The Racal RA1218, and more ... or, what happened to the RA1219?
13 minute read
This post is part of the series 'Racal's early solid-state receivers':
- The Racal RA1217
- A new 1MHz Osc. for the RA1217
- The Racal RA1218, and more ... or, what happened to the RA1219?
July 2025
This is the second RA1218 that has come through my workshop. This one is destined to be extremely well-travelled as it came from Italy and will ultimately be going to a client in New Zealand. Although the rear panel was decidedly 'ugly' on account of corrosion and grime, internally it was in very good condition. I suspect the general cleanliness can be attributed to the fact that solid-state equipment does not attract dust in the way valve-based kit does.
The purpose of this article is not so much an in-depth discussion, but think of it more of a 'thinking aloud' approach on account of the distinct lack of information available on the RA1219 and RA1220.
As with the RA1217 and all other Racal kit preceding the RA1770 series, the mains input is via the ubiquitous Plessey connector. In this case the previous owner had simply soldered wires onto the pins of the chassis plug. I know that Plessey connectors don't meet EC regulations, but I think we'll call this approach 'temporary'.
I'd had exploding capacitors in the PSU in my RA1217, so I decided to pre-empt such an occurrence with the RA1218 by examining the PSU before switching it on. The PSU is slightly different to that in the RA1217 in that it generates +ve as well as -ve rails, and a +200V rail for the Nixies. Like in my RA1217, one of the 20nF capacitors on the mains side of the transformer had exploded. As well as that, the 'blow-holes on most of the electrolytics were bulging. Thus, I did a 'blanket' replacement of all the electrolytics in the PSU ... see below.
A correct quick-release Plessey connector was fitted to the mains cord, and on powering the RA1218 up, I found that it was apparently in good working order, although as with my RA1217, there was an issue with the AGC. In this case, the long time-constant setting was not working. Actually, when that setting was selected, there was a distinct lack of audio output. This was traced to C14 (64uF) on the AGC board in the IF module being a low value resistor! I replaced this with a 68uF tantalum bead as that was the only capacitor I had in stock that would fit. After that, all was sweetness and light. As with my RA1217, the underside of the AGC board had seen much re-work in the past.
It is no secret that the RA1217 was not well received. This was likely for two reasons. The first being that from the perspective of one who was accustomed to the RA17, the MHz and KHz controls are transposed. Thus, they likely found themselves spinning the MHz control by mistake when tuning across a band. The second reason was very likely down to the RA1217's relatively poor performance, due to higher noise figure, which was down to the nature of the transistors used in the design.
As can be seen from the above photographs, there is little physical difference between the RA1217 and the RA1218, except for the inclusion of the temperature controlled oscillator and the number of IF filters installed, and this latter difference may be simply down to end-user choice. Also, the RA1218 front panel is less crowded and there are no concentrically-mounted controls thanks to the extra 1.75 inches in panel height.
However, the boys at Bracknell had been busy elsewhere. The little boxes may look similar, but there are subtle differences. In many cases, different transistors have been used. Much appears to have been done in an effort to improve the performance and stability of the receiver, and it doesn't take much imagination to understand why. I can't find any dates on the documents I have relating to the RA1218. However, the manuals I have for the RA1217 are dated 1967, so the RA1218 would have appeared some time after 1967. This was at a time when the Cold War was anything but cold ... things were heating up. At that time, the bulk of the traffic handled by the RA17s and MA79s etc. would have been RTTY. An inevitability was that it was becoming necessary to cram more data into narrower bandwidths and this required much improved stability.
The RA17 was actually more stable than the RA117. Or more correctly, the RA117 was not as stable as the RA17, by nature of its design. In order to have the sidebands at the final IF the 'right way round' (apparently a GCHQ requirement), it was necessary to slot an extra IF into the mix. This used up space, and in doing so a concession had to be made around the 2nd VFO/2nd IF. The drift on the RA117 2nd VFO is quite apparent ... took me by surprise. This 'feature' is however eliminated when the RA117 is used in conjunction with an MA350B, which as well as providing a synthesised 2nd external LO, also provides rock-steady signals at 1MHz and 1.7MHz.
The RA1218 embodies design features aimed at providing performance improvements over the RA1217. Some of the design is borrowed from the MA350. The overall receiver design is the same ... essentially still a solid-state RA117. However the 1MHz signal at the root of the Wadley drift-cancelling system applied to the 1st VFO, is now derived from a temperature-controlled 5MHz crystal oscillator, which also provides a 'standard' for the counter used for the frequency display. The derived 1MHz is further divided down to 100KHz for the calibrator and this 100KHz signal is also used to provide a drive source for the 1.7MHz source. But clearly this wasn't enough.
Bring on the RACALATOR! Who on earth came up with that name! Did someone ask a child to give it a name? It doesn't even sound like what it does! Anyway, this is the MA210 ... a frequency stabiliser intended for the RA17L and RA17C-12 ... at first.
As can be seen from my cunning artist's impression above, the MA210 originally came in rack-mount form, and was designed to keep the RA17's 2nd VFO to within 1 to 2Hz over long periods. A means of fine tuning the RA17 over a range of 1KHz is made available. According to the MA210 brochure, a tuning voltage is applied to the receiver's modified 2nd VFO. This voltage is represented on the MA210 panel meter. According to the brochure, tuning is achieved thus ...
From the brochure:
To tune the system to a required signal frequency the RACALATOR function switch is set to 'Coarse-Tune', and tuning controls adjusted until the receiver is set within +/-100Hz of the incoming signal as indicated on the RACALATOR display unit. The function switch is then set to 'Fine-Tune' and the multiturn control on the MA210 front panel, which has an overall range of 500Hz is adjusted until the display indicates the desired frequency within +/-1Hz. On switching to 'Hold', the RACALATOR starts the process of controlling the receiver second variable frequency oscillator to eliminate the effects of frequency drift. In this condition, the mean variation of receiver tuned frequency will not be more than 1-2Hz over very long periods of time. [sic]
I can see how this works. I can even see how fluctuations in a counter's least significant digit might be converted into a tuning voltage to apply to a varicap diode. But I can't help thinking that the last sentence is a tad vague when it states that the variation in frequency will be no more than '1-2Hz over very long periods of time'. I also can't help thinking that there is at least one mistake in the text in the brochure. Initially the brochure states that the control range is +/- 1KHz, yet the quoted paragraph states that the range is only 500Hz. I suspect the first instance should maybe read 'a control range of +/-500Hz' and the second instance should read 'overall range of 1KHz'.
What Racal did was to take the RA1218 and replace the frequency counter module with a modified MA210 ... one that was compatible with a 2nd VFO running between 3.6MHz and 4.6MHz. As with the RA1217, the 2nd VFO in the RA1218 is already fine-tuned using varicap diodes. Thus, the 'modified' RA1218 was re-designated as an RA1219.
Unfortunately for us geeks who are intrigued by these things, there is literally zero information available on the RA1219. I did however manage to find one photograph of what was identified as an RA1219 (above). If you look carefully at the controls towards the bottom right you will notice that once again there are two concentric controls (AF and RF Gain, bottom right). What was 'RF Gain' on the RA1218 is now the Mode Selector switch for the built-in MA210. Fine tuning is by way of the original 'Fine-Tune' knob (as in the RA1217 and RA1218) between the KHz and MHz controls. In the case of the RA1219, this is now labelled 'Fine Tune Interpolate'. There does not appear to be a Tuning Voltage Meter, although there does appear to be a green lamp to the right of the existing meter, which I suspect is for indicating 'Lock'.
Further trawling of the internet unearthed an article in the June 1970 edition of Wireless World by Pat Hawker, G3VA, entitled 'Communications Receivers - An examination of the extent to which circuit design and cost are being influenced by increasingly stringent performance requirements'.
Also in the same publication there is a list of communications receivers available on the British market, and their manufacturers. On the fifth and final page of this list, Racal-BCC Ltd is listed as supplying the RA17, RA117, RA1217, RA1218, RA1220, RA329B and RA6217. Note how the RA1219 is missing from the list. I even found a 're-print' of Racal-BCC's product catalogue which confirmed that the list published in Wireless World was merely a sub-set of their product range. But again, the RA1219 was missing.
I know these two images are woefully small, however, such is the dearth of information available, these will have to do. Both are taken from Racal-BCC's product catalogue, and I have included them here simply to highlight the uncanny similarity between the RA1220 and the RA1219, and in turn to the RA1218.
As luck would have it, the June 1970 issue of Wireless World also carried a photographic montage on the front cover, showcasing the RA1220. This really grabbed my attention. Not just because Nixie tubes are cool (pun intended, since they are 'cold-cathode' devices), but because of the striking similarity between the RA1218 and the RA1220.
My focus was instantly drawn to the display sub-assembly ... and not just to the 1.7MHz board with the two crystals. It was the chassis that really struck me. In the RA1218, the frequency counter module is a double-decker structure, with the counter electronics on the bottom level and the 1.7MHz board on top. Note how the chassis for what Racal now called the 'Racalok' is identical to the frequency counter chassis in the RA1218. The same 1.7MHz board, mounted in the same position and the cut-outs for the edge connectors are the same. There is even a similarity to some extent in the hardware attached to the rear wall of the assembly.
It is obvious that the RA1220 incorporates an additional Nixie tube ... and there are clearly six 14-pin ICs visible on the large PCB, BUT ... the rest of the circuity on that board appears to be analogue, so I'm guessing that as in the RA1218, the actual counter is on the lower level. I also note that the 'Lock' indicator on the RA1220 appears to be a green LED as opposed to the green lens over an incandescent light bulb in the RA1219. Did we have green LEDs back in 1970?
So here's where I stick my neck out ... The RA1217 was a brave attempt to make a solid-state version of the RA117 ... and it is clear that lessons were very quickly learned. At the time, advances in solid-state technology were appearing at such a rate that it was difficult at times to keep apace with them. To that extent, the RA1218 was an inevitability. I will venture that the pretentiously-named Racalator or MA210 and the Nixie display in the RA1218 are the product of the same requirement and/or thought process ... and that the RA1220 is little more than a re-badged RA1219, with the Racalator re-named as the Racalok, hence why the 1219 is missing from Racal-BCC's catalogue.
The purpose of this article is not so much an in-depth discussion, but think of it more of a 'thinking aloud' approach on account of the distinct lack of information available on the RA1219 and RA1220.
As with the RA1217 and all other Racal kit preceding the RA1770 series, the mains input is via the ubiquitous Plessey connector. In this case the previous owner had simply soldered wires onto the pins of the chassis plug. I know that Plessey connectors don't meet EC regulations, but I think we'll call this approach 'temporary'.
I'd had exploding capacitors in the PSU in my RA1217, so I decided to pre-empt such an occurrence with the RA1218 by examining the PSU before switching it on. The PSU is slightly different to that in the RA1217 in that it generates +ve as well as -ve rails, and a +200V rail for the Nixies. Like in my RA1217, one of the 20nF capacitors on the mains side of the transformer had exploded. As well as that, the 'blow-holes on most of the electrolytics were bulging. Thus, I did a 'blanket' replacement of all the electrolytics in the PSU ... see below.
RA1218 PSU before.
RA1218 PSU after.
PSU main component board.
PSU 5V board.
A correct quick-release Plessey connector was fitted to the mains cord, and on powering the RA1218 up, I found that it was apparently in good working order, although as with my RA1217, there was an issue with the AGC. In this case, the long time-constant setting was not working. Actually, when that setting was selected, there was a distinct lack of audio output. This was traced to C14 (64uF) on the AGC board in the IF module being a low value resistor! I replaced this with a 68uF tantalum bead as that was the only capacitor I had in stock that would fit. After that, all was sweetness and light. As with my RA1217, the underside of the AGC board had seen much re-work in the past.
AGC board with new C14.
It is no secret that the RA1217 was not well received. This was likely for two reasons. The first being that from the perspective of one who was accustomed to the RA17, the MHz and KHz controls are transposed. Thus, they likely found themselves spinning the MHz control by mistake when tuning across a band. The second reason was very likely down to the RA1217's relatively poor performance, due to higher noise figure, which was down to the nature of the transistors used in the design.
RA1217.
RA1218, RF module cover temporarily removed.
As can be seen from the above photographs, there is little physical difference between the RA1217 and the RA1218, except for the inclusion of the temperature controlled oscillator and the number of IF filters installed, and this latter difference may be simply down to end-user choice. Also, the RA1218 front panel is less crowded and there are no concentrically-mounted controls thanks to the extra 1.75 inches in panel height.
However, the boys at Bracknell had been busy elsewhere. The little boxes may look similar, but there are subtle differences. In many cases, different transistors have been used. Much appears to have been done in an effort to improve the performance and stability of the receiver, and it doesn't take much imagination to understand why. I can't find any dates on the documents I have relating to the RA1218. However, the manuals I have for the RA1217 are dated 1967, so the RA1218 would have appeared some time after 1967. This was at a time when the Cold War was anything but cold ... things were heating up. At that time, the bulk of the traffic handled by the RA17s and MA79s etc. would have been RTTY. An inevitability was that it was becoming necessary to cram more data into narrower bandwidths and this required much improved stability.
The RA17 was actually more stable than the RA117. Or more correctly, the RA117 was not as stable as the RA17, by nature of its design. In order to have the sidebands at the final IF the 'right way round' (apparently a GCHQ requirement), it was necessary to slot an extra IF into the mix. This used up space, and in doing so a concession had to be made around the 2nd VFO/2nd IF. The drift on the RA117 2nd VFO is quite apparent ... took me by surprise. This 'feature' is however eliminated when the RA117 is used in conjunction with an MA350B, which as well as providing a synthesised 2nd external LO, also provides rock-steady signals at 1MHz and 1.7MHz.
The RA1218 embodies design features aimed at providing performance improvements over the RA1217. Some of the design is borrowed from the MA350. The overall receiver design is the same ... essentially still a solid-state RA117. However the 1MHz signal at the root of the Wadley drift-cancelling system applied to the 1st VFO, is now derived from a temperature-controlled 5MHz crystal oscillator, which also provides a 'standard' for the counter used for the frequency display. The derived 1MHz is further divided down to 100KHz for the calibrator and this 100KHz signal is also used to provide a drive source for the 1.7MHz source. But clearly this wasn't enough.
Bring on the RACALATOR! Who on earth came up with that name! Did someone ask a child to give it a name? It doesn't even sound like what it does! Anyway, this is the MA210 ... a frequency stabiliser intended for the RA17L and RA17C-12 ... at first.
The MA210 Racalator.
As can be seen from my cunning artist's impression above, the MA210 originally came in rack-mount form, and was designed to keep the RA17's 2nd VFO to within 1 to 2Hz over long periods. A means of fine tuning the RA17 over a range of 1KHz is made available. According to the MA210 brochure, a tuning voltage is applied to the receiver's modified 2nd VFO. This voltage is represented on the MA210 panel meter. According to the brochure, tuning is achieved thus ...
From the brochure:
To tune the system to a required signal frequency the RACALATOR function switch is set to 'Coarse-Tune', and tuning controls adjusted until the receiver is set within +/-100Hz of the incoming signal as indicated on the RACALATOR display unit. The function switch is then set to 'Fine-Tune' and the multiturn control on the MA210 front panel, which has an overall range of 500Hz is adjusted until the display indicates the desired frequency within +/-1Hz. On switching to 'Hold', the RACALATOR starts the process of controlling the receiver second variable frequency oscillator to eliminate the effects of frequency drift. In this condition, the mean variation of receiver tuned frequency will not be more than 1-2Hz over very long periods of time. [sic]
I can see how this works. I can even see how fluctuations in a counter's least significant digit might be converted into a tuning voltage to apply to a varicap diode. But I can't help thinking that the last sentence is a tad vague when it states that the variation in frequency will be no more than '1-2Hz over very long periods of time'. I also can't help thinking that there is at least one mistake in the text in the brochure. Initially the brochure states that the control range is +/- 1KHz, yet the quoted paragraph states that the range is only 500Hz. I suspect the first instance should maybe read 'a control range of +/-500Hz' and the second instance should read 'overall range of 1KHz'.
What Racal did was to take the RA1218 and replace the frequency counter module with a modified MA210 ... one that was compatible with a 2nd VFO running between 3.6MHz and 4.6MHz. As with the RA1217, the 2nd VFO in the RA1218 is already fine-tuned using varicap diodes. Thus, the 'modified' RA1218 was re-designated as an RA1219.
RA1219.
Unfortunately for us geeks who are intrigued by these things, there is literally zero information available on the RA1219. I did however manage to find one photograph of what was identified as an RA1219 (above). If you look carefully at the controls towards the bottom right you will notice that once again there are two concentric controls (AF and RF Gain, bottom right). What was 'RF Gain' on the RA1218 is now the Mode Selector switch for the built-in MA210. Fine tuning is by way of the original 'Fine-Tune' knob (as in the RA1217 and RA1218) between the KHz and MHz controls. In the case of the RA1219, this is now labelled 'Fine Tune Interpolate'. There does not appear to be a Tuning Voltage Meter, although there does appear to be a green lamp to the right of the existing meter, which I suspect is for indicating 'Lock'.
Further trawling of the internet unearthed an article in the June 1970 edition of Wireless World by Pat Hawker, G3VA, entitled 'Communications Receivers - An examination of the extent to which circuit design and cost are being influenced by increasingly stringent performance requirements'.
Also in the same publication there is a list of communications receivers available on the British market, and their manufacturers. On the fifth and final page of this list, Racal-BCC Ltd is listed as supplying the RA17, RA117, RA1217, RA1218, RA1220, RA329B and RA6217. Note how the RA1219 is missing from the list. I even found a 're-print' of Racal-BCC's product catalogue which confirmed that the list published in Wireless World was merely a sub-set of their product range. But again, the RA1219 was missing.
RA1220 listing.
I know these two images are woefully small, however, such is the dearth of information available, these will have to do. Both are taken from Racal-BCC's product catalogue, and I have included them here simply to highlight the uncanny similarity between the RA1220 and the RA1219, and in turn to the RA1218.
RA1218 listing.
RA1220 on front cover of June 1970 Issue of Wireless World.
Inside of counter in RA1218.
As luck would have it, the June 1970 issue of Wireless World also carried a photographic montage on the front cover, showcasing the RA1220. This really grabbed my attention. Not just because Nixie tubes are cool (pun intended, since they are 'cold-cathode' devices), but because of the striking similarity between the RA1218 and the RA1220.
My focus was instantly drawn to the display sub-assembly ... and not just to the 1.7MHz board with the two crystals. It was the chassis that really struck me. In the RA1218, the frequency counter module is a double-decker structure, with the counter electronics on the bottom level and the 1.7MHz board on top. Note how the chassis for what Racal now called the 'Racalok' is identical to the frequency counter chassis in the RA1218. The same 1.7MHz board, mounted in the same position and the cut-outs for the edge connectors are the same. There is even a similarity to some extent in the hardware attached to the rear wall of the assembly.
It is obvious that the RA1220 incorporates an additional Nixie tube ... and there are clearly six 14-pin ICs visible on the large PCB, BUT ... the rest of the circuity on that board appears to be analogue, so I'm guessing that as in the RA1218, the actual counter is on the lower level. I also note that the 'Lock' indicator on the RA1220 appears to be a green LED as opposed to the green lens over an incandescent light bulb in the RA1219. Did we have green LEDs back in 1970?
So here's where I stick my neck out ... The RA1217 was a brave attempt to make a solid-state version of the RA117 ... and it is clear that lessons were very quickly learned. At the time, advances in solid-state technology were appearing at such a rate that it was difficult at times to keep apace with them. To that extent, the RA1218 was an inevitability. I will venture that the pretentiously-named Racalator or MA210 and the Nixie display in the RA1218 are the product of the same requirement and/or thought process ... and that the RA1220 is little more than a re-badged RA1219, with the Racalator re-named as the Racalok, hence why the 1219 is missing from Racal-BCC's catalogue.