TF 2100 A.F.Oscillator - UK Vintage Radio Repair and Restoration Discussion Forum

Julesomega · 23rd Jul 2021, 1:01 pm

I was delighted to pick up a TF 2100 recently: it looked like a unit I already have and it was working perfectly at switch-on. I was thinking that now I had two units I would investigate modifying the output impedance of one to about 4Ω for driving low-impedance loads at high levels, in the manner of the Farnell LFP1 which has long been on my wants list, but on opening it up I found that my two units are quite different models, the previous one being the TF2102M1

The MI catalogue for 1975 shows the TF2102 AF Oscillator (3Hz-30kHz) specified as having no amplitude bounce, making it useful for vibration testing setups. This is because a diode detector with an arbitrarily fast response is used for amplitude stabilisation rather than the usual thermistor or tungsten bulb. The output amplitude is unusually high, using a Class-B output stage with series output transistors to give 10V rms into 600ohms, or 20V emf. By comparison the catalogue describes the TF2100 as having "extremely low distortion" which was by then discontinued and supplied only in a double-unit case with a "monitoring attenuator" as the TF2000, or as the TF2005R two-tone signal source, being a triple-unit with a second oscillator unit plus the monitoring attenuator, for intermodulation testing on high quality AF equipment. I am now even more delighted with this purchase and will certainly not be modifying it in any way: the 2102 would be the better candidate for modification

Regarding the thermistor for the Wien Bridge oscillator, it is shown in the parts list as an "A54" thermistor which measures 4mm × 22mm. This is NTC, and according to an old catalogue is nominally 5kΩ at 20°C, reducing to a minimum of 80Ω when hot. The only modification I am now considering on the 2100 is to replace the rear panel O-P connector which is one of those rare "miniature co-ax connectors" with something more easily found, like an SMB

The 2102 can now be modified for the low-impedance output: it would seem prudent to replace the original Ge TO18 devices with Si TO39s with suitable heatsinks. It seems a shame that this first generation of solid-state MI signal sources is so seldom encountered: these pre-microprocessor analogue instruments are eminently restorable and of superb electronic- and mechanical- design

HECTOR63 · 23rd Jul 2021, 1:29 pm

Picture 5 !! Wow

Very good,now thats what I call a servicable item wonderful and practical to work on simply impressed

Radio Wrangler · 23rd Jul 2021, 1:29 pm

With two units you're good for two-tone tests like intermodulation distortion, or running up an SSB transmitter for PEP measurement.

It might be more versatile to leave the units original and just build a little power amp. In the extreme a box with a power amp and speaker as well as a combiner. THen you have an amplified test speaker, a low impedance driver and a combiner all of which can be exploited separately as needed.

David

Jon_G4MDC · 23rd Jul 2021, 2:14 pm

I'm not sure that co-ax output was intended to be used apart from when the oscillator was built into a frame such as in the Transmission Measuring Set. There it fed the adjacent MF Output Attenuator.

In normal standalone use wouldn't output be taken via the 4mm sockets at the front?
I think that coax connector is probably a mini Belling Lee - like a scaled down version of the 75 Ohm TV plug. Not all that hard to find and better than modifying an otherwise "mint" unit.

Julesomega · 23rd Jul 2021, 2:16 pm

I won't need to mod it if I can find a Farnell LFP1

The designers were certainly designing for servicability, maybe in response to pressure from the repairs department! Thankfully someone on the Marconi Instruments forum sent me a full scan of the manual

Diabolical Artificer · 24th Jul 2021, 7:50 am

I wonder when the fold out design first came out, the TF2331 distortion analyser used the same fold out construction, whoever thought of it is a genius. Makes assembly easier too.

Andy.

Dickie · 24th Jul 2021, 8:08 am

It appeared in 1963, and was designed by W.D. Cain, the chief mechanical engineer at the time. He was quite a free-thinker.

https://worldradiohistory.com/Archiv...ni-1963-06.pdf

It's also worth browsing the other issues at this time, as they contain details of of other instruments that crop up on the forum.

Julesomega · 24th Jul 2021, 8:44 am

Thanks for that link Richard, and hooray for Mr.Cain! The issue was devoted almost entirely to the TF2000/2100 and the completely unknown "MF" variants which extend the range to 550kHz (they were not in the later catalogues I have). It's hard to imagine what market these 600Ω instruments were intended for, and carrier telephony systems must have been into the MHz region by then

Jon_G4MDC · 24th Jul 2021, 10:03 am

The TMS I referred to earlier went to 550kHz and the oscillator was marked MF not AF.

Low capacity multiplexed links were probably still operating below 550kHz. I am thinking of units such 1+1s etc.

23rd Jul 2021, 1:01 pm	#1
Julesomega Nonode Join Date: Dec 2012 Location: Stockport, Greater Manchester, UK. Posts: 2,102	TF 2100 A.F.Oscillator I was delighted to pick up a TF 2100 recently: it looked like a unit I already have and it was working perfectly at switch-on. I was thinking that now I had two units I would investigate modifying the output impedance of one to about 4Ω for driving low-impedance loads at high levels, in the manner of the Farnell LFP1 which has long been on my wants list, but on opening it up I found that my two units are quite different models, the previous one being the TF2102M1 The MI catalogue for 1975 shows the TF2102 AF Oscillator (3Hz-30kHz) specified as having no amplitude bounce, making it useful for vibration testing setups. This is because a diode detector with an arbitrarily fast response is used for amplitude stabilisation rather than the usual thermistor or tungsten bulb. The output amplitude is unusually high, using a Class-B output stage with series output transistors to give 10V rms into 600ohms, or 20V emf. By comparison the catalogue describes the TF2100 as having "extremely low distortion" which was by then discontinued and supplied only in a double-unit case with a "monitoring attenuator" as the TF2000, or as the TF2005R two-tone signal source, being a triple-unit with a second oscillator unit plus the monitoring attenuator, for intermodulation testing on high quality AF equipment. I am now even more delighted with this purchase and will certainly not be modifying it in any way: the 2102 would be the better candidate for modification Regarding the thermistor for the Wien Bridge oscillator, it is shown in the parts list as an "A54" thermistor which measures 4mm × 22mm. This is NTC, and according to an old catalogue is nominally 5kΩ at 20°C, reducing to a minimum of 80Ω when hot. The only modification I am now considering on the 2100 is to replace the rear panel O-P connector which is one of those rare "miniature co-ax connectors" with something more easily found, like an SMB The 2102 can now be modified for the low-impedance output: it would seem prudent to replace the original Ge TO18 devices with Si TO39s with suitable heatsinks. It seems a shame that this first generation of solid-state MI signal sources is so seldom encountered: these pre-microprocessor analogue instruments are eminently restorable and of superb electronic- and mechanical- design Attached Thumbnails __________________ - Julian It's good here

23rd Jul 2021, 1:29 pm	#2
HECTOR63 Banned Join Date: Feb 2017 Location: Bedfordshire, UK. Posts: 601	Re: TF 2100 A.F.Oscillator Picture 5 !! Wow Very good,now thats what I call a servicable item wonderful and practical to work on simply impressed

23rd Jul 2021, 1:29 pm	#3
Radio Wrangler Moderator Join Date: Mar 2012 Location: Fife, Scotland, UK. Posts: 22,858	Re: TF 2100 A.F.Oscillator With two units you're good for two-tone tests like intermodulation distortion, or running up an SSB transmitter for PEP measurement. It might be more versatile to leave the units original and just build a little power amp. In the extreme a box with a power amp and speaker as well as a combiner. THen you have an amplified test speaker, a low impedance driver and a combiner all of which can be exploited separately as needed. David __________________ Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done

23rd Jul 2021, 2:14 pm	#4
Jon_G4MDC Nonode Join Date: Nov 2015 Location: Welwyn Garden City, Hertfordshire, UK. Posts: 2,015	Re: TF 2100 A.F.Oscillator I'm not sure that co-ax output was intended to be used apart from when the oscillator was built into a frame such as in the Transmission Measuring Set. There it fed the adjacent MF Output Attenuator. In normal standalone use wouldn't output be taken via the 4mm sockets at the front? I think that coax connector is probably a mini Belling Lee - like a scaled down version of the 75 Ohm TV plug. Not all that hard to find and better than modifying an otherwise "mint" unit.

23rd Jul 2021, 2:16 pm	#5
Julesomega Nonode Join Date: Dec 2012 Location: Stockport, Greater Manchester, UK. Posts: 2,102	Re: TF 2100 A.F.Oscillator I won't need to mod it if I can find a Farnell LFP1 The designers were certainly designing for servicability, maybe in response to pressure from the repairs department! Thankfully someone on the Marconi Instruments forum sent me a full scan of the manual __________________ - Julian It's good here

24th Jul 2021, 7:50 am	#6
Diabolical Artificer Dekatron Join Date: Jul 2009 Location: Sleaford, Lincs. UK. Posts: 7,657	Re: TF 2100 A.F.Oscillator I wonder when the fold out design first came out, the TF2331 distortion analyser used the same fold out construction, whoever thought of it is a genius. Makes assembly easier too. Andy. __________________ Curiosity hasn't killed this cat...so far.

24th Jul 2021, 8:08 am	#7
Dickie Octode Join Date: Sep 2004 Location: St. Albans, Hertfordshire, UK. Posts: 1,478	Re: TF 2100 A.F.Oscillator It appeared in 1963, and was designed by W.D. Cain, the chief mechanical engineer at the time. He was quite a free-thinker. https://worldradiohistory.com/Archiv...ni-1963-06.pdf It's also worth browsing the other issues at this time, as they contain details of of other instruments that crop up on the forum. __________________ Regards, Richard, BVWS member

24th Jul 2021, 8:44 am	#8
Julesomega Nonode Join Date: Dec 2012 Location: Stockport, Greater Manchester, UK. Posts: 2,102	Re: TF 2100 A.F.Oscillator Thanks for that link Richard, and hooray for Mr.Cain! The issue was devoted almost entirely to the TF2000/2100 and the completely unknown "MF" variants which extend the range to 550kHz (they were not in the later catalogues I have). It's hard to imagine what market these 600Ω instruments were intended for, and carrier telephony systems must have been into the MHz region by then __________________ - Julian It's good here

24th Jul 2021, 10:03 am	#9
Jon_G4MDC Nonode Join Date: Nov 2015 Location: Welwyn Garden City, Hertfordshire, UK. Posts: 2,015	Re: TF 2100 A.F.Oscillator The TMS I referred to earlier went to 550kHz and the oscillator was marked MF not AF. Low capacity multiplexed links were probably still operating below 550kHz. I am thinking of units such 1+1s etc.