A310 speaker sound fix

Apologies if this is a simple question (I’ve googled it to death).

Is there a link somewhere on the A310 board (or configuration setting?!) that disables the internal speaker UNLESS the rear headphone socket is used?

The speaker on my A310 beeps when you CTRL-G from BASIC but there is no other sound from the speaker at all until you plug in headphones(!) . . .

The internal speaker works ‘normally’ with an old (AKF11) acorn monitor (fingers crossed it keeps working for some time yet) using a video cable with separate audio jack plugged in (and also seems to work okay when I just plug in ‘modern’ headphones).

It’s easier for me to use a modern monitor via upscaler hence the question (i.e. how do I enable the internal speaker to work ‘normally’ without the need for headphones?) . . .

The A310 in question is ex school computer so I am assuming the teachers preferred to disable the internal speaker to stop kids disrupting lessons?!

The A310 has been upgraded to OS 3.11. No other board modifications that I can obviously see.

I’m not aware of any setting that would do this, it sounds like a fault/modification somewhere.

Does it have to have actual speakers connected for the internal speaker to work, or would a dummy jack plug in the headphone socket do?

Memory plays tricks – was there not a star (*) command like *speaker Off/On

was there not a star (*) command like *speaker Off/On

yes – *speaker on|off and *audio on|off
The first controls the internal speaker and the second the wider sound system

@ Andrew

I found a dummy jack plug and when plugged into headphone socket there is no sound to internal speaker.

I’ve noticed now that when I plug in the ‘modern’ headphones, the internal speaker seems to be ‘left’ and the headphones only have sound in the ‘right’ ear/speaker! I believe the internal speaker is normally mono . . .

I tried CTRL-G via BASIC again with headphone socket empty but I think I imagined the beep (it’s silent I mean when headphone socket is unused).

Also, when I plug in the monitor audio cable into the headphone socket the internal speaker is in fact disabled (I guess that’s ‘normal’?) . I mistakenly believed the internal speaker made noise too . . .

@ Colin & John

Yes I have typed *audio on and *speaker on via command line but no luck sadly (internal speaker silent).

I tried the suggestions in the service manual (
http://chrisacorns.computinghistory.org.uk/docs/Acorn/Manuals/Manuals.html ) but no luck . . .

@ All

The internal speaker is working (i.e. only when ‘modern’ headphones plugged in) but is seemingly disabled when headphone socket does not have headphones plugged in (I don’t have any ‘old’ 32 ohm headphones to test what they would do when plugged in) . . .

I guess there must be a modification to either the speaker or the headphone socket to disable the internal speaker unless headphones are used?! OR perhaps modern (non 32 ohm) headphones create some sort of feedback or create an unusual bridge in the associated circuitry which enables the internal speaker some how?!

I can cope with having headphones dangling at the back but it would be great to find a solution. :o)

P.S. The A310 board does seem to have lots of empty chip/component sockets. Is it possible that the board does not have the necessary circuitry to use the speaker? But why install a speaker if the board can’t utilise it?! School techs adding redundant components perhaps for future enhancements? Is it likely that some schools would have ordered ‘silent’ A310s to stop pupils annoying teachers?! My teachers (1980s) were pretty uptight . . . LOL

P.P.S.
The OS 3.11 upgrade is via a daughter board (with flying leads to at least one existing board chip). I wonder if the internal speaker ‘fault’ has something to do with that (perhaps the A310 board is not totally compatible with the OS upgrade or something has become disconnected)?

The internal speaker is working (i.e. only when ‘modern’ headphones plugged in) but is seemingly disabled when headphone socket does not have headphones plugged in (I don’t have any ‘old’ 32 ohm headphones to test what they would do when plugged in) . . .

So, you mean that you’ve done the following:

*AUDIO ON

*SPEAKER ON

And then you’ve pressed CTRL + G (when at the prompt) and no audio came out from the speaker. However, if you connect a jack (what do you mean with “modern” headphones?) and press CTRL + G then you can hear the sound from the internal speaker?

If so, something might be wrong with PL6, PL7 or PL8. In this case I would suggest to post the question on stardot forum (https://stardot.org.uk/forums/), lots of people there still repair Archimedes and may have more experience with this type of problem.

The A310 board does seem to have lots of empty chip/component sockets.

You mean soldering pads not used for a socket? That is normal.

Is it possible that the board does not have the necessary circuitry to use the speaker? But why install a speaker if the board can’t utilise it?!

The Archimedes speaker is usable and works fine in normal conditions. You may have a machine with some problem, this is very possible given the age of the A310 and the fact that was used in a school.

The OS 3.11 upgrade is via a daughter board (with flying leads to at least one existing board chip). I wonder if the internal speaker ‘fault’ has something to do with that (perhaps the A310 board is not totally compatible with the OS upgrade or something has become disconnected)?

To install RISC OS 3.xx on the A310 it was required to have a daughter board, so this is normal. I don’t think the daughter board has anything to do with the problem you are having (although if, without seeing the machine, it’s impossible to determine what is the issue).

Again, probably the best place to ask is startdot.co.uk forums.

Good luck!

The A310 board does seem to have lots of empty chip/component sockets.

The A3xx series had the same mainboard as the A4xx series, it was just that the ST506 (aka MFM) harddisc interface stuff was not fitted on A3xx machines. I think there were also slight differences in video output (high res mono output only on A4xx) and backplane (co-processor only on A4xx).

@ Paolo

Thanks for the tip. I will post issue on stardot and update this with link later.

The ‘modern’ headphones are ones manufactured in the last 10 years. I noticed on various acorn (RISC OS era) machines in general that ‘modern’ headphones prevent the machine starting up at all when plugged in (blank screen with screeching sound)? I’m surprised the headphones don’t have the same affect on the A310 (and that they appear to enable the internal speaker!) . . .

There’s a wealth of information on the old “Chris’s Acorns” web site, saved for posterity on the computing history site.

For the A300 you may find this file particularly useful.

Circuit diagrams are right down the bottom end of the PDF file, the audio info chapter is about halfway through.

You may not have the right impedance headphones to match the output circuit.

A quick look at that manual suggests that just having the wrong impedance would not give the symptoms you are seeing.

There are three leads going to SHT2, labelled “sound”, “speaker” and an input labelled “smute” which appears to mute the speaker.

There are three leads going to the headphone jack socket – left, right and ground. There doesn’t appear to be a connection between these two groups.

I’d be wondering where exactly the “smute” lead goes. It’s just possible that the headphone jack socket includes a switch, and is used for this. Such switches are normally grounded when the jack plug is absent, and open circuit when it’s inserted. That however would be designed to mute the speaker when the headphones were plugged in, and you are describing (I think) the opposite. However see my comment below about TRRS connectors.

One thing about “modern” headphones though is that headsets (not pure headphones) sometimes have a 4-pole jack plug. It’s usually referred to as TRRS (tip, ring, ring, sleeve) whereas the three-pole one is TRS. These have the two headphone inputs on the tip and first ring, the microphone output on the second ring, and ground as usual on the sleeve. Depending on the design of the socket, the ground connecion might be going to the microphone output on one of these.

Regarding impedance, I’ve encountered these with 25 ohm headphones, and with 1K ohm headphones, both of recent design.

Alternatively, since those four pin jobbies are longer, there’s always the possibility that shoving one of those in the back might have bent something out of place?

Yes, I think there was a mechanism to disable the internal speaker when a headphone jack was plugged in, but never looked in detail at how it was actually implemented.
For my hi-fi system, the audio signal actually passed through the headphone socket (it was five pin Lin, Rin, Lout, Rout, Gnd) and putting the plug into it made contact with the pins and lifted them slightly, breaking the connection to the output side.

It seems rather peculiar to have this apparently working back to front. Sounds like something is wrong there…

I’m not sure impedance is a big issue. Headphones can be around 8-400 ohms (or more), with the typical value being “about 32”. It would mostly affect how loud the sounds are, whether it melts your ears or needs additional amplification. It’s a little more complicated (not a great idea to short circuit an output), but as a base rule of thumb that will suffice.

Note that TRRS is often wired as above, but not always. Some companies wire up differently. Some are wired to have video output instead of mic input (and many permutations), there’s even a balanced audio output version. So, probably best to stick with a traditional three pin plug. ;)

The thing beeps on Ctrl-G, doesn’t it? If it does, that must mean the internal speaker is working, surely?

Edit: that’s the third time I’ve edited it, to get “beeps”, not “boots”.

I noticed on various acorn (RISC OS era) machines in general that ‘modern’ headphones prevent the machine starting up at all when plugged in (blank screen with screeching sound)? I’m surprised the headphones don’t have the same affect on the A310 (and that they appear to enable the internal speaker!) . . .

Not sure this is the case for all Acorn Archimedes/Axxxx/RiscPC/A7000 machines. Here I have a full rack filled up with all RISC OS machine from the first Archimedes 310 up to the last Acorn system RiscPCs and A7000+. ALL of their audio outs are connected to 32 ohm jack cables that end into a special ultra low noise mixer which is then used to feed back a VGAtoHDMI converter, so all the machines audio is “switched on” all the time and with a “modern” cable and audio system.

They all boot fine and there is no crackle or issue of any sort. On top of that, I also have professional audio cards for the RiscPCs (the famous DMI 50 with the Yamaha XGR 50 synth module) and they are also connected to the same ultra low noise mixer. To reduce even more any possible interference, I use a pure sine UPS to power every single piece of equipment in that rack and everything is grounded correctly (given that the analogue side is very sensitive to any electrical interference).

The rack is usually paced in the other side of my flat (but it has wheels, so when I need to do maintenance and stuff I just move it close to my lab/recording studio), so the audio signal also has to travel through an RJ45 cable from that side of the flat till my lab/recording studio, and again no audio degradation, crackle or any audio issue whatsoever.

So, I am thinking you have a set of machines with some audio issues. Unfortunately, without inspecting the machines or testing them, I cannot tell you what the problems could be, but I know few guys (which are regulars on stardot btw) who are serious wizards with all old Acorn machine issues etc. So, again, it’s worth to try to put all your issues in a post there (if you haven’t done it yet).

best of luck!

It’s just possible that the headphone jack socket includes a switch, and is used for this.

I remember old Usenet discussions about the headphone jacks and I am fairly sure that the “headphone jack with a switch” which automatically disables the internal speaker when a headphone was attached was used in all machines starting with the A3000. Before that (A3xx/A4xx/A540/Rxxx), everything was independent and therefore fully software-controllable via speaker on/off.

Thinking further about it, maybe it was the Risc PC that started this? Unfortunately all my old machines are currently in various states of non-operation, so I cannot quickly check.

Alternatively, since those four pin jobbies are longer, there’s always the possibility that shoving one of those in the back might have bent something out of place?

Are there ones which are longer? All of my 4-pin ones are the same length & shape as the 3-pin ones. They’re designed to be backwards-compatible with 3-pin sockets.

Okay, yup. Just dug out my headset and compared it with the plug on headphones. They’re the same length. Could have sworn TRRS was slightly longer…

My bad. 🤦

BTW, wiki tells me there’s now a TRRRS.

BTW, wiki tells me there’s now a TRRRS.

Yeah, more “shorts” to allow controls from headphones… ;)

@ Steffen

Thinking further about it, maybe it was the Risc PC that started this? Unfortunately all my old machines are currently in various states of non-operation, so I cannot quickly check.

Yes, the RiscPC/A7000 have the headphone auto-switch, so no sound from the speaker if you plug the headphone jack in, while all the previous machines (up to the A5000 included) and then also the Iyonix still emit sounds from the speaker even when the headphone jack is plugged in.

Now given that, when one plugs in the headphone jack, there is a temporary short of both audio devices Q9 and Q11. So, it was advisable, not to plug/unplug the headphones when the computer was on (if Jonathan’s Archie comes from a school I seriously doubt this was followed ;) ).

So, something is not ok with both the BC239 on both or either the Q9 or Q11. However, what I am not sure of is why if Jonathan plugs in the jack then he can hear a mono signal, so I am guessing the BC239 on Q9 is not fully damaged or something?!?!? (not sure, so cannot say, hence this is a wild guess).

I also remember that sometimes audio issues came from bad PSUs and, once, even from a bad MEMC1a that wasn’t fitted correctly or that was damaged (unfortunately I do not recall well, it has passed so much time since I worked for computer shops).

However, what I am not sure of is why if Jonathan plugs in the jack then he can hear a mono signal

Aha… Right. Can you play some music with singing?

If you can hear the music more or less okay (only in mono), but the vocals sound like they’re coming from the bottom of a well, then what you’re actually hearing is the audio coming from the L and R channels, but with no ground.

Experience? I’ve worn out many pairs of headphones in my life. If the ground breaks, this is the result. Mono sound and faraway vocals.

@All

Thanks everyone for input!

To clarify, when I connect the A310 to an AKF11 monitor (via SCART to VGA lead with separate audio lead with TRS ‘three pole’ audio jack plugged into headphone socket at back of machine) the sound coming from the monitor speakers appears to be perfect (left and right stereo as demonstrated nicely by ‘Small’ game). I assume this means the audio components on the board (and the headphone socket?) are working correctly?! (The internal speaker is silent during monitor use AND when the headphone socket is otherwise empty.)

When I plug in TRRS (playstation) headphones the internal speaker ‘appears’ to be left speaker and the right headphone ‘appears’ to be right speaker (it’s a bit hard to tell with different volumes/distances).

I’m going to try older TRS headphones next to see if there is a difference.

BTW I would never attempt to plug/unplug anything into any acorn machine with power on (learnt that the hard way).

I have a multimeter. Does anyone have any idea what continuity I should be getting from parts of the headphone socket to other components on the board? BUT if the monitor speakers work normally perhaps that won’t show anything?

I still intend posting this issue on stardot at some point but I feel like there is some progress here (apologies if I am not following protocol correctly; I am new to this forum which BTW I am finding very helpful so thank you all). Not sure it’s possible to upload photos of the board on this forum so stardot would be better for that too . . .

I have a feeling the board was modified for school use to stop pupils being disruptive so I hope to reverse the mod (if I can work out what it is) to get the internal speaker working ‘normally’ again.

UPDATE

When I tried TRS (three pole) headphones the A310 internal speaker was silent! What might that explain?!

I’ve repeated the experiment on my (as far as I know normally working) A410/1 and I have the following different results (assuming the A310 and A410 can be compared in this way?)

A410/1: When AKF11 monitor with audio lead is connected, the internal speaker plays normally (as well as the monitor LEFT/RIGHT speakers). I guess that’s where *SPEAKER OFF can come in handy (to silence the internal speaker I mean).

When TRS three pole headphones are plugged in to headphone socket I get perfect left/right stereo in headphone speakers AND internal speaker works normally.

When TRRS four pole headphones are plugged in to headphone socket I get perfect (LOUD!) left/right stereo in headphone speakers AND internal speaker works normally.

So something about the TRRS four pole jack is enabling the A310 internal speaker?!

Is anyone able to repeat the above experiment on a normally working A310 (perhaps it functions differently after all!)?

BTW (@ Rick) I’m not able to test if the A310 (TRRS headphone) right speaker output is mono or not (I don’t have a cd rom/interface to play cds on the A310 I mean)

P.S. The TRRS headphones pick up a fair amount of interference on both machines . . .

P.P.S. The A310 has an econet board (the A410 does not).

I’m thinking that it may not be any easy fix and I need to send the A310 away to be looked at by an electronics engineer (it’s probably the PSU as Paolo suggests or dry caps on the board) . . . My attempt to replace C1, C2 and C9 components on the PSU has not revived this poor Archie.