[ARC5] More on Bashless T-17 Microphone Resurrection

[Richard Knoppow]

     Carbon microphones are vulnerable to moisture and to "welding" from 
excessive current.  Baking them will sometimes undo the effects of 
moisture but nothing will help those that have welded and packed. 
Sometimes its possible to put new carbon from old telephone transmitters 
in.  I've had a little success with this but it does not always work. 
Baking must be done for several hours at around 150F.
     I think these mics (the T-17) were not designed with long life in 
mind unlike the elements in Bell System telephones which were pretty 
well protected against moisture and operated where current was 
controlled.  Carbon mics are also position sensitive, which used to 
raise hob with early telephones and is the reason shaking them was 
sometimes necessary.  The element used in the Bell System series 500 
phone was designed to be as independent of position as possible.  It is 
one of the most thoroughly engineered pieces of acoustical equipment 
around (along with the matching receiver).

On 5/31/2016 11:34 PM, David Stinson wrote:
> So work continues on "resurrecting" stone-deaf WWII T-17 microphones
> with little "booster" circuits.  I have 6 or 7 T-17s.  One 1942 contract
> stock T-17 will adequately modulate a WWII radio
> (yes; I did double-check and it is actually stock, LOL). The others are
> deaf as a post.  Others have had luck
> whacking them around to "wake them up," but I've beaten, bounced,
> bashed, baked and battered T-17
> elements for years. If I've ever gotten one to work properly that way, I
> don't remember it.
> So far, I've been working-over mikes with missing or unrecoverable
> elements (except the candle stick), using at first an electret element
> and later using a simple transistor "current booster" to amplify
> the currents through the weak carbon elements.
> Time to turn to a "stock" T-17 element and see what can be done.
>
> The subject T-17 is a stock 1944 contract with the black plastic case.
> Used with the BC-669, held right
> to my lips and me saying "Foooooooore," this was the
> modulation waveform, which is typical of most stock T-17 mics today:
> https://goo.gl/photos/ZeYyD6tjVKTvcDNn9
>
> We've investigated and tested two very simple circuits for "boosting"
> weak carbon elements and both have worked well.   See my previous posts
> in this thread.  Elements that
> measure low resistance, 60-250 Ohms or so, should be installed in the
> Emitter lead of the NPN
> transistor.  Those that measure high resistance-
> 2K Ohms or more- work best installed in the
> Base lead where they can be a large fraction
> of the total bias voltage divider resistance.
>
> I removed and tested the stock T-17 element with an Ohmmeter. It read
> around 1K- too much
> to go into the Emitter lead so I installed it in the Base lead.  Here is
> the circuit used to "boost"
> this element:
> https://goo.gl/photos/NSqmQrmimRvV9ySx5
>
> Important:  Note that the circuit above is almost
> the same one we used back on the 29th to boost
> the large "candle stick microphone" element.
> But there is an important difference.
> The carbon element in the Base lead
> is part of the resistive voltage divider coming from
> the Collector, bringing a positive bias voltage to turn the transistor
> "on."  The circuit works because
> the element acts as a variable resistor, varying that
> bias current and thus controlling the larger current flowing
> in the Collector-Emitter circuit, much like the small
> Grid current in a Triode controls the larger Plate current.
>
> Because the condition of these weak carbon elements
> varies widely, the resistor "R" must be selected for the individual
> element and for the transistor you use.
> The difference can be dramatic.  For instance- the resistor for the
> "candle stick" mike
> which used a lower-gain 2N2222 transistor (beta tested 70) was 4.7K
> Ohms.   This T-17 element and a transistor with a beta tested at 110
> required
> a resistor of 220K Ohms.   The idea is to use as high a value as
> possible and strike a balance between
> providing adequate Base current and loading-down the microphone bias
> voltage provided by the radio.
> The easiest way to do it is to use a pot and tweak for
> best response and output.
>
> With the little booster helping the original T-17
> element, this is what the "Fooooore" modulation
> envelope looks like now:
> https://goo.gl/photos/rUnhMAEixfyAx4Gp8
>
> As with the earlier work, I prototyped the booster on
> a board and mounted it behind the T-17 element:
> https://goo.gl/photos/w6QUp3vy3C5rwZEdA
>
> Boosting the carbon elements has been going
> so well, I've decided to set the electret subber
> aside for now.  I get excellent audio reports and I was never a fan of
> electrets anyway- too noisy.
>
> As I mentioned in a previous part of this thread- the transistors I'm
> using are unmarked NPNs that test with a beta of
> about 110.  If anyone can suggest a common
> number transistor with at least this beta
> for this circuit I'd be grateful.
>
> GL OM ES 73 DE Dave AB5S
>
> ______________________________________________________________
> ARC5 mailing list
> Home: http://mailman.qth.net/mailman/listinfo/arc5
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:ARC5 at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>

-- 
Richard Knoppow
1oldlens1 at ix.netcom.com
WB6KBL