A little status update:

I have reached the conclusion that it wouldn't make much sense to package DIY kits for the Covox Speech Thing and FTL Sound Adapter replicas.
There's so little in there that it would almost cost more time and money to safely package the individual components and ship them in a box than to assemble the unit and to put it in an envelope.

As you can see, the devices are already ~80% assembled and they are arranged on a test print for the labels.
The heart is still missing, though: I don't have the R-2R resistor networks, yet, but they should arrive here within the next couple of days.

Some more components have arrived. This is one of the resistor networks:

And when it comes to Covox Speech Thing replicas, I think this is about as close to the real thing as we can get:

Looks awesome to me.

Finally done!

Challenge for the reader: Try to get closer to the real thing!

That's awesome! 😁

Just one thought - What about adding an optional capacitor ?
The original Speech Thing had one as a low-pass filter inside the external speaker, if memory serves.

Or maybe it's just enough to include a small note in the package,
which says which filter circuit and specs (cone diameter, impedance etc.) the Covox speaker box had ?

Anyway, these are just my two cents and are in no way badmouthing this replica. 😊
Just mentioned it, so 99% replication of the Covox sound can be enjoyed by someone.

oh these look great!
is there a price set yet?

Jo22 wrote:

Just one thought - What about adding an optional capacitor ?
The original Speech Thing had one as a low-pass filter inside the external speaker, if memory serves.

If you scroll up you can see that there's a footprint for a filter capacitor on the PCB. Populating it with a 4.7nF capacitor makes the audio output electrically identical to the FTL Sound Adapter.

keropi wrote:

oh these look great!
is there a price set yet?

1) Thanks! 2) Let's make one!
I'd say that, since there's still a difference between a hobbyist project and self exploitation, that'll be around €10 per device, plus shipping.
That's admittedly clearly above the BOM cost, but at the same time clearly below e.g. the €22 for a CVX4.
You'd basically be paying for components, component shipping, soldering, cutting notches into case shells, assembly, labeling and testing, i.e. the device itself, while I'd consider the development process a spare time activity.

Speaking of which... I still have to test them all.

Put me on the list 😀

I want in too - one of each
price is super great btw no need for breakdowns

The next up is the Sound Jr. replica?

I shouldn't, but I think I'll bite. If there is still enough available: I'll take one.

that looks so nice !
I'd love one !

hard1k wrote:

Put me on the list 😀

Awesome! ... Which list?

Paralel wrote:

The next up is the Sound Jr. replica?

Yes, but only as DIY kit. Pre-assembling them would take way too much time, especially with my equipment.
I still have to solder a prototype, though.
But I'll wait for more favorable weather. For the past couple of days I haven't wanted a soldering iron anywhere near me.

root42 wrote:

I shouldn't, but I think I'll bite. If there is still enough available: I'll take one.

Which one?

dreamblaster wrote:

that looks so nice !
I'd love one !

See above.

I still haven't found the time to test all the devices, but we'll get there.

Hello,

If there are still some available, I'd be glad to buy one of each.

Thank you

obaltus wrote:

Hello,

If there are still some available, I'd be glad to buy one of each.

Thank you

There are and we can surely arrange for that. (e.g via PM)

But I've got news, as well: After roughly one hour of soldering, here comes the first SoundJr replica:

This is the finished product:

And this is what's hiding inside:

There's been some component substitution that I described earlier in this thread.

The two 1nF filter capacitors are pure guesswork.

I put the word "replica" and the year on the PCB so that no one can mistake it for the real thing.

So far, I've been able to get sound by using it as a generic LPT DAC, the volume adjustment, however, remains untested.

I haven't found terribly much time for this, lately, but here's a little update:

The files for the SoundJr replica have been on github for the past couple of weeks.

There's one problem, though: Dungeon Master does not recognize the FTL-style adapter, which means that either the detection routine is buggy or the series resistors for additional short circuit protection are to big.
I have a feeling that it might be the latter. While that would not be a catastrophe it would certainly be annoying, because I'd have to swap them (and find the time for that).

Other than that, I'm still looking for a way to create high quality recordings of the output.

Interesting that it didn't detect it. I would imagine you are correct regarding the short protection.

I found a solution and I found a problem.

When I put a 390Ω resistor between pin 12 and 17 of the pass-through port of a Speech Thing replica, Dungeon Master will detect it as FTL Sound Adapter. Let's play it safe and use 220Ω for the replacement resistor packs in my FTL Sound Adapter replicas. (They currently use 2.2kΩ resistor packs.)

So that's the solution. There is a problem with the Speech Thing replicas, though.
I have been testing them with active speaker boxes that are not labeled as left or right and kept wondering why they do not produce any sound when I switch them from stereo to mono mode.
It looks like the cables I used for the Speech Thing replicas are of the kind where red means left and white means right. 😵

Yeah i also applied this Patch but we simply run a wire to use Dungeon Master.
Its very interesting it sounds exactly the same as DSS.

And there is a complicated detection routine for dss.. 😉

The Speech Things are now wired up correctly with the signal on the tip of the plug and ground on the sleeve. The ring is not connected.