Eddystone EA12 & spurii

[Skywave]

I am in the process of doing a routine service on two of these. Initially, both had poor sensitivity, plus various other minor (mainly mechanical) faults. Following the usual replacement of changed-value resistors (to get valve electrode voltages about right) and a general re-alignment, both now meet (or exceed) the published signal-noise sensitivity and other important specs.

However, a new problem has come to light. Each receiver generates strong spurious signals on 3.70, 3.75 MHz and at various other frequencies (I haven't made a note of the exact freqs. yet) on other bands. Disabling the RF stage does not effect a "cure". A little arithmetic indicates that these spurii could be caused by unwanted mixing of harmonics from the 1st. and 2nd. local oscillators. If so, a "fix" for this is not going to be easy - and probably not feasible.

I have posted a similar request for help on the EUG Forum, but I thought I'd try here as well.

So - has anyone else here had any in-depth experience of maintenance of these receivers with regard to this problem?

Thank you.

Al / Skywave / Sun. 12-08-07 //

[jim_beacon]

AL,

check all the screening and earth points carefully.
check teh decoupling caps on the oscillator feeds.

This is a common fault with RA17's - you get a lot of spurious signals from the 1MHz oscilator, and they are usually caused by faulty screening or decoupling.

Jim.

[Skywave]

Thanks Jim.

Indeed, the earthing and decoupling caps. have come under suspicion, but nothing definitely wrong was found. As for the screening, the only "changeable" things here are the bottom cover of the coil box, the screening can of the 1st. VFO / 2nd. mixer and the valve cans. All look intact, solid connections to chassis, etc.

To me, it just seems highly suspicious that there are two of these sets that exhibit the same faults (= same freqs.). I find it difficult to believe that Eddystone would have released this model with these faults so manifest. . . . which, therefore, implies a fault condition

However, one issue does disturb me. Upon exam. with a 'scope, the output from the buffer of the 1st. VFO (to the 1st. mixer) is a saw-tooth waveform. Therefore, rich in harmonics. At this point, such a waveform is definitely not what is wanted - which leads to my earlier comment about identifying the source of the spurious signals (calculating unwanted mixing products), etc.

So - please excuse me if I put this so direct - do you have any hands-on experience of the EA-12? We can always go into "PM mode" if things get really detailed (to avoid boring everyone else).

Aside: As for the RA-17, I have owned several of these over the years and have worked on many, (unlike the EA-12). Although I have experienced the fault you describe in the Racal, I haven't come across it very often.

Al / Skywave.

[jim_beacon]

Al,

No, I don't have direct EA12 experience, in fact after some early bad experiences (640, EC958, and the VHF and UHF monitor thing), I've tried to steer clear of the products of Stratton and Company - they seemm to be designed to be difficult to work on!

I was hoping the general theory of poor screening / decoupling would help, though it soounds like the sawtooth from the oscillator could be part of the problem. Having looked at teh circuit, Iwould try changing V2, as this is both doubler and mixer (seemes like a bad idea....). As a quick thought, does it give spurious rseponses on some bands and not others (depending on whether or no the crystal frequency is doubled).

As regards the RA17s, maybe I've just been unlucky, but I've seen it in three of the four that have been across my bench.

Jim.

[Skywave]

Jim -
Thanks for the prompt response.

With regard to your feelings about products from Stratton & Co., I must say that I agree with you.
Having owned / worked with a whole raft of comms. receivers over the years (ex-WW2 stuff, Racal gear, Plessey, Marconi and a lot of USA equipment - amongst others), Strattons Offerings do seem to require a particular form of patience and perseverance - not to mention time - to get a faulty one working properly. For example, removing the front panel and stripping down the drive mechanism on an Eddystone - and then re-building it - is decidedly not for the faint-hearted.

Anyway, I digress. Back to the main issue.

Yes - your comment on general screening / decoupling is, of course, valid - and appreciated. I will investigate this aspect deeper.
Re: V2 - buffer / doubler and mixer in same envelope. And sharing the same cathode. Looks like an invitation for trouble to me * . Ditto V4.

Re: assessment of all spurii.
No, I haven't done a full all-band assessment - yet. When I found the problem on the 80m. band, (3.70, 3.75 and 4.00 MHz) I stopped there, and decided to investigate this. I intend to do an "all-band inspection" later in the week.

Just for the record, my analysis (& thinking) goes like this -

There is a sprog that appears when the tuning dial is at 3.70 MHz.
Removing the RF amp. valve does not clear it - so it's not "front-end pick-up". But removing V2 (ECH81) does. Fitting a replacement V2 (a "known good") does not solve the problem.

Suspecting unwanted harmonic mixing products, the following reasoning applies -

When the tuning dial is at 3.70 MHz, the 2nd. osc / VFO is at 1.30 MHz. The 4th. harmonic of this is 5.2 MHz.
The 1st. osc. (crystal) is at 5.1 MHz.
So, any stray 5.1 MHz arriving at the grid of the 2nd. mixer (V4a) (via the 1st. mixer *) will mix with the VFO 4th. harmonic (5.2 MHz) and produce 100 kHz - i.e. the IF.
The resulting sprog will be tunable by virtue of "adjusting" the 2nd. osc / VFO - giving the "tunable" spurious symptom.

A similar reasoning goes for the sprog at 3.75 MHz. (VFO at 1.25 MHz)
At 4.00 MHz, for 4th. harmonic above, read 5th. (VFO at 1.00 MHz)

I'll leave you (or anyone else reading this and who has got this far) to do the arithmetic

* When V2 is removed, the sprog disappears. Therefore, the 1st. osc. buffer and 1st. mixer (both of which are contained within V2) are contributory to the problem, and not the 1st. osc. (crystal) itself - which is in a seperate valve (V3).
Also note that there is no attempt at any filtering of harmonics from the 2nd. VFO buffer, prior to its output being fed to the 2nd. mixer (V4A). This VFO tunes from 1.6 to 1.0 MHz; therefore (and I hate to say this ) a modification by way of fitting a low-pass filter on the output of the buffer springs to mind - but only as a last resort.

That's the full story to date. Naturally, I'll welcome any further comments / suggestions in my attempts to solve this problem.

Al / Skywave.

[Skywave]

Update -

I have spent countless hours trying to clear this fault - all to no avail.

Unfortunately, various other matters in my life - which have been put "on hold" during this extended saga - have now become critical. Therefore, I have no choice but to set these two receivers aside for another day, another time - if ever.

I'd just like to close now & say thank you to everyone who provided assistance on this matter - on this Forum & elsewhere.

However, I am quite prepared to talk to any other Members on this issue via PM's.

Al / Skywave