G4FPH's Amateur Radio Pages - Racal TTA 1885A TX Terminal Project
This project started with my spotting an interesting looking item on eBay. Never before attracted to Racal equipment, the size of the roller coaster coils in the eBay photo encouraged me to find out more, and eventually bid and win the item. If you find yourself with one of these units, I may be able to help you get it going or know someone who can.
The project is still very much work in progress. If you are interested, drop me an e-mail for more details.
The Racal TTA 1885A is a 1 kW, HF, multimode transmitter terminal. The terminal has continuous coverage from 1.6 (hence the big roller-coasters!) to 30 MHz and comprises two units: The MA1723 drive unit is all solid state, with synthesised tuning in 100 Hz steps and delivers a nominal 100 mW into the two-stage companion, TA1823, power amplifier. The power amplifier features a broadband solid state stage operating in Class A producing a maximum 25 W to drive a single 4CX1500B valve operating in passive grid, grounded cathode, class AB1. The valve section of the amplifier unit employs a Pi-L output network and is fully auto-tuned using relays, solenoids and motorised roller-coasters. Oh, yes! Operating modes are SSB, ISB, AM, CW. In summary, the TTA 1885A is a fine example of 1980s British RF engineering that probably cost the UK taxpayer very dear the first time around!
Not many pictures yet, but click on the thumbnail below for a bigger picture.
With the last internal anode, glass envelope valve having been retired from active service, the G4FPH station now relies on external anode valve finals. Recent experiences with triodes in grounded grid having been entirely positive, it was time to get to grips with external anode tetrodes. I once thought of making a VHF PA with 4CX250B, but got fed up designing power supplies and interlocks to protect the valve long before the project made it to air. Tetrode - what a pain! With Racal having done the hard work on the PSU and interlock front and with the incentive of getting reasonable power output at the end of the project, I welcomed the Racal TTA1885A with its Eimac 4CX1500B tetrode into the station.
The circuitry of the MA1723 drive unit is CPU controlled (RCA 1802 processor), with the operating programme stored in 2532 EPROMS. The rest of unit's circuitry is mostly CMOS, analogue and discrete. Big boards, easy to mend. Possible problems for the future are ageing tantalum capacitors and availability of the processor IC plus a custom Racal IC in the synthesiser, plus odd ECL ICs. The two large, LCD display panels are already quite badly deteriorated.
The TA1823 amplifier is designed to 'plug and play' with the drive unit via a straightforward, multi-wire interface. The amplifier can be fully remote controlled from the drive unit's front panel. The amplifier is controlled by a rack of cards - no processor this time, just lots of CMOS, analogue and discrete. Relays in the amplifier are responsible for switching the output valve grid network (choice of three depending on operating frequency, while chunky, 115 V AC solenoids push contacts to select combinations of capacitors for 'C1' and 'C2' in the Pi-L output network. 'L1' (tune - on right of picture) and 'L2' (load - on left of picture) in the Pi-L output network are large roller coaster coils driven by DC-powered servo motors. Remember analogue servos!? The amplifier divides the HF spectrum into 8 bands. The amplifier auto tuning sequence is in three parts:
Part 1 measures (coarsely) the operating frequency of the driver to decide on which of the 8 bands it should pre-select capacitors for and which of the 3 grid networks it should use. Before moving on to part 2, it presets the tune and load coils to initial positions, again acording to frequency.
Part 2 applies a limited amount of drive and motors the tune roller coaster to a point close to anode resonance.
Part 3 applies full drive and motors the load roller coaster to present a resistive load to the valve and has a final tweak of tune and load together.
The MA1723 drive units have built in test routines (Racal BITE) that are actually useful! So far, drive unit problems have been faulty synthesisers caused by duff tantalum capacitors and burned out resistors.
The TA1823 units have no self test and should be treated with the usual caution due to the dangerous voltages inside. Racal have very kindly provided a 'chicken stick' in the bottom of the amplifiers front compartment. On my amplifier, an indication of unexpectedly high power output on some bands pointed to a faulty VSWR bridge (very frequency concious) rather than a re-writing of the laws of physics! Some arcing off the bottom of the 'load' roller coaster went away after testing - must look into this one. (Tune) servo instability around 10 MHz needs to be investigated too. I can confirm that the unit is unable to tune into open circuit - 50 Ohm is better!
Does what it says on the rack. Will cause ICAS rated dummy loads to mess themselves if you forget to turn the TX off!
1. If you are interested in receiving as well as transmitting, the first modification is to make a changeover system for use with the transmitter terminal. As it comes, the TTA 1885A is essentially either switched off or in TX! There is a standby mode when no RF comes out, but all this means is that there is no drive being fed to the amplifier - the PA valve still sits there with 3 kV on it and 300 mA of standing current - good for warming the shack, but not very 'green' and could even generate a bit of noise in a nearby receiver. My prototype c/o puts full unregulated negative bias rail (about -100 V) on the grid on RX and cuts the valve off a treat. Conventional coax relays for the aerial changeover are slow, but a quick fix to get going. New to many will be the need to mute a companion receiver when using with the TX. As most transceivers have no facility for doing this, my plan is to interrupt the external loudspeaker feed, or something equally crude. I hope the AGC will recover in time for me to hear what's being said!
2. Next challenge is to get microphone interfaced to the drive unit. The stock mic inputs (it has 2 in case you want to run ISB) are for carbon microphones - not compatible with the content of my junk box. Also, the mic inputs have low value resistors pulling up to 15 Volts to energise the carbon micropheone and no mic gain control. Have I put you off yet? Fortunately the rear panel yields a couple of 600 Ohm balanced line level inputs. My plan is to make a mic preamp that will give a balanced output at 0 dBm to feed these (there are gain pots for the line level inputs on the back panel too!).
3. Fitting a red light to show when the unit is on TX must surely be a good idea. Do this before you cook the dummy load you borrowed from a friend!
Almost there - I'm looking forward to my first QSO. More words and picture when I get time!
Thanks to Malc, G4FQI, for transport, storage and skateboard!
Copyright ©2004 G4FPH - even the spelling mistakes.
Last modified: 14 November 2004