On Sun, 27 Jul 2014, Stuart Longland wrote:

Hi,

Silly question? I've been experimenting with a radio station on the
bicycle for HF operation and one problem I've been facing is that of RF
feedback getting into the microphone.

I've tried a few things to try and eliminate the RF. The set is a Yaesu
FT-857D, and I use runs of CAT5 cabling to hook the head unit and my
headset up to the radio which is mounted in a motorcycle topbox behind
me. I've ensured that the microphone + and - signals are on their own
dedicated pair in the CAT5, as are the +5V and GND signals.

The microphone biasing is done near the head unit of the radio, so maybe
1.5m away from the radio, and a short lead then plugs into the
(helmet-embedded) headset. My biasing circuit looks like this: (please
excuse the ASCII art)
.-----. .---------------~~----.
+5v ---o------------o----| 1k |---o ..100nF |
| | '-----' '---||--o---o Mic + +| .
----- ----- '' --- ( )| Electret
.-'-. 100uF ----- 100nF ,. --- 100nF -| ' Insert
| | .-----. .---||--o---o Mic - | (headset)
0v----o------------o----| 1k |---o ''100nF |
'-----' '---------------~~----'

The capacitors are all (with the exception of the 100uF electrolytic)
small monolithic capacitors:
http://www.jaycar.com.au/productView.asp?ID=RC5490

My intent is that the capacitors across the 5V rail should suppress the
RF on the 5V rail, and the capacitor across the output of the bias
circuit should suppress any common-mode RF on the balanced feed to the
radio. I think there's one across the microphone in the headset too.

Originally I contemplated a phantom-power arrangement, but I'd need to
sit down and design the bias circuit as most of my stuff is in the 3-5V
range, not the 12-48V range that phantom-power normally assumes.

I note I do not get the RF feedback if I use the stock Yaesu handmic
that came with the set, which is a 600ohm dynamic microphone.

Currently I use an electret insert in the headset, one out of the junk
box often salvaged from old mobile phone "headsets" (those annoying
earphone things with an inline microphone).

The lack of problems with the Yaesu handmic got me thinking: I have
amassed a large quantity of earphone speakers from those mobile phone
headsets (I hate the bloody things: can't stand putting anything *in* my
ears).

These are speakers with dimensions not too dissimilar to small dynamic
elements with an impedance of 32ohms. Question is, how well would these
work as microphones? Has anyone tried this and can share any insights?

IN the old days, endless cheap 100mW walkie talkies would use the speaker
for the microphone on transmit. For that matter, endless intercom systems
used the same speaker as a speaker and as a microphone.

Endless articles in the magazines showed taking a speaker of some sort and
using it as a microphone.

Of course, those were speakers, most seemed at least 2".

An earphone may not offer the same level of sound collection that a
speaker with a larger cone allows, you may have to play with things. I
remember taking cheap dynamic earphones and taking the bit that went in
your ear off, and using that as a contact microphone for various things.

The other issue is the output may not be so great, mostly because the
speaker is arranged for sound reproduction, not sound pickup. It may need
a preamp, depending on the actual output voltage. It may also need
impedance matching, which would also help boost the output, use an output
transformer in reverse so the speaker winding is fed from the speaker, and
the now secondary winding will have more voltage to feed the higher input
impedance of the rig.

Michael

Michael Black wrote in
news:alpine.LNX.2.02.1407271413320.23911@darkstar. example.org:

An earphone may not offer the same level of sound collection that a
speaker with a larger cone allows, you may have to play with things. I
remember taking cheap dynamic earphones and taking the bit that went in
your ear off, and using that as a contact microphone for various things.


Could be so. When I asswered I overlooked the detail of headphone type. The
little bud types might not be so good, but the small on-ear types that came
out when cheap Walkmans werre new, are a very good candidate, because the
construction of those is almost identical with that of many cheap dynamic
mics, and the sound fidelity is also very good with a diaphram about 0.75''
to 1'' wide. Mylar too, so no degrading with humidity.

I haven't tried to work out the implications of matching impedance for gain,
but these days it is likely easier not to do it, just use a high resistance
input with low noise and high gain. Cheap op-amps that will do it are easily
had. Dynamic mics with transformer matching might work but even if immune to
RF pickup they will catch magnetic fields as if intended to do so!

I also wonder if fully balanced feed is needed in either case, dynamic or
electret. If one wire is firmly grounded, and is part of a twisted pair with
the wire that carries the DC feed and the AC signal out, then it ought to
cancel out any incoming HF anyway. It's possible that trying to make it fully
balanced might make it more vulnerable, not less, because it has no firm
ground on either pole. The main thing is to get the gain up as close to the
electret as possible, and that will want very small parts. Transistor rather
than transformer..

Lostgallifreyan wrote in
:

I also wonder if fully balanced feed is needed in either case, dynamic
or electret. If one wire is firmly grounded, and is part of a twisted
pair with the wire that carries the DC feed and the AC signal out, then
it ought to cancel out any incoming HF anyway. It's possible that trying
to make it fully balanced might make it more vulnerable, not less,
because it has no firm ground on either pole. The main thing is to get
the gain up as close to the electret as possible, and that will want
very small parts. Transistor rather than transformer..


Bit more thought on that... If the circit to be fed by this has two grounds,
one for frame, the other for local AF input signals, then instead of CAT5
pairs, use a twisted pair cheap 3mm thick cable with a screen. Screen to
frame ground, twisted pair for AF line and signal ground. That should shunt
HF pickup to ground in the equipment where it is likely already done, and
allow at least a metre of signal line to work well.

On 28/07/14 04:18, Michael Black wrote:
IN the old days, endless cheap 100mW walkie talkies would use the
speaker for the microphone on transmit. For that matter, endless
intercom systems used the same speaker as a speaker and as a microphone.

I do remember those, in fact I've got one gutted somewhere. Had a
crystal for 27.145MHz.

I once tried wiring up an electret element, not knowing there was a
difference, and was disappointed when it didn't work.

An earphone may not offer the same level of sound collection that a
speaker with a larger cone allows, you may have to play with things. I
remember taking cheap dynamic earphones and taking the bit that went in
your ear off, and using that as a contact microphone for various things.

In the interest of science, I gave it a shot just then. With a 100nF
capacitor in series to block the DC, I wired it to a DIN5 plug (all my
radios have been set up with adaptors to DIN5 headset jacks) and tried it.

It did work, but without any amplification or impedance matching, the
modulation is well down. I might try winding a small transformer and
see what that does.

Stuart Longland wrote:
On 28/07/14 04:18, Michael Black wrote:
IN the old days, endless cheap 100mW walkie talkies would use the
speaker for the microphone on transmit. For that matter, endless
intercom systems used the same speaker as a speaker and as a microphone.

I do remember those, in fact I've got one gutted somewhere. Had a
crystal for 27.145MHz.

I once tried wiring up an electret element, not knowing there was a
difference, and was disappointed when it didn't work.

An earphone may not offer the same level of sound collection that a
speaker with a larger cone allows, you may have to play with things. I
remember taking cheap dynamic earphones and taking the bit that went in
your ear off, and using that as a contact microphone for various things.

In the interest of science, I gave it a shot just then. With a 100nF
capacitor in series to block the DC, I wired it to a DIN5 plug (all my
radios have been set up with adaptors to DIN5 headset jacks) and tried it.

It did work, but without any amplification or impedance matching, the
modulation is well down. I might try winding a small transformer and
see what that does.

In the days when speakers were used as microphones in walkie talkies
and intercoms, it also was quite popular to have transformers between
the final stage transistors and the speaker.

The circuits of such devices were often very cleverly designed, re-using
many components between receive and transmit (using a multipole switch).

It is quite likely that the output transformer was used as a step-up
transformer while the speaker was used as microphone.

On Sat, 2 Aug 2014, Rob wrote:

Stuart Longland wrote:
On 28/07/14 04:18, Michael Black wrote:
IN the old days, endless cheap 100mW walkie talkies would use the
speaker for the microphone on transmit. For that matter, endless
intercom systems used the same speaker as a speaker and as a microphone.

I do remember those, in fact I've got one gutted somewhere. Had a
crystal for 27.145MHz.

I once tried wiring up an electret element, not knowing there was a
difference, and was disappointed when it didn't work.

An earphone may not offer the same level of sound collection that a
speaker with a larger cone allows, you may have to play with things. I
remember taking cheap dynamic earphones and taking the bit that went in
your ear off, and using that as a contact microphone for various things.

In the interest of science, I gave it a shot just then. With a 100nF
capacitor in series to block the DC, I wired it to a DIN5 plug (all my
radios have been set up with adaptors to DIN5 headset jacks) and tried it.

It did work, but without any amplification or impedance matching, the
modulation is well down. I might try winding a small transformer and
see what that does.

In the days when speakers were used as microphones in walkie talkies
and intercoms, it also was quite popular to have transformers between
the final stage transistors and the speaker.

The circuits of such devices were often very cleverly designed, re-using
many components between receive and transmit (using a multipole switch).

They were really complicated switches, for the sake of a few transistors.

It is quite likely that the output transformer was used as a step-up
transformer while the speaker was used as microphone.

That's what I would have thought, but I recall articles about modifying
those cheap walkie talkies and they added transformers to step up the
output from the speaker on transmit.

You're right, in that era, the audio amplifiers were using an output
transformer. But it wasn't just to match impedance, it was part of the
amplifier, and thus needed on transmit too.

Michael

On 03/08/14 11:10, Michael Black wrote:
The circuits of such devices were often very cleverly designed, re-using
many components between receive and transmit (using a multipole switch).

They were really complicated switches, for the sake of a few transistors.

Yep. This one I gutted, I recall de-soldering the switch and then
reverse-engineering the pinout so I could replace it with a relay, which
I did. A 4-pole double-throw relay IIRC.

I had some hair-brained idea (this was when I was in primary school) to
hook the thing up to the PC-speaker output of the computer (since I knew
how to make tones) and then use some circuit interfaced to the game port
(since I knew how to read the switches on those) and try to send data
using AFSK.

Exactly what data rate I'd achieve, given the whole lot would be
implemented in QBasic I have no idea. I doubt it'd outpace PSK31.

Not that I knew what AFSK was back then. Or that to do what I wanted to
do, I really should have a radio license which I didn't back then.

Stuart Longland wrote:
On 03/08/14 11:10, Michael Black wrote:
The circuits of such devices were often very cleverly designed, re-using
many components between receive and transmit (using a multipole switch).

They were really complicated switches, for the sake of a few transistors.

Yep. This one I gutted, I recall de-soldering the switch and then
reverse-engineering the pinout so I could replace it with a relay, which
I did. A 4-pole double-throw relay IIRC.

Not only that the switch has many poles, the circuit is often very tricky.
It is not a receiver and a transmitter with a switch to toggle the
power, antenna and speaker/mike to connect to one of them, no it is
a blob of electronics that morphs between being a transmitter and being
a receiver when the PTT switch is switched over.

In those days I sometimes tried drawing the schematic by looking at the
PCB traces and components, and it is very difficult to draw a schematic
that makes any sense...

It is completely contrary to the electronics world today, where one would
prefer having a thousand extra transistors to save a single mechanical
component (like an extra pole on the switch). The times have changed...

On Sun, 3 Aug 2014, Rob wrote:

Stuart Longland wrote:
On 03/08/14 11:10, Michael Black wrote:
The circuits of such devices were often very cleverly designed, re-using
many components between receive and transmit (using a multipole switch).

They were really complicated switches, for the sake of a few transistors.

Yep. This one I gutted, I recall de-soldering the switch and then
reverse-engineering the pinout so I could replace it with a relay, which
I did. A 4-pole double-throw relay IIRC.

Not only that the switch has many poles, the circuit is often very tricky.
It is not a receiver and a transmitter with a switch to toggle the
power, antenna and speaker/mike to connect to one of them, no it is
a blob of electronics that morphs between being a transmitter and being
a receiver when the PTT switch is switched over.

Yes, the switch would be so much simpler if they were just switching audio
and power.

I get the feeling these were the solid state equivalent of the one tube
transcievers used to homestead the higher bands. They were a modulated
oscillator on transmit, a superregenerative receiver on receive, and a
common audio amplifier. There the space and cost of a tube meant they
switch it between the two functions, but since it was a modulated
oscillator, it was a simpler arrangement than switching between a
superregen and a crystal controlled transmitter.

Those single tube transceivers were certainly simple, and got people onto
the higher bands. ONce a band got busy, there'd be a rule put in that you
had to use crystal control (or have equivalent stability) on that band.
So these rigs would start off at the "UHF" 10metre band, then move to
5metres, then up to 2.5Metres. Even fifty years ago, they were being used
on the 420MHz band. SImple and cheap, you didn't get much range, but they
helped get people on the band.

In those days I sometimes tried drawing the schematic by looking at the
PCB traces and components, and it is very difficult to draw a schematic
that makes any sense...

Expecially when you were a kid without much ability to figure out what the
switch contacts were doing. All these circuit board traces would go into
what amounted to a black box switch, crtainly beyond my skill at the time
to trace out.

It is completely contrary to the electronics world today, where one would
prefer having a thousand extra transistors to save a single mechanical
component (like an extra pole on the switch). The times have changed...

I guess it makes sense at the beginning, but transistor prices dropped
fast, yet the same scheme was used into the seventies. I assume when
cheap walkie talkies moved to 49MHz, they didn't add transistors but still
used that complicated switch (but I've never looked at a superregen 49MHz
walkie talkie). Considering that "transistor radios" at the time were
certain to tell you that they had "X transistors" you'd think the cost of
adding a transistor for transmit (and thus be able to say "four
transistors" or whatever) would increase sales enough that it would offset
th cost of the extra transistor.

It is a lesson worth repeating, adding transistors may nominall make the
circuit more complicated (and expensive), but often results in the overall
design being simpler.

Of course, once ICs came along, that took the idea to the extreme, endless
transistors in the IC, but you never see them.

Michael

On Sun, 3 Aug 2014, Stuart Longland wrote:

On 28/07/14 04:18, Michael Black wrote:
IN the old days, endless cheap 100mW walkie talkies would use the
speaker for the microphone on transmit. For that matter, endless
intercom systems used the same speaker as a speaker and as a microphone.

I do remember those, in fact I've got one gutted somewhere. Had a
crystal for 27.145MHz.

I once tried wiring up an electret element, not knowing there was a
difference, and was disappointed when it didn't work.

An earphone may not offer the same level of sound collection that a
speaker with a larger cone allows, you may have to play with things. I
remember taking cheap dynamic earphones and taking the bit that went in
your ear off, and using that as a contact microphone for various things.

In the interest of science, I gave it a shot just then. With a 100nF
capacitor in series to block the DC, I wired it to a DIN5 plug (all my
radios have been set up with adaptors to DIN5 headset jacks) and tried it.

It did work, but without any amplification or impedance matching, the
modulation is well down. I might try winding a small transformer and
see what that does.

I recall a couple of articles where people moved those cheap walkie
talkies to 10 or 6metres, and they both mentioned adding an audio
transformer to improve the audio on transmit.

Michael