Bringing the Phones Back to Life (TTYs, Rotary Phones, and a Magneto Autopsy)
Today was one of those days where a small project turned into a full-blown exploration of old telephone technology. By the end of it, I had a working TTY printer again, organized wiring across my phone network, ringing bells that hadn’t rung in decades, and grease all over my hands from dismantling a century-old magneto.
And honestly, it was a lot of fun.
A Surprise: A Third TTY Machine That Works
The day started with the arrival of a third TTY machine I had purchased for $20, strictly for parts, which included a few rolls of paper but was missing a few things, like the printer cover and plug. The seller said it would say “No Power” when powered on. My plan was to salvage components and repair the unit that recently stopped working for the exhibit.
Out of curiosity, I connected one of my spare power adapters.
To my surprise, the machine powered on like normal. Perhaps the seller was simply using batteries to test it.
Encouraged, I fed paper into the printer mechanism and called another machine. The printer sprang to life and printed cleanly.
Now I have a working TTY that can produce printed output again. That’s perfect for the exhibit—visitors will be able to walk away with a tangible printed message, which makes the experience much more memorable. Now if only I can get the other printer working…
Getting the Wiring Organized
With the excitement of that discovery out of the way, I turned to the growing nest of wiring around the exhibit.
I spent some time labeling everything:
- Ten phone cords now have an extension number on both ends
- Monday night, the Grandstream HT818 ATA devices were labeled with:
- their IP address
- all extension numbers associated with each port
This will make troubleshooting dramatically easier later and make the system easier to explain when people ask how everything works and how to call other phones. Ten more cords are on order for the second 8-port device.
The Rotary Phone That Wouldn’t Ring
Next, I turned my attention to an old metal rotary phone.
The phone could already:
- get a dial tone
- make calls
- receive calls
But it would not ring.
Unlike my other rotary phones, this one had three exposed wires with fork terminals:
- yellow
- red
- black
To experiment, I bought:
- a phone jack box with four screw terminals
- a pack of 0.47 µF polypropylene film capacitors
These capacitors are commonly used to connect the ringer circuit to the talk line in old telephones.
I tried several wiring configurations but could not get the phone to ring.
Eventually I opened the phone.
Inside, I found a plastic component labeled “Stromberg Carlson” with model number 200595-B. I had assumed the bells were inside the housing—but after looking up the model number, I realized something important:
This phone never had bells. This was just a control box. The bells would have been outside of the box in the empty area on the plate.
It was likely designed to be used with a separate ringer box.
Slightly dismayed, I put everything back together.
Rediscovering the Candlestick Phone Ringer Box
Next I looked at the ringer box that came with my candlestick phone.
The candlestick phone itself was interesting—it had an old cloth-covered cable that had been adapted to an RJ11 jack at some point. The phone clearly survived changes in wiring standards.
But there was no connection at all to the ringer box.
Testing confirmed it behaved just like the metal rotary phone:
- it could dial out
- it could answer calls
- but it would not ring
So I turned my attention to the ringer box.
I wired:
- black and red wires to the L1 and L2 terminals
- connected them to an RJ11 breakout terminal
- placed a capacitor between the red wire and the terminal
When I tried ringing it, almost nothing happened. The coils are rated at 500 ohms.
I had already configured the Grandstream HT818 ATA devices to supply higher voltage for ringing, but it still wasn’t enough.
Without the capacitor, the line would immediately pick up with a single clink of the bell.
So I added another capacitor in parallel and dialed again.
Bingo.
The bell rang—and loudly.
Experimenting with the Crank Phone
Encouraged, I tried the same idea with my hand-crank phone.
This phone’s ringer coils are rated at 1250 ohms instead of 500, which allows more phones on the same line but requires higher ringing voltage.
I added capacitors in parallel again.
- four capacitors started producing movement
- five capacitors caused the phone to register as off-hook
There simply wasn’t enough voltage from the ATA to drive the bells properly.
Testing the Magneto
Next I tried connecting the ringer box to the crank phone.
When I turned the crank on the crank phone, the ringer box rang—much louder than before.
But when I tried cranking the ringer box itself, almost nothing happened.
At most I could measure about 5 volts AC.
Something was clearly wrong.
Taking the Magneto Apart
It was time for surgery.
I started by removing a screw from the bottom that held the magneto in place, but it wouldn’t budge.
Eventually I discovered:
- two large screws on the back
- six hidden screws along the wooden backboard
- a decorative plate around the crank
After removing all of these, I was finally able to separate the backboard and work the magneto free.
Even outside the case, the crank was extremely difficult to turn.
Cleaning and Lubricating the Mechanism
I began applying Traxxas 30wt silicone shock oil, intended for RC projects. I had it for my LEGO Great Ball Contraptions.
At first, it didn’t help much.
Then I noticed a small opening that looked like a set screw port, packed with what looked like cloth.
After removing the material and digging deeper, I realized what it really was:
A century of grease and dust packed into an oil port.
Once I cleared it and added oil properly, the gears began loosening up.
Reassembling the Magneto
To get better access, I removed the three large magnets.
When reinstalling them I realized the original orientation had been wrong.
The magnets need to alternate polarity, and there were small horizontal dots on each magnet marking one pole.
After correcting the orientation, I spun the crank and measured the output.
The magneto was now producing about 27 volts AC.
That’s still far below the 80–100 volts typically expected, but I noticed the magnets themselves were fairly weak.
I considered fixing them with neodymium magnets to improve the magnetism, but it was getting late, and I didn’t have any on hand.
Reassembly and Success
After reassembling the box, the difference was immediate.
The crank now turned smoothly and easily.
Earlier in the day it was difficult to even make a single rotation.
Now I could spin it rapidly.
And the bell rang.
One bell still wasn’t being struck properly, so I adjusted the bell position and hammer alignment, which increased the volume significantly once both bells were working together.
Progress
By the end of the day:
- I unexpectedly gained a working TTY printer
- All phone wiring is fully labeled and organized
- A ringer box now rings reliably
- A magneto generator has been cleaned, lubricated, and partially restored
And perhaps most surprisingly, I spent part of the day disassembling and repairing a magneto generator—something I never imagined I’d be doing.
My hands were dirty, but I had done something tangible and learned a lot in the process.
And the exhibit is getting closer to coming alive.
