1116nF, or 1.1µF. Close enough for jazz.
my MTester thing says it's got an ESR of 7 ohms, which is high, but I don't think this is gonna mind.
The two 100µF caps:
c6: 144 µF, ESR 0.5 ohm.
c8: 148 µF, ESR 0.6 ohm.
that's a lot higher than they're supposed to be but again, it may not matter with a device this analog.
the 47 µF caps:
c2: 60 µF, 1.3 ohm ESR
c9: 62 µF, 1.2 ohm ESR
so none of these caps are good, but none are super smoking-gun either. Still, I desoldered them all, so I'm not putting any back. I'll swap them for some new ones once I can order some
replacements will be here tomorrow!
I don't currently have a good way to test the oscillator, but I suspect it's working: I get (well, got, I would get nothing now, since I stole all the caps) clicks in pulse dialing mode, which sound vaguely enough like pulse dialing that I'm gonna assume it is. Without an oscillator we shouldn't be getting any clicks
There are some ceramic caps on here too, but I didn't desolder or test any of those. I don't think I really need to? unless you fry them by overvoltage/overcurrent I'm not sure they're gonna die anytime soon.
I also don't know if this thing ever worked: this one came from ebay apparently unopened over the last 32 years
It's entirely possible it never worked, and they soldered it together wrong
@moira fortunately there's no traces there (since it's single sided) so that shouldn't cause problems
@foone ceramics almost always hold up, I wouldn’t bother.
That brown goop looks like a particular glue which is known to become corrosive and conducive over time. But if it’s that, why also hot glue? I has a confusion.
So the current plan is:
1. Stick new caps on
2. Test
3. If not work, start yanking components and documenting what they are until I run out of components or find some that are broken
@lritter not me! I'm diabetic. some 1992 punk probably was, though
@foone who been eatin chocolate
@huronbikes they sold this in a lot of varieties (the phone part is the same, they just need to make new foam cases for each license), so it wouldn't surprise me if they did!
@foone I feel like there was a missed opportunity to make it completely yellow and call it Banana Phoam
@foone yea MLCCs don't really die. You can regenerate them by heating them up, might be worth a try if its that old but unless its precision caps for some analog function I wouldn't suspect them to make the pass or fail difference.
@timonsku yeah. I'm not gonna worry about them unless I can rule out everything else.
@mattsains I can't and it's on my desk! I'm anosmic :)
@foone I can smell this pcb from here
@karotte this one might be old enough that they routed it by hand with stencils.
@foone This particular style of PCB routing seems to be common among really cheap single-layer PCBs. I've been wondering, is there any practical reason for it or is the result of a particular layout software?
@foone These kinds of boards were usually done using wave soldering.
@plaes I figured this one wasn't because of all the flux residue and inconsistent soldering, but maybe they wave soldered most of it and then touched up some bits by hand?
@plaes maybe! But it was in a closed box, I don't think it's been opened since 1993
@foone ...or someone has tried to fiddle with it...
got the new caps in. time to solder.
Damn. No difference.
it is definitely trying to pulse-dial, even if I can't hear it I can see the LED flash.
lemme try blind-dialing my own phone number. maybe this thing just has a bad speaker, and is in fact working?
YO! it works.
it took a lot of tries to get the fucker to dial, but I was able to call myself, and hear myself (in one direction only, because the speaker isn't currently connected)
reconnected the speaker, but got nothing. So it's still one-way
lemme locate a spare speaker
it turns out I have no spare speakers.
how strange that a noted misophonic would have no spare speakers!
okay I plugged it into the speaker of a broken flip-clock I just so happened to have within reach.
It works fine! I can even tone-dial, although the weird part is that I can't hear the tones. There's just a slight click noise.
so I think this thing might have just had a bad speaker from the beginning. whoops. that should have been easier to diagnose!
076萌SNS is a social network, courtesy of 076. It runs on GNU social, version 2.0.2-beta0, available under the GNU Affero General Public License.
All 076萌SNS content and data are available under the Creative Commons Attribution 3.0 license.