247A KTU Touchtone Adapter

[Dave F]

247A KTU

In the late 1950s, as Touchtone was being developed, one issue that had to be addressed was how the new telephones would be able to work on the rotary-dial intercoms of 1A1 Key Systems.  In 1960, Bell Labs designed a prototype Touchtone Adapter, designated F53889, for that specific purpose.  This adapter would decode the tones made by the new push button dials and convert them into relay contact closures which could select intercom stations just like a rotary 207C KTU did.  After testing and some substantial modifications, in 1962 the new Touchtone Adapter, now renamed 247A KTU, was added to the inventory of 1A1 hardware.  In 1963 Touchtone service was just becoming available in a few lucky locations, and 247As started magically appearing in Key System relay racks.

To anybody enamored by 1A1 technology (myself included) the 247A was  totally awesome.  I remember seeing a few installations and was amazed by how much "new" technology went into this single piece of telephone equipment which most users would never see or appreciate.

However, it soon became apparent that there was trouble lurking in paradise.  The 247A was huge and it took up a lot of rack space.  In many smaller 1A1 installations it was necessary to install an extra relay rack just to accommodate this one KTU.  This surely added some unwelcome labor and hardware costs; and that didn't include the cost of the 247A which must have been substantial all by itself.

The most serious problem, however, was that the 247A just didn't work very well.  Tone-decoding technology for mass production was still in its infancy, and the circuitry of the 247A was quite primitive and didn't do a particularly good job.  On the front plates of the two decoder boards there were "bias adjustments"  which had to be tweaked for every installation.  Any time a phone was removed or added to the Key System, the bias had to be readjusted or the 247A might not continue to properly respond to the tones.  Also, the detection bandwidth of the tone decoders was wide, and the KTU was prone to reacting to voices on the intercom line and interpreting them as tones.  This condition is called "talk-off", and it results in incorrect intercom stations being selected.  To help minimize this problem the 247A had a time-out circuit which would disable the tone detectors after a few seconds of being off-hook.  If you waited too long to dial your desired intercom extension, you would get a "reorder" tone and have to hang up and start all over again.

The brain-trust at Bell Labs was well aware of these deficiencies and, even as 247As were rolling off the W.E. assembly lines, its replacement, the 247B KTU, was well along in development.  1A2 was on the verge of being implemented and the new 247B utilized much of the new 1A2 technology.  Miniature relays reduced the size of the 247B, and improved tone detection circuitry eliminated the need for bias adjustments and greatly reduced the talk-off problem.

By 1966 the 247A was rated MD (Manufacture Discontinued), and so began its retreat into history as quickly as it had originally emerged.  Any time an installed 247A experienced any problems, repair personnel were instructed to remove and replace it with the new B version.  W.E. didn't really even want the old units returned, so most of the retired 247As ended up in the dump.  In fact, so few remain today that some serious long-time collectors have never even seen one.

The specimen shown in the pictures here was removed from service in about 1970.  The KTS installer who pulled it was told to toss it in the trash.  Instead, he put it in his garage where it sat for about 45 years.  His wife finally convinced him to clean out some space, and the KTU was destined for the recycle bin when I found out about it.  Just in the nick of time I saved something that I had been waiting decades to find.  It remains in the same condition as when I acquired it, jumper wires and diodes still on the rear terminal board.  If I ever do build another 1A1 system, this 247A KTU will certainly be a part of it.

Documentation:

There is no BSP on the F53889 prototype, but the schematic can be found in SD69411.

The original BSP on the 247A is:
518-410-410.  Early issues contain schematic.

Other BSPs on 247A are:
518-114-116 and
518-310-113.

Schematic of 247A can also be found in:
SD69447 and
SD69288

DF

[rcourtney]

thanks.

[SUnset2]

Thanks for posting that.  It's interesting to see what they had to do before the technology enabled the small intercom units that came later.

[5415551212]

Wow thats really interesting, thanks.
Dave do you have a 1A1 system your going to add this to?
Cheers
--Stephen

[Dave F]

Wow thats really interesting, thanks.
Dave do you have a 1A1 system your going to add this to?
Cheers
--Stephen

No, but I have all the necessary stuff packed away.  Just need the space, time, and the stamina to do the work!

DF

[5415551212]

Tell me abut it I have been putting together a 1A2 system for about 6 years now, I am on the home stretch now.
I'll make a post about it when its done.
I think only one forum member has a working 1A1 system thats John (Babybearjs).
If you need any un-terminated 25 pair cable send me a PM.

[Dave F]

.........If you need any un-terminated 25 pair cable send me a PM.

Thank you.  I'll remember that for future reference.

DF