Recapping the Hallicrafters SX28 RF Box.

[ms660]

PART ONE

Hope this is of use to someone.

Before I stripped out the RF box I recapped the audio, bfo, IF, agc, noise limiter etc. The cathode bypass capacitors in the audio stage are contained in the same cans as the HT filter caps, the HT filter and resevior caps are originals and were fine (to date)..no hum, no bulges no pimples and no weeping, however I decided to disconect the cathode bypass tags and fit seperate new electrolytics instead.

All the wax caps in the above stages were replaced but the mica's were left in situ as these are normally quite reliable. While I was in there I checked all the R's for out of spec...all but a few were way out so they were also replaced and where neccesary the wattage was up rated, mainly on anode and screen feeds etc.

The main screen feed voltage divider network for feeding the RF and IF screen circuits is a long flat metal clad wirewound job that is riveted to the side of the chassis, it consists of one 15k resistor which is fed from the main HT rail (approx 280v) the other end goes straight to chassis, this resistor is tapped at 11k up from the chassis end, the 11k tap being the screen circuits feed (approx 100v).

On checking this it was found that the 11k section was going intermittantly high as the tag was wiggled, probably due to a poor internal rivet connection, I did not have an 11k so a 10k 10 watt was fitted (after isolating the original) at a later date I will fit a couple of 10 watts made up to 11k.

No major problems doing all of the above, taking time to figure out in which order to replace all the components seemed to pay dividends as things are pretty tight in there.
Powered up and everything was ok, the voltage across the common cathode resistor in the push pull output stage had dropped about 30% of what it used to be...which was good news. The S meter was now working very well as was the AGC and noise limiter, IF gain was also up. This part took me most of the day to do.

The next day I decided to have a ponder at the RF box and after a couple of hours I decided to go for it. There are a couple of articles on 'tinternet describing the best way to tackle this part of the receiver and in the main are well worth looking at. In a nut shell it is impossible to replace anything in the first RF stage without removing the coils, the same goes more or less for the second RF and also the mixer.

So here I describe what I did and what you need to do to gain access. First pull the RF, Mixer, Osc. valves then unsolder the four wires that feed the last three (rear) sections of the main tuning gang. Unsolder the six wires that feed the last six (rear) sections of the bandspread tuning capacitor. Mark all these wires with coloured marker pens in a code that suits and make a drawing of what goes where including which holes in the chassis they come up through! Straighten the wires so they will pull through easy (later)/

Remove rectifier valve and turn the chassis upside down and front facing, remove band change knob, slacken off all collars on the band change and aerial trimmer shafts, note that each one has a phosphor bronze spring earthing washer behind them. Undo the aerial trimmer shaft at the flexible shaft coupling then withdraw the shaft completely, make sure you retrieve all collars and washers..note which order and where! Remove flexible shaft coupler from aerial trimmer capacitor.
On the rear apron of the chassis mark an orientation line on the rear of the band switch plate and the chassis, remove the two switch plate securing screws.

Carefully withdraw the band change shaft from the rear, DO NOT LET IT TURN WHILE WITHDRAWING give yourself plenty of room as it's a foot long, retrieve all collars and washers.. note which order and where! Remove the two aerial terminals, mark them and the connections.
Slacken off the phono input socket, remove all the hex screws from around the uppermost edge of the RF box including the two that go through the rear lip of the chassis.

Remove the Philips screws that secure the two side screens to the chassis (four each side) you will probably have to prise the wiring loom to one side to access some of these.

On the right hand screen plate there is a nut which anchors the second internal screen, if there is any spare "stud" protruding from the nut then cut it off (clean up any filings etc) remove nut.
There are two wires that feed through the left hand screen one from the first RF compartment and one from the second RF compartment, they are both right down at the bottom of the side screen, they go out to a tag strip on the chassis, they both go to the same tag they are the "earthy" end of the grid coils for the first and second RF stages both of these wires need to be disconnected from the tag, mark them and make a note! straighten them as best as possible as they will need to be pulled through the side screen later. Similarly there is a wire that goes through the right hand side screen from the second RF compartment to a tag strip, it is also down at the bottom, this is a HT feed, disconect and mark etc etc.

There are two wires that go from the first RF compartment to the second RF compartment through the dividing screen at the top, they go to a band change wafer in the second RF compartment, disconnect them from the wafer switch in the second RF compartment mark them and make a note! It is best to snip the connections to the wafer than rather to risk any damage that might be caused by excesive overheating/mechanical strain etc.

There are three more wires that also feed through this same screen they are down in the bottom right hand corner, two of these connect to a bandswitch wafer in the first RF compartment, disconect these at that wafer using the same method as previously described etc etc. The third wire of the group goes to a tag of the coil that is furthest to the right in the first RF compartment, it is the tag at the bottom right hand side of the coil nearest to the dividing screen, this wire should be sniped and marked etc etc. For the sake of this article I shall call this wire, wire "A". Straighten all these wires out enough so that they can be easily be pulled through the grommitts without to much strain.

OK...First RF removal now...Remove the four Philips screws securing the rearmost divider, they are down at the bottom of the second RF compartment, two on each side, also remove the similar screws that secure the next divider, they are in the mixer compartment.

The left hand screen needs to be tilted towards the left, its top lip fits under the lip of the rear chassis and is a "dead fit" use a suitable peice of wood as a drift and place it on the edge of the lip and a few light taps with a small hammer will move it, it needs to tilt over far enough so its lip clears the dividers plus a bit more if it will go.

The right hand screen needs to do the same but to the right, it will not tilt as far due to the ext. power socket, before attempting to tilt this screen over you need to make enough clearance so this side screen will clear the "stud" encountered earlier, this can just about be done (not much room!) by pulling back the wiring loom where it crosses the this side screens bottom lip down at the front of the screen and at the same time lift the front of the screen up and over to the right untill it clears that stud. Once this done tilt this side screen to the right as far as it is safe to do so.

OK...the first RF stage divider and all its coils and switches will now just about come out in one lump, the bottom lip on this divider goes under the coil plates of the second RF compartment, first wiggle this divider a bit to lossen things up then try and slide it to the left as far as it can go, then tilt it forward and at the same time push the bottom out rearwards until the bottom lip of the screen is released from under the the coil baseplates, use a reasonable amout of pressure if need but be very carefull not to strain the switch wafers and coils.

Once the bottom of the screen is free then it's tilt turn and pull and it will come out in one go..brew time.

Part 2 to follow.

Lawrence.

[ms660]

PART TWO

OK..Removing the first RF section is the most difficult job and with this now out of the way removing the second RF is made a lot simpler.

There are a couple of extra things that need to be done, the first is a wax capacitor that runs from one of the RF coils to a tag strip in the rear of this compartment on the chassis, it hooks itself over one of the main bus wires that is common to the coils it is easy to spot when trying to remove this assembly and will need one end cut at that tag, make a note etc. The other is a wire that goes from the bottom of one of the switch wafers to the second RF valve base, it's approx. one inch long and I called this wire "B" I snipped this in half and made a note. With these two out of the way it is now realatively easy to remove the second RF section, take care again with the switch wafers and coils.

Once it is out replace the capacitor on it that was snipped at the tag just mentioned above, leave the left hand lead of the new capacitor free and full length.

Replace all capacitors and R's in both sections make notes if need be when removing originals.
It is important that any replacements do not protrude further in any plane than the originals otherwise problems will arise when trying to refit the sections back into place.
Double check everything that is replaced using a meter if need be to check routing, continuity, shorts etc etc. Replace all wax caps on the coil and switch assemblies, check for any solder droppings floating around.

...Brew time.

OK... the next is the mixer stage, some reports say this section has to come out, some say it can be done in situ..it can be done in situ and this was the method I chose.

Some components cannot be accessed for unwrapping their connections or even accessed for snipping, in such cases it's a matter of gripping the components wire with long nosed snipe pliers and use the "wiggle till it breaks" tecnique on their wires. a lot of the new component wires will have to be a "laid on joint" to their connections, tin everything well first and check the joint with a magnifier and do a "pull test" with a pair of pliers, make sure that any connections to valve bases are really secure as these will undergo mechanical movement when replacing valves etc. It will be seen that throughout a job like this some of the components can be re routed for convienience especialy to different available earthing points but bear in mind that this receiver tunes up to 40 megs.

A quick word about the replacement components:
Try and source new caps that are smaller than the originals but of the same or greater working voltage I used Vishay Roderstine film capacitors from Farnell they are yellow with axial leads that bend easily and are nice and compact, they do a 630v series which were the ones I used throughout this job, all the replacement R's were film type, I would not recomend using carbon composition types.

OK..Time to put everything back in reverse order. Regarding wire "B" it can stripped on both halves, cleaned and tinned and rejoined when the assy. is back in situ with a small soldering iron (I used my Antex XS 25 for all of this work) this method leaves a handy test point, or you can replace the wire completely, if this is done cut a new piece the same length + .25 of an inch, prepare and tin both ends, solder one end to its valve base connection and solder the other end to the wafer switch once the assy. has been fitted back in place, you can just get the soldering iron in there with a little care.

Regarding wire "A" mentioned in Part 1, snip its connection to the coil in the second RF assy. and replace with a new piece the same length + 0.25 of an inch, strip and tin the free end.
Refit the second RF assy but do not tighten the screws fully, connect the free end of the capcitor mentioned earlier to its chassis tag, make sure all the relavant wires have been poked through the side screen grommets and up through the chassis grommets to the tuning gangs, tighten the fixing screws and check that none of the interconnecting wires that run down the centre of the chassis are not trapped (ie; wiggle test) under the centre of the screen.
Slacken off the screws again. Resolder wire "B" joint or solder newly installed free end of wire "B" to wafer switch (which ever method was chosen).

OK..Now refit the first RF assy. making sure that all wires are poked through their respective grommets and that nothing is trapped etc. Reconnect the switch wafer wires and connect the free end of replacement wire "A" to its coil tag, secure the assy. with its screws but do not tighten. Reconnect the aerial terminals and connections, refit both side screens and refit all its screws but do not tighten, refit the nut on the right hand side screen to its "stud" but do not tighten. Reconnect the wires that come through the side screens to their respective tags.
Wiggle all the screens etc so they lie in their "natural positiion" then tighten everything up.

Brew time.

Part 3 to follow.

[ms660]

PART THREE

This is the last part of this article.

OK..Re tighten phono socket and check that there are no wires trapped where the side screens meet the chassis at the rear, on the left hand side screen there should be a gap between it and the rear of the chassis of a few mm. there are two wires that pass through this gap, give them a wiggle test to see if they move freely, there is a similar situation on the right hand screen but this has a dedicated cut out.

Re insert the band change shaft from the rear making sure that the orientation mark is the right way round, DO NOT TURN IT when re inserting.

If all is ok it should pass through all the wafer switches, if it does not then a slight joggle should suffice. Make sure all the collars and washers are re fitted back in their correct order, when fully home refit the two securing screws on the rear chassis apron, refit the knob and check rotation.
Refit the flexible shaft coupler to the aerial trimmer capacitor, refit the aerial trimmer shaft with all it's collars and washes, check rotation and knob alignment.

Turn chassis over and resolder all wires to their respective tuning gang connections, if any of these wires were snipped instead of unsoldered then fray the end of the cloth insulation and grip the end of the wire with pliers and pull back the cloth, it will pull back enough to clean the wire and re tin.

OK...Valves back in, tip chassis on its side and check for shorts or unusually low resistance readings from ht and screen divider resistor to deck.

Switch on time...in my case everything was ok...RF gain up considerably, I put this down to the fact that cathode resistor in the first RF stage was completely open circuit. Noise floor down as far as I could ascertain, receiver a lot more perky now.

All RF coils and trimmers needed a provisional re tweek as did the Osc. but to a lesser extent.
The Osc. section was not stripped out, so far as I could see there were no succeptable capacitors in that section and the couple of internal resistors that are in there were tested in circuit via the topside of the valve base and were low value types and more or less within spec.
Pulled in a Stateside Ham in Ohio late at night up on 20..good enough for me for the moment.
Forgot to mention in Part 1: remove any grub screw burrs from the shafts.
The RF rebuild started at 10 am and was finished at 12.15 am.

Should Have: ...Printed the circuit diagram out on a large scale and indicated the component values and valve pin numbers on it ..Oh well.

Any mistakes etc in this article...let me know.

Brew time.

Lawrence.

[ms660]

Thanks Mods/Admin for sorting my posts out.

Lawrence.

[MichaelR]

Nice piece of work, will be interested on your thoughts on its performance when you have had time to"shake it down" for a week or so.

Mike

[ms660]

Will do Mike...I have do not have a spectrum analyser or noise measurement gear or anything like that neither do I have a "modern" receiver to compare it too although I was thinking of picking up a Frog 7 as they seem to go for reasonable money these days.

So any obsevations will be comparative, the lowest noise/sensitive receiver I have to compare is my old B40D which is all up and running followed closely by the HRO's, plus my aerial is crap at the moment hence the word comparative, the S meter circuit is such that it cannot be used with the BFO on the SX28 as it is switched out of circuit on AGC off and BFO positions, I doubt that if it was switched in that it would say much on the kind of signals we are talking about, having said that it is now working as it should on AM wether it be medium wave or short wave broadcast stations or CW.

This receiver is flavour of the month at the moment as I like to have my kit up and running to original spec as far as is practical so I can enjoy using them, they are no use to me sat on a shelf gathering dust. But as always it will depend on band conditions, there were quite a few up on 10 the other day when a competion was on also on 80 which is usually bad where I am, top band will be a no no though as the noise is to bad at my location plus the !st RF is switched out on MW and the first SW band on the SX28.

Thanks for the interest.

All the best.

Lawrence.

[MichaelR]

Hi Lawrence if you are listening to SSB turn off the BFO and use the agc the S meter should bounce up and down to the sideband not ideal I know but it does give you a feel for the strength of the signal , I often do this with my HRO mx ( with the HRO keep the rf gain set to 9.5 having zeroed the "s" meter with no signal in).

Mike

[ms660]

Thanks Mike.. I will give it a go, the other thing I need to sought out with it is to run it off a seperate power supply to do a proper test, the SX 28 is a dual voltage and plugged straight into the mains it gives 340v HT at the resevior cap. and a measured 6v on the heaters, the ht is roughly 40v over spec so in the interests of the original filter/smoothing caps I am at present running off a variac set to produce the correct HT of approx 300volts, obviously the heater voltage has reduced and is down to five volts, I have tried switching the voltage selector to 110 and run the receiver at 110 ac in.
The results are pretty much the same ie:to much HT, my other Hallicrafters does the same so I guess its a common trait. I will run a seperate PSU I have off the variac to give the correct HT and feed the heaters from a seperate heater transformer I have, all this can be connected via the receivers Ext.PSU Octal socket.
By the way are your HRO set zero pot shafts live? Mine are, I do not know if National modified these at all at a later date to make them a bit more friendly!

Lawrence.

[MichaelR]

Yes Lawrence the set zero pots are live no "health and safety" in those days, "men were men". I read somewhere that quite a few of the Bletchley park operators received a "belt" from the HT on the link at the rear of the set when they were moving the sets on the bench powered up.

I would not worry about being 40Volt over on the HT, remember with these old sets that when they produced the manuals with the voltages listed they were using meters that probably loaded the circuits more than you will with your modern meter.You are about 13% high

An important check is making sure that the working voltage of the reservoir cap is more than sufficiently high and the heaters are not being over voltaged.

Mike