"The phone is a remarkably complex, simple device,
and very rarely ever needs repairs, once you fix them." - Dan/Panther

Main Menu

Is AE sealed transmitter capsule repair possible?

Started by KaiserFrazer67, April 15, 2017, 07:20:58 PM

Previous topic - Next topic


Quote from: TelePlay on April 16, 2017, 05:22:33 PM
Your AE transmitter is similar to my toasted 4-35 SC except the center post on the SC is not like the AE. It seems the SC transmitter was assembled by crimping a ring around the front of the transmitter, locking the front grill to the body. There is not other way to take the SC apart and from the photo of your AE, it seems to have the same type of ring crimped around the front. When shake the SC, I can not hear the carbon inside.
Mine is indeed crimped around the front grille too, as well as crimped around the base of the center post.  If I were going to open it up, it would probably be best to cut through the front crimp and the entire outer contact ring with a Dremel tool and take the outer ring off, with the capsule clamped in a vise by the center post grille-up so nothing spills out.  Then I can take off the grille and whatever's underneath it, and dump the granules out into a glass jar.  At least I'd be able to salvage something worthwhile to someone, rather than tossing the whole thing into the garbage.  I never thought about recycling the carbon granules to use in restoring early-type transmitters...  good idea!
-Tom from Oakfield, Wisconsin --  My CO CLLI & switch: OKFDWIXADS0--GTD-5 EAX

"Problems are merely opportunities in workclothes." -Henry J. Kaiser


Quote from: TelePlay on April 16, 2017, 12:03:18 AM
A more powerful generator at a lower frequency that is capable of being operated in air might do a very good job of vibrating the metal capsule and it's carbon contents.

To follow up on that thought, I did a bit of experimenting this morning. I used a small, 1" diameter tweeter. Small speaker but big enough to set my toasted SC transmitter face down on the speaker facing up on my audio work bench (first picture below). Used that speaker so I could power it directly off of my headphone out port, no amp.

I played a sine wave log increasing frequency file from 0 to 5,000 Hz through the speaker recording what the transmitter picked up. All parameters were kept constant. The only thing changed was the "condition" of the carbon granules.

SC 01 below is the base line, what the transmitter was putting out before trying to break up the carbon granules.

SC 02 repeats the scan but after shaking the transmitter.

SC 03 repeats the scan after creating a different sound file of 15,000 Hz and played that for 30 seconds to "break up" the granules. Similar to SC 02 but the amplitude was greater showing some improvement in transmitter output.

SC 04 repeats the scan but after playing a 3 track file for 5 minutes to "condition" the transmitter more thoroughly. The tracks contained a 10,000 Hz, 15,000 Hz and 20,000 Hz constant or continuous sine waves passing those frequencies into the transmitter's carbon capsule. The after treatment scan was very similar to SC 03.

SC 05 is a repeat of scan SC 04 except the transmitter was physically shaken for a few seconds and the while horizontal moved side to side to "even out" the granules in the capsule. That got rid of a lot of noise probably due to the shaken carbon granules settling in SC 04 as sound waves vibrated them during the sweep scan. I've seen this same type of noise by recording and doing nothing more than move the transmitter or tilting it a bit so the carbon granules would "shift" causing what appears to be noise or static. Or, fluffed up carbon granules settling by themselves for what  ever reason causes noise. And, yes, the same thing would be happening when holding a moving a candlestick during normal use but the S/N ratio would put the "normal" noise in the background as long as the capsule was picking up voice at it's designed levels.




SC 06 is nothing more that a re-record of the past set up, nothing was shaken or touched, just re-recorded. What is seen is a small decrease in amplitude.

SC 07 is a re-recording with nothing shaken or touched. The decrease in amplitude is more visible in this file. What appeared to be happening was the carbon granules were settling back together (starting to pack) thereby decreasing amplitude. The sound waves were doing the "settling."

SC 08 is another re-recording of the transmitter without doing anything other that playing the sound file again. There is a marked decrease in amplitude.

SC 09 shows what happens when the transmitter is shaken up again and after shaking, moving it from side to side to "even out" the granules before playing the file and recording it with the transmitter.

So, even after being "broken up," the granules seem to resettle quickly resulting in decreased output. This has to be something to do with the "quality" of the granules themselves and/or the "atmosphere" within the capsule.

Next step is to put a larger "tweeter" on an amp and really shake the carbon granules through direct vibration, basically creating an ultrasonic affect on the carbon within the capsule using air as the sound wave medium. I have a spare amp that puts out about 35 watts which when used with a larger tweeter should produce an exciting time for the carbon granules, at least enough to fully break up whatever has bonded together over time.

If someone else want to try and duplicate this to see if they get the same results, that would be great.


Being curious about the signal degradation/decrease with use, I took the SC transmitter under test (which was once toast - no response), shook it well, moved it from side to side to "level" the carbon granules and recorded a response sweep from 0 to 10,000 Hz.

01)  Record the base line sweep
02)  Re-record with no changes
03)  Re-record with no changes
04)  Re-record with no changes
05)  Record after exposing the transmitter to 10, 15 and 20 kHz over 5 minutes, no shaking.
06)  Shook the transmitter well and moved it side to side to "level" the carbon granules.

While the 01 track shows good transmitter response, the amplitude decreases with each subsequent recording (02 to 04) which would seem to indicate the carbon was being settled by the sound waves into a state where they would not produce the same response level as track 01.

Without touching the set up, I subjected the transmitter to 10, 15 and 20 kHz sound for 5 minutes. Then recorded track 05 which shows some improvement probably caused by the vibration of the sound waves "breaking up" the settled carbon granules.

Finally, I shook the well and moved it from side to side to level the granules and the recording, track 06, is once again similar to track 01.

Not shown are the following tracks of decreasing response with several re-recordings as in tracks 02 to 04.

This transmitter seems to have been under water for some period of time. The exterior dirt and corrosion was removed but nothing can be done about the carbon capsule short of opening it up. There is a difference between carbon and activated carbon which has been purged of moisture and any other chemicals in the air which are absorbed by carbon. Activated carbon is a very good filter, removes stuff from air and water, but becomes "full" at some point and has to be heated to expel whatever has been absorbed. I do not know what telephone companies used, or how they put it into the capsules which I would assume were air tight to keep stuff out. As such, a carbon mic would work well until the carbon granules became compromised in some way. Even if clumped and then broken up by vibration, as in this case, it seems the compromised carbon is quick to settle back into a position that no longer responds well to sound waves.

Given that I can not hear above 5,600 Hz anymore, I do not know what these sound like above that. However, any sharp spikes seen in 06 under 3,000 Hz are very easily heard, loud ticks or pops most likely from carbon settling.

Just providing what I've measured (and in my case seen) with respect to this one transmitter. If I had a good heated vacuum chamber to both heat up the capsule to 120 F under reduced pressure to suck out any moisture that may be in the carbon, that would be an interesting experiment but I don't have such, so won't be trying that.

I have one more thing to try next week and will post how that goes.