UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Amateur and Military Radio

Notices

Vintage Amateur and Military Radio Amateur/military receivers and transmitters, morse, and any other related vintage comms equipment.

Closed Thread
 
Thread Tools
Old 23rd Jun 2019, 9:23 pm   #1
cnpope20
Triode
 
Join Date: Jan 2018
Location: College Station, Texas, USA.
Posts: 11
Default Racal RA17 alignment

I'm in the process of acquiring a Racal RA17, and at some stage I'll be wanting to do an alignment. I think the trickiest thing is aligning the 40 MHz bandpass filter. I've been looking around for a sweep generator that I can use for doing this, and I am wondering whether one of the cheap(ish) Chinese generators like the Feeltech FY6800 60MHz might provide a reasonable solution.

Does anyone have experience of using one of these? As far as I can make out from the specs, it should be possible to set it up to do something like, for example, a 39 MHz - 41 MHz sweep. But I'd hate to buy one, and discover only later that the sweep range was restricted to 25 MHz maximum, or something like that! Has anyone used one of these?

I know the FY6800 is probably far from ideal for many RF purposes, since it is poorly shielded and cannot be relied on for supplying calibrated microvolt-level outputs. But for sweep generator purposes for the RA17 bandpass filter, or a more conventional IF amplifier, I'm thinking it might be reasonable.

Alternatively, does anyone have any other suggestions for a modestly priced sweep generator that could do the job?

Thanks.
cnpope20 is offline  
Old 23rd Jun 2019, 11:25 pm   #2
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Racal RA17 alignment

It isn't just a signal source you need, but also a means of detection suited to the levels in play.

Both source and detector need to be very lightly coupled to the filter to avoid pulling it. Light coupling implies high losses in the couplings, so the detector needs to be sensitive.

I'm afraid it's not low cost but my usual tool is a spectrum analyser with a tracking generator.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 23rd Jun 2019, 11:46 pm   #3
Craig Sawyers
Dekatron
 
Craig Sawyers's Avatar
 
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,941
Default Re: Racal RA17 alignment

The two official tools that Racal used were a Rohde & Schwartz SWOB 2 Polyskop, or a Samwell and Hutton CT501 wobbulator or the non military 78m.

The fact I have both is absolutely no use to you because I am on the other side of the pond and 5000 miles from where you live. And they weigh hundredweights. They weigh more than an RA17!

But as Dave says, the more more modern alternative to those 1950's boatanchors is a spectrum analyzer with tracking generator.

The single most difficult thing to align on an RA17 is the 40MHz filter. And unless someone has been fiddling with it, it is best to leave well alone - unless your tests suggest that there is a problem there.

Their is much to be done before you get that far. For example the Hunts metallised paper capacitors will be leaky of have no measurable capacitance. And there are lots of them. Replace them all with axial polypropylene. Replace any resistor that is out of tolerance - they are carbon composition and drift high. Typically for example you might find a 47k resistor that measures closer to 68k. There are likely to be many of those. The red coloured silver mica capacitors are another known problem area.

Then test the valves/tubes, and replace any that are shot.

Then see how well it works with an antenna input. Only then consider the lengthy alignment process.

Craig
Craig Sawyers is online now  
Old 24th Jun 2019, 5:34 am   #4
cnpope20
Triode
 
Join Date: Jan 2018
Location: College Station, Texas, USA.
Posts: 11
Default Re: Racal RA17 alignment

Quote:
Originally Posted by Craig Sawyers View Post
The single most difficult thing to align on an RA17 is the 40MHz filter. And unless someone has been fiddling with it, it is best to leave well alone - unless your tests suggest that there is a problem there.
Craig
Yes, indeed. But I would still like to be able to make observations and measurements, even without having an intention of re-aligning it. I just like to measure things anyway, for the fun of it!

It is hard for me to justify the cost of a spectrum analyzer and tracking generator, really just for the purpose of satisfying my curiosity about what the response curve looks like, so I'm looking around in case there might be some cheaper option. For all its obvious shortcomings, the FY6800, on paper, appears to offer a cheap way of obtaining a sweep generator source that could be capable of covering the required frequency range, and so it has to rank somewhere, however low, on the list of possible ways of achieving what I'd like to do. But it would "come off the list" if it were to turn out that for some unforeseen reason it did not in fact have the capability of sweeping the required frequency range. Hence question, about whether it is in fact technically able to do this.

I'm just wanting to build up a clearer picture of possible options, even if they are not necessarily the "ideal" ones, before reaching a decision about which to go for.
cnpope20 is offline  
Old 24th Jun 2019, 5:58 am   #5
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Racal RA17 alignment

Very strongly seconded!

The two VHF filters in Wadley triple-mix receivers are found in any of three conditions:

1) All is well. don't touch!
2) One of the silver mica capacitors has failed
3) Someone has tried aligning it and made a right mess of the settings.

For some reason, many people go straight for all the twiddlers whenever they come across a deaf receiver. Most often it's something else and given the difficulty of correctly adjusting those filters, they've just bricked an otherwise fixable receiver.

The filter adjusters interact diabolically and even with a swept display running, your attempts tend to diverge, not converge. The way out of this trap is to set the resonance frequency of each resonator individually with its neighbouring stages shorted. To understand this, you have to dig into the design processes for high-order coupled resonator filters.

It used to be possible to buy RA17s that had been visited by the phantom twiddler fairly cheaply. Their owners usually denied ever trying to align them... 'it's a bit deaf, probably a valve' would act as a warning flag for me. Sometimes this was a set suffering a resistor gone high. Whether the phantom had had a go at it could be assessed by how easy it was to haggle the price down If it was too easy, someone was having guilty feelings! After checking and resetting the two scary filters, you then had to trace and fix the original fault. Afterwards, you had a saleable receiver. When someone came along with an RA17 and 'fessed-up to twiddling the filters, I'd show them how to do it for free. I just saw it as a public service to make a bit of profit from someone trying to foist something they'd screwed up off onto unsuspecting punters.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 24th Jun 2019, 9:27 am   #6
Craig Sawyers
Dekatron
 
Craig Sawyers's Avatar
 
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,941
Default Re: Racal RA17 alignment

I recall that there is a thread somewhere on here where there is the shorted resonator trick - it could have been you that described it David.

Luckily mine, when checked with my SWOB 2, measured just fine. So no earlier phantom twiddler or dead mica caps in the 40MHz filter.

Once the Hunts were replaced and the odd madly out of tolerance resistor, it was ridiculously sensitive. One evening I pulled in Radio Australia like it was Radio 4. I thought it had to be a European relay - but no - emailing them I was listening to a side lobe on their antenna pattern that provided programmes in the general direction of Hong Kong.

Offhand though although a signal generator like the FY6800 can be used to check the shape of the top and down to ~-6dB frequencies, the rather critical rapid fall off at each side of the filter can't really be checked that way. And you cannot realign the 40MHz filter with a signal generator - unless David has a trick.

Ah - this is the thread - and it was David with the resonator trick https://www.vintage-radio.net/forum/...d.php?t=147825

Craig
Craig Sawyers is online now  
Old 24th Jun 2019, 2:26 pm   #7
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Racal RA17 alignment

Might just have something...

The AD8309 is a logarithmic detector, covers 5-500MHz and the manufacturers claim 100dB dynamic range.

In practice it floors out at about -70dBm

Needs a high impedance buffer or a good passive probe and very careful layout.

I've been using them for decades. Does just what it says on the can.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 24th Jun 2019, 3:30 pm   #8
Craig Sawyers
Dekatron
 
Craig Sawyers's Avatar
 
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,941
Default Re: Racal RA17 alignment

That is an impressive beast. I'm guessing that the individual gain blocks are Gilbert Cells, or some derivative thereof.

But yes - something this fast and non-linear needs very careful layout. But they are cheap enough to buy one or two to play with ;-) Which in practice means a custom SM circuit board.

What I think you are suggesting to the OP is a combination of the FY6800 in sweep mode with the AD8309 as a detector, and then use the shorted section alignment trick on the 40MHz filter?

Craig
Craig Sawyers is online now  
Old 24th Jun 2019, 5:15 pm   #9
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: Racal RA17 alignment

Quote:
Originally Posted by Radio Wrangler View Post
Might just have something...

The AD8309 is a logarithmic detector, covers 5-500MHz and the manufacturers claim 100dB dynamic range.

In practice it floors out at about -70dBm

Needs a high impedance buffer or a good passive probe and very careful layout.

I've been using them for decades. Does just what it says on the can.

David
In the next week I will be testing the bits I got from ebay comprising a 1.5GHz low loss probe and a board with a similar analog detector which came to a total of about £17 delivered (interconnect cable included). Both boards have a regulator so all I need do is wire up a rough DC power source and try it against my (uncalibrated!) 2022. I'll let you know how it goes.
PJL is offline  
Old 24th Jun 2019, 5:32 pm   #10
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Racal RA17 alignment

That was the sort of thing I was suggesting.

Re the 1.5GHz low loss probe... beware of the tip capacitance pulling things. Sampling through a high-ish value resistor (about 5k) mounted on the tip can be helpful.

If you're good and sensitive, a couple of turns near the L can pick up enough. As you bring the loop nearer you'll see the centre of the response pull, so look at how it moves and estimate where it must have been before it was pulled (easier than it sounds) and twiddle that stage to plant that peak smack on the geometric centre frequency.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 24th Jun 2019, 5:42 pm   #11
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Racal RA17 alignment

The log detector is a chain of differential amplifiers with controlled gain (feedback around them) and a carefully controlled limiting level.

At a given input level early amps are running linear at low levels, then comes one amp with plenty of output voltage and its detector stage contributing the the output adder, then there are more stages stuck in limiting with their detector giving a fixrd output.

As the input level is changed the switchover from no detection to linear detection to saturated, fixed output contribution moves to and fro along the chain like the slider on a zipper.

Added nicety is that the curvature of a real world diode detector helps blend the piecewise linear segments because there is overlap where one diode is working almost linearly while its neighbour is almost linear at the other end of its range with opposite non-lin curvature.

This is a process we've had to do in spectrum analysers since they began, but now in neat chips.

I made a 'VSWR meter from hell' for my kW size ATU. It used the older AD640 logger so I got forwards and backwards power meters with scales of dBWatts that were linear.
Subreaction of the two voltages gave me computed VSWR as a linear scale of dB return loss. To put the cherry on the cake, the limited outputs drove an ECL phase detector I'd made, so the thing gave me a reflection phase measurement.

RL and phase---- all that's needed to plot straight onto a Smith chart. Dave's vector VSWR meter!

G3ROO made a couple of sets of PCBs and I added peak reading for SSB.

Over the top or what?

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 26th Jun 2019, 2:52 pm   #12
cnpope20
Triode
 
Join Date: Jan 2018
Location: College Station, Texas, USA.
Posts: 11
Default Re: Racal RA17 alignment

Quote:
Originally Posted by Craig Sawyers View Post
I recall that there is a thread somewhere on here where there is the shorted resonator trick - it could have been you that described it David.

Luckily mine, when checked with my SWOB 2, measured just fine. So no earlier phantom twiddler or dead mica caps in the 40MHz filter.

Once the Hunts were replaced and the odd madly out of tolerance resistor, it was ridiculously sensitive. One evening I pulled in Radio Australia like it was Radio 4. I thought it had to be a European relay - but no - emailing them I was listening to a side lobe on their antenna pattern that provided programmes in the general direction of Hong Kong.

Offhand though although a signal generator like the FY6800 can be used to check the shape of the top and down to ~-6dB frequencies, the rather critical rapid fall off at each side of the filter can't really be checked that way. And you cannot realign the 40MHz filter with a signal generator - unless David has a trick.

Ah - this is the thread - and it was David with the resonator trick https://www.vintage-radio.net/forum/...d.php?t=147825

Craig
Thanks all, for some helpful advice. That link to a previous thread, with its discussion of how to align the 40 MHz filter, looks really useful, if it ever comes to that.

I'm becoming somewhat more sceptical about the FY6800, I must say, having read a review somewhere where it said that there is no sync pulse of any sort when operating it in sweep mode. Seems a bit crazy. A "kludge" solution was described, using one of the two generator channels to supply a ramp waveform to drive the other channel operating as a VCO. But the fact that one has to do that in order to have a useful sweep signal does not inspire confidence. So, if anyone knows of a better, but modestly priced, sweep generator then I would certainly appreciate hearing about it.

Meanwhile the RA17 has arrived, and it seems to be operating pretty nicely already, so there is no immediate need for me to do anything to it. I would still like to assemble the wherewithall for measuring the filter performances, though, just for the hell of it.
cnpope20 is offline  
Old 26th Jun 2019, 5:14 pm   #13
Craig Sawyers
Dekatron
 
Craig Sawyers's Avatar
 
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,941
Default Re: Racal RA17 alignment

Replace all the Hunts tubular metallised paper capacitors. However well it seems to work, it will work a whole lot better when they are replaced with axial polypropylene caps.

They are used for (a) local supply decoupling (b) screen grid bypass (c) Cathode resistor bypass, and a bunch of other places.

They will either essentially be resistors, or have nearly zero capacitance, or both of those. There was not a single one in mine that was remotely OK.

Craig
Craig Sawyers is online now  
Old 26th Jun 2019, 5:50 pm   #14
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Racal RA17 alignment

You'll also find a large proportion of resistors have gone higher in value. The circuit is surprisingly tolerant of this, but you can have a nice surprise from the improvement to be had in checking/replacing as necessary.

And if you have to visit all the nooks and crannies swapping capacitors then you're doing 80% of the work already.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 27th Jun 2019, 12:12 pm   #15
Skywave
Rest in Peace
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
Arrow Re: Racal RA17 alignment

Areas that require attention in an RA-17: I've mentioned these topics before, but it is no problem for me to mention them again.

One area that nearly always requires an extensive renovation is the 2nd. VFO. Due to the amount of heat generated within this 'box' and its lack of ventilation, upon inspection you will almost certainly find that most - if not all - of the components therein will require replacement. There is a group board with a few Rs and Cs in that VFO assy. which itself may be so badly burnt as to also require replacement. Occasionally, either - or both - of the two valves require replacement, also.
Overall, all of that work is not difficult. However, there is one precaution: do not adjust the core of the oscillator coil: it is fixed with locking nut to prevent that. (There is a trimmer capacitor provided for osc. freq. adjustment: 99% of the time, adjusting that is all that is necessary). But you can get the rare fault where its inductance does need to be reset and doing that successfully is not a trivial matter. Should you find that that is necessary, send me a P.M.: I have a tested & proven procedure for that adjustment.

Two other areas also need mentioning:
1. The output level of the 1 MHz crystal osc. is (not surprisingly) critical for satisfactory performance of the entire receiver: in brief, refer to the manual.
2. The 37.5 MHz LPF and its entire amplifying chain can be realigned satisfactorily without a sweep generator, etc.: just takes longer and a good deal of 'fiddling' is required. What is critical in the tuning of that LPF are the two coils and each 'end' of that filter. A lack of drive from that LPF can cause all sorts of problems. I do have notes on how much O/P from it is usually required. Again, P.M. me if necessary.

Finally, the 40 MHZ BPF: I've never tried this, but in theory, it sounds O.K. By using an accurately calibrated signal generator, it should be possible to assess if that BPF requires adjustment by simply measuring its response over its required passband, plus 5% (say) to assess the response of its skirts. As others have stated, the only way it can be correctly re-aligned is by using a sweep technique - and even then, certain precautions are necessary.

Al.
Skywave is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 10:11 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.