Monday 27 September 2021

Collins PM-2 rebuild - Decision made !


As per my previous blog, I've been thinking about what to do with the PM-2 rebuild, rebuild as is or rebuild to the most recent service bulletin.

After some discussions with other Collins restorers who gave some great advice, I have decided to rebuild to the service bulletin PM-2 Service Bulletin 1. 

This improves and extends the input voltage limits, relocates the electrolytic capacitors to under the speaker and under the tagboard, rather than having them all mounted underneath the tagboard, I'll also upgrade the diodes and put 105 degree capacitors in as usual. I will also take the opportunity to change or rebuild C7 which is a dual 40uf electrolytic capacitor on the +275 V line. Not decided if I'll restuff the original can or find a suitable modern equivalent - I've used 50uf/500v dual ones from Cricklewood electronics previously to good effect. 

If you saw the previous blog post you will have seen that the tagboard was not re-useable - it had been overheated badly and repaired (poorly) several times.

So decision made - the first thing is to re-create/remake the tagboard - I always try to make it as close a possible to the original so I sourced a suitable piece of 3 mm Phenolic resin sheet, I couldn't quite find the right colour its a too dark but the right thickness. I also found the right tags from RS Components, or very close to the original ones under part number 433-589

So after some careful measurement of the original board and some drilling I had this ;

I'm no expert and as you can see the phenolic resin sheet is a little brittle and I clipped one corner (top left) but overall not too bad. From this, the tags/pins need to be pushed into the holes. The pins come with a punch which helps to do this as otherwise, you will damage the pins.

So after some careful work, I ended up with this ;

So now I'm ready to start putting together the board ready for the rebuild, but that will be in another blog post.

Overall just under a couple of hours of careful but pleasurable work to get this far, though its not a race !

Sunday 19 September 2021

Drake TR-7 Repair & Restoration - Part II

A short recap

I bought a non-working TR-7 from a well know website, it received quite well, but the digital display didn't work amongst other things.

After my initial fault find and usual cleaning of connectors and interconnections, the fault remained. I have a large stock of TR-7 parts including the DR-7 display board. I swapped it and sure enough, the display worked - so it's the DR-7 that is faulty.

So faulty board back into the unit. Next check that the switch on the back of the unit which allows the counter to be used as a frequency counter (a nice touch that most people forget about!) - check that it works - the switching worked correctly- on the display when switched to external the display should blank - this did for the most part - but for the 10's digit which showed a 0

The TR-7 has a service manual which is very good - easy to read with good circuit descriptions (in my opinion!) - I'm lucky I have an original one which is wonderful but some good scans are available if you look carefully - it does recommend that you have a scope to fault find on this board with 100Mhz I can vouch that if you don't it might be a little tricky as the waveforms really do need that sort of bandwidth.

I have an order which I check the DR-7s , effectively, check voltages, and then work backward from the display - multiplexers, and drivers first, then timebase/gate, dividers and finally if needed input controllers and other inputs.

So after checking the voltages (all ok) I skipped the display and multiplexers as the display was working (although not displaying the right frequency of course!) and started on the timebase /gate.

I start by checking the 500Khz input (present) and then work through the dividers, first one -U9014 divides by 5 (100Kz or 10us) - this was fine, then further divided by U9017, 9019, and 9022 to divide by 10,000. 

Check the output of U9017 - all OK - Check input into U2019 (pin 10) - all ok.

Check the output of U9019 at the point indicated by the manual (collector Q9003) - nothing like the right waveform - ahhhhh.

Work back - check the output directly at U9019 pin 14 (q4) - OK , pin 13 (q13) - nothing

Now without the pin 13 output via the two diodes (D9003, D9004) the transistor won't switch properly and so the following divider stages (U9020)  and U9018B won't work correctly. 

So this definitely cause problems with the reset and loading pulses which would explain why the digital display wasn't working perhaps, it wasn't loading and resetting correctly! Lucky I guess that it was this early on in the check.

I carried on and checked the gate / timebase area, as well as the input and internal/external switching, were OK - which they were.

So I ordered the correct chip - 14158 CMOS chip - MC14158BCP which is a dual BCD counter - I even managed to find a Motorola version of the chip which was the same brand the original ones, I order two just in case at the grand cost of £5.98 in total!

Chips arrived, chip removed, and carefully replaced with a new one ic (Circled below) before cleanup the diodes are just to the right of pin 14.

Interestingly enough the date code on the other chips are 7904 making them 1979 vintage with the new chip being 1998 vintage putting them 24 years apart !

And so carefully put the DR-7 back into the TR-7 and .........


I checked it worked on all positions of the band switch - it did and that the up / down / store buttons functioned correctly - they did.

I then check the rest of the divider chain - and as expected all waveforms etc. were as per the manual

So now that's fixed onto the next stage for repair - perhaps the store button (you can see top left how it's broken) together with checking the transmitter and full re-alignment, together with the change of the blue film - this one has faded rather badly.

However I spent a lovely hour or so listening around on 80M, 60M, and 40M, the receiver is sensitive and all other functions PBT, etc. work well even the USB /LSB balance doesn't seem too bad!

Overall about an hour to find the fault and about 45 mins to fit the new chip, put the board back in, and finalise testing.

Friday 17 September 2021

Drake TR-7 Repair & Restoration


As you may know, whilst I repair most equipment, my real specialism and passion is the repair and restoration of Drake and Collins equipment.

From early valve types right through to the last equipment made, I've spent many happy hours on customers' (and my own) equipment restoring or repairing them, and when I get offered or see such equipment to purchase then I will usually do so - depending on the price of course!

So when some Drake equipment appeared on a well know internet website for sale, I made an offer and I was pleased to be notified that it had been accepted.

It was (more or less) a complete setup with a fairly late model TR7 transceiver (1337 model - originally didn't have the DR-7 but obviously had one added later), RV-7 Remote VFO, PS-7 power supply, and a drake branded Astatic microphone. Its serial number puts it as being made in January 1980, the remote VFO serial number puts it as a similar date of manufacture with the PS-7 serial number as July 1979 - so perhaps the TR-7 and RV-7 were bought together with the PS-7 added later - or perhaps the PS-7 was just slower to sell!

All were in fairly good condition but described as non-working, the seller tested them fairly comprehensively and could resolve signals and get good TX power out of it, but no SSB audio on transmit and the digital display didn't work (displayed 51985) correctly together with the remote VFO didn't appear to function several of the aluminum discs were also missing on the main TR-7 , the VFO was 'gritty' and it was also missing a button. 

From the pictures, it looked like it had been stored somewhere dusty (garage most likely) as some of the screws on the pictures were a little rusty (!).

So deal done, and with eyes fully open (!!) I paid for it  -  it arrived safely and initial review and testing took place.

Sure enough (and somewhat unusually) the description was entirely accurate!

It did look dusty, but more importantly, it was entirely original, all the screws were original and in the right place - this is a pet hate of mine on Drake equipment especially TR-7's some screws are self tappers and some are threaded - they obviously do not go in each other's holes but some people don't realise and swap them about this causes problems with both the screws and the cases and in some cases, the only way to get the case (especially the bottom panel) is to rivet in captive bolts as the holes are so badly mangled!

Anyway, not this one, it looked like it had never had the case off (or who did carefully put things back the way they should be!

With the TR-7 I have developed a standard approach for anything that isn't 100% working (ie pretty much everything that comes through my bench!) and that is to ;

  • Remove all plug-in boards and clean pins and sockets, use Deoxit (D100L) to aid cleaning of the pins, etc. and then reseat them carefully.
  • Check and clean all the mini-coax interconnections - ensuring that the earth connection and the middle connection are both clean and tight - also use Deoxit in the center pin socket on the PCBs. In the TR-7 (and Ten-Tec amongst other radio) - there is no 'pin' on the coax it is just the inner connector of the coax which pushes into a small socket on the PCB connection so it is always worth checking that these are clean and the coax center is actually positioned correctly into the socket - I've had several come to me with intermittent problems that the owner has been driven mad by which were a combination of dirty coax and in two memorable incidents where the center coax was bent back and so wasn't inside the PCB connector and occasionally would make a connection !!
  • Check all the controls in a similar way carefully using Dexoit to clean the potentiometers tracks where possible, together with the relevant switches, etc.
  • pay special attention to the DR-7 board which has Molex connectors which 'push' down into the digital and other boards below it, in my experience simple poor or oxidized connections cause many common faults on the TR-7 series of radios.
  • Clean all the switch contacts with isopropyl and then use Dexoit (D100) carefully to fully clean the contacts. I don't use the  Dexoit D5 as this has a flushing action which is great but sometimes can damage the SRBP or whatever it is.

After doing all this in around three hours - out the radio back together - and tested it, it was certainly improved with all functions working nicely - except the display which remained stubbornly at 51985, this didn't surprise me as previous experience on such a display problem showed its usually a logic/counter problem or chip failure.

I have many spares for the TR-7 series including a few DR-7 boards and so to rule out interconnections etc. I used a known good spare DR-7 board to check.

Sure enough, the display worked fine with the replacement DR-7 so that located the fault on the DR-7 board nicely! - it's worth remembering that the counter/decode circuit is quite straightforward but can be quite tricky and so without a scope its sometimes difficult to locate the problem - unless it's a connection problem of course.

In the next blog, I will cover the diagnosis and repair of this board with some tips on where and what to look for to locate the problem.

Friday 10 September 2021

Kenwood TS-711 - Repair of VFO


I think is quite an interesting repair and restoration and so I thought I'd write it up as it might be useful for someone else !

I got hold of this Kenwood TS-711 with an interesting and slightly confusing fault list ;

  • The radio has been modified and has 'additional relays' located within it (!!!)
  • Off Frequency by approx 200hz
  • S Meter background illumination not working
  • VFO tunes but clatters ans clicks at switch on and when non channelised operation (VFO) is selected remains in channel 'click' mode
So nothing particularly unusual for the first three - all easy enough, remove modifications to restore originality, re-align the transceiver (I'd do that anyway!), repair the s meter backlight, but it's the last one that is more interesting - what's that about?

For those readers not familiar with the TS-711/TS-811 range of radios they are a 2M (711) and 70cm (811) multimode transceiver. They can be mains powered or run from 13.8V and are overall even today a very nice base station transceiver. They output around 30W are sensitive and can do the appropriate 12.5Khz channel spacing and with one of my CTCSS units fitted (or an original one with some modifications) will enable use through repeaters.

They do have an interesting tunning operation and that is that the tunning dial has a solenoid on it and when a channelised operational mode such as FM is used (or you can do it manually by pressing the CH Q  on the front panel) you heard a fairly loud 'clunk' and the dial then has a 'click' or indented feel to it. When you select SSB or CW or press the CH Q button again you hear a 'clunk' and the VFO has a 'freewheel' feel to it or like a normal VFO.

I can't think of any other way of describing it!

It's really quite a nice approach and as far as I am aware unique to just these models of transceivers.

So upon receipt, sure enough, it's off frequency, the s meter light doesn't work and it has two relays quite nicely mounted inside, and when turned on there are several clunks and then all goes quiet.

It does tune correctly but only with the 'clicks' on the tunning dial - pressing the CH Q button or selecting SSB results in some clunking but no change in tuning style from click mode to 'normal' vfo mode

So I spend some time trying to work out what the relays did - in the end, I gave up - it looked like it was applying external modulation to the microphone for some reason - but easy enough to remove just in case it was having an effect on the other faults - I removed the modification easily enough - so back to standard, no change on any of the other faults.

I replace the S meter illumination with a standard Kenwood backlight bulb - an easy fix!

I'll realign the transceiver at the end so that left the dial issue.

Thinking about it, I'd had one 711 many years ago which didn't quite have the same fault but did have intermittent problems in this area and it turned out to be a solenoid-related problem, so I thought I'd start there.

The solenoid 'frame' is connected to the main VFO encoder and can be removed fairly easily, you need to remove all the knobs and then the front panel - the front panel can then be folded down and by removing two screws from the front and the large but the VFO and solenoid unit can be removed for inspection (after of course removing the relevant leads from their sockets!).

So after doing that I had a look, the design is ingenious a solenoid is used as a piston to move a star-shaped 'cog' around which triggers but a position indication at the right time and also removes the 'click' from the VFO, like I say ingenious.

Well this one wasn't moving freely and was very stiff, and this was what was evidently causing the problem - the solenoid tried to move the cog, failed two or three times as it was stiff and then the CPU stopped it doing it as the position sensor (a microswitch) didn't give the right signal at the right time to indicate it had moved.

At least that's what I think it was doing !!!!

So how to repair it?

Initially, I disassembled the solenoid, piston, and other elements, take care don't lose anything there are a very small circlip, a washer, a spring, and another washer. lose them and you're sunk 

After removing them you can remove the 'cog' which does the movement and triggers the microswitch - no need to remove the microswitch (or at least I didn't)

This is the bit that took me some time to work out - the cog is on a spindle that goes inside the VFO encoder housing and it was jammed - or at least very very stiff - I tried to lubricate it carefully but nothing would allow it to move.

So after a little thought, I decided to remove the VFO encoder PCB (four screws) to see what was on the inside of the VFO encoder, after carefully doing that I could see that the 'spindle' actually had a small square piece of plastic that turned and 'lifted' a small piece of metal that provided the 'click' in channelised mode - very clever !!!

This spindle was so stiff it wouldn't move, but after some careful manipulation, I succeeded in removing it. The spindle had a blackish tinge to it and so I cleaned it up with some wire wool and also the hole that it went through as well. After replacing it, it moved very freely and nicely so with a small amount of oil to help it, I put the encoder back together again. Be careful if you push the spindle too far it will drop back into the encode and you will need to disassemble it again !!

I then rebuilt the unit in the reverse order, take extra care to ensure that the circlip holding the 'star cog' is in the right position or the solenoid won't work correctly and once this is back ensure that the washer, spring, washer order is correct to apply a small downward pressure on the metal arm - and the circlip is correctly fitted on top of this - or it will jam up.

It is a complicated little piece of engineering!

When complete you should be able to push the solenoid arm backward and forward easily which will move the 'star cog' and trigger the microswitch - difficult to describe but when you see it - it makes perfect sense!

With that, I put the unit back into the front panel and reconnected the encoder, etc. but before I put everything including the front panel back on I tested that the repair had worked.

Applied power and no clunking noise (a good thing) press the CH Q switch a 'clunk' and the VFO was free - press again and 'clunk' it was back to being a channel switch with steps.

Excellent ! - all working and just about two and a half hours in total, not including an alignment of course!

Now just to do a straightforward alignment - I checked the frequency which was a little high and it will be completed.

Monday 6 September 2021

TS-50 repair continued


I promised an update on the TS-50 with the catastrophic audio switching transistor failure, and here it is ;

The original transistor is virtually unobtainable and so I found an equivalent transistor ZTX851. Interesting to note it has the same specification (actually slightly better) but is a lot smaller - I guess the onward march of technological improvement !

Replaced the transistor (if you do this watch out as it's not a direct pin-compatible replacement you'll need to carefully swap the base and the collector) carefully insulating it to ensure no short circuits.

I could have possibly shortened the leads a little but decided to leave them, they are perhaps a little long, to be honest, but no great matter.

So put it back together again and......

Still no audio 😔

But.... now I have the correct switched +ve on pin 8 of the amplifier chip so some progress at least!

Essentially only one thing left which is that audio chip itself - all other static components and interconnections/voltages were correct and so I ordered a new upc1241H audio chip from Cricklewood electronics, I looked elsewhere and they were by far the cheapest for an original chip even with the postage - but I also needed some HV electrolytic for the collins PM-2 rebuild and so I even got free postage !

Chip arrived the next day (great service guys!) and was quickly replaced, with the resultant bags of audio on the test, don't forget to put the clip back on the audio chip so it presses against the chassis heatsink (I always put a little smear of heatsink compound on it - not sure it actually helps but makes me feel a little better!)

So a little bit of an odd one, often replace the electrolytics due to leakage and high resistance but this is the first time I've ever had an audio chip failure.

My working assumption is that the chip failed and drew some much current it took out the switching transistor - however looking carefully at the audio chip I can see no cracks or damage at all - I would have perhaps expected that looking at what it did to the transistor !

Now all back together and on the bench for a soak test...another one of these great little radios back on the air

Sunday 5 September 2021

Collins PM2 Power Supply - Rebuild to service bulletin or not ?


Well, lots have gone on since my last blog, lots of repairs, lots of family stuff, and of course COVID-19 - arrgggh !!!

I thought I'd try and share some more of my experiences, mostly to record them for posterity, but also to help others as well a little, so here is a new post ;

I've blogged before on a PM-2 Upgrade, I've done several of these now for other people - contact me if you are interested! - but I recently got hold of a PM-2 which was clearly not working, thankfully no one had plugged it in, most likely as the cord was in a terrible state - just take a look ;

Looking at the date code on the capacitors this unit is quite an early one - 6047 for several of them which makes the capacitors made in late November 1960 - not surprising that the cord has now perished beyond any use after 60 plus years  !!

After removing the case several things were obvious.

There was no way that it could be used without a major and complete rebuild!

Looking at the photo (right) you can see how badly overheated the mainboard was - both the components and the board itself is pretty much shot and is not usable in my view. It's not too clear but the board is badly charred, possibly at least twice with the board being scraped of the carbon by the look of it !

I know that these units were given a hammering in the field and then also again in private hands, don't forget these are designed to be 'lightweight use and not a replacement for the 516-F - but are often used as if they were, unfortunately. It looks like this one has had rather a bad time, been repaired several times, one of them badly, at least in my view ! The capacitors are also showing signs of stress as well, not surprising really when they are nearly 61 years old -  just take a look at the first capacitor after the smaller 10uf ones - can you see the bump ? - definitely not to be used or reformed!

The unit itself looks like quite an early one - it doesn't have the service bulletin applied which moves the capacitors from below the board to under the speaker so that places it well before 1963

So what to do ? - should I try and salvage the board and rebuild it as is or should I update as per the service bulletin and make a new board that replicates the original collins board ?

I've asked for other collins owner's and restorers' views on the collins reflector group, but ultimately I'll make the final decision 😀 - watch this space for the next update.

Friday 3 September 2021

Two Kenwood TS-50 repairs - capacitor problems and a catastrophic transistor failure !!!!


I love the TS-50.

They are really great radios - work really well and are very rugged, but as per many radios of this age they do suffer from capacitor break down / leaks which causes odd problems both by damaging the PCB traces but also the PCB through connections as well !

I recently picked up a couple of these both non-working in one way or another ! - The first one looked to be working quite well but had no receive audio and the other was working quite well but with no backlight and was making a funny buzzing sound - not from the speaker but from the display !

I decided to have a look at the TS-50 with the display problem first, 

After carefully dismantling the front panel, as perhaps expected it was a capacitor that had/was causing the problem (below)

I've circled the offending capacitor - the connectors are corroded and its leaked badly onto the board and is also affecting the transistor next to it as well as the PCB itself.

The capacitor was so badly damaged that as I touched it - it fell off entirely leaving just to pcb connection on the board ! - see picture right.

Anyone that works on these (or other radios with similar problems) will know how difficult it is to clean up this type of corrosion - and will recognise that smell (I think it's a little fishy) when you solder and clean up any capacitors that have leaked !

After some careful cleaning work around the capacitor and removing the transistor to try and ensure removal of all the mess and also to take a very close look at pcb tracks to see if they have been damaged at all - not that unusual really the capacitor leaking can cause massive problems!

So having done that it was quite clear that some damage to the capacitor mounting lands had been damaged and there was some erosion of the small tracks near there as well. After some careful cleaning and examination, the tracks were actually not too bad, and so I cleaned and tinned them in order to protect them and give them a little longer life hopefully !

As you can see - I also re-worked and remounted the transistor as well - I checked it was OK with the transistor tester, it was and so after cleaning and tining it and its pcb mounting location I resoldered it back into place.

You can also see the cleaned and tinned tracks as well in this picture - thankfully they didn't need re-working as once clean they still had continuity - good news!

This just left the mounting of a new 47uf capacitor to replace the one that had caused all the problems! Sometimes the leaking can cause real problems in soldering capacitors into place, as was the case here, it needed to be cleaned several times as well as some burnishing with a fiberglass pen to enable a clean surface to solder too. 

The picture right shows everything after mounting and tidy up but before final cleaning, as you can see far better than the original!

All that was left was to now replace the backlight bulbs - so I removed and checked that they were blown- they were. 

I don't like to replace backlights with LED's which seem to be quite popular - for me, they seem too bright and lead to a 'patchy' illumination style and are too harsh to my eyes, and so I replace them with similar original bulbs they are quite straightforward to do and provide consistent illumination as per originally designed.

This TS-50 also had some quite bad screen marking/scratches and so needed some extra work using specialist plastic polish I have to remove the scratches and put it back to pretty brand new condition - it takes time - about an hour of careful work but it gets all but the very deepest of scratches out - with this unit it now looks like new.

That just left cleaning the rest of the front panel whilst it was disassembled and of course the case as well - something that I always enjoy doing as it brings everything together again and for me concludes the repair. Final testing took place - including checking the brightness controls works correctly as well as full receive and transmit. - All worked well and so chalk up a successful repair! - I have ordered a complete set of replacement surface mount and through-hole capacitors, as they will need replacing based on the condition of the display capacitor - I'll write that up in another blog write-up when I get the chance!

The second TS-50 I acquired is in good condition but with no receive output. I tested it upon receipt and sure enough transmit is fine a signal is received on the S meter display but no output on the speaker.

Usually, this tends to indicate a problem with the capacitors leaking badly around the audio amplifier in my experience which means replacing all the capacitors (as usual!), I checked visually and sure enough, there was leakage evident - not too bad but enough to warrant a complete replacement of all the capacitors but rather than jumping in and starting to do that straight away I thought I'd trace the audio chain through to check and see if there was anything else that I'd need to repair.

Upon tracing out the audio circuit, I work from the speaker backwards when the receiver is working - don't forget to check the speaker (yes some do open circuit) but not in this case also the rear speaker connector as sometimes they get damaged and this prevents audio to the speaker - but not in this case -  No output from the audio amplifier so check at the input (using a scope) on pin 1 and yes there is input to the amplifier which goes up and down in level with the AF front panel control which is good news, so I then check that the amplifier has the correct voltages - and .... no voltage on pin 8 (circuit below copyright Kenwood)

So check the voltage on the input to the switching transistor (Q13) and sure enough, there is the correct voltage but no output and so it's likely faulty.

Now Q13 is located right at the back left-hand side of the TS50 and is tucked away so I had a closer look (after my cursory usual check around) and what do I find when I look more closely?

Wow a lovely circular hole right in the middle of the switching transistor !

No wonder it is not switching! - this is a 5A 50V switching transistor which is generally not available anymore and so I have to order a suitable equivalent replacement which I am waiting to arrive. It does pose the question of what caused it to fail in such a catastrophic manner and what might it have damaged during its failure ?

I've seen such failures before but never in a TS-50 and so this is a little unusual in my view.

I will go ahead and replace all the surface mount and through-hole capacitors, but I am a little concerned that the AF amplifier might have caused the failure although I have carried out a few static tests including feeding the right voltage into the amplifier chip from my desktop power supply and it didn't draw too much current so its at least it's not a short circuit !!

 I'll update here when the transistor arrives and see how the repair progresses 😊

Sunday 24 March 2019

New Feed arrives from Germany for Es'hail-2 / AMSAT Phase-4A

Trying to cover all the bases I ordered a dual feed from AMSAT-DL sometime ago (several weeks) and forgot about it.

It arrived on Thursday :)

Its beautifully made I must say - I've also ordered a 1M offset dish as well which I will pickup so slowly putting together some of the more permanent elements of the station. 

This feed does not use an LNB and so I will need to work in a more 'traditional' method, fitted with SMA's for both 10Ghz and 2.4Ghz  - I've noticed its out of stock at the moment - but with this quality I'm not surprised !

The Spec is :-

2.4 GHz left-handed circular polarization 

> 10 GHz linear polarization 
> antenna impedance 50 Ohm 
> SMA connection sockets 
> for rotationally symmetric and offset mirrors 
> aluminum material 
> all screw connections in stainless steel 
> shank diameter Ø 23 mm

I also ordered an adapter which converted the 23mm shank to the more common 40mm LNB size as well to make life simple when I come to use it !!

Saturday 23 March 2019

Collins 516F-2 Upgrade

As part of my ongoing Collins work I recently decided that I would install a farm projects power supply board for the 516F-2, these are designed by W0IY and provide a solid state conversion including a relay which takes the load of the switch on the equipment, this is a known weak point on Collins equipment that uses a rather lightweight switch in the equipment to switch the 516F-2 on and off.

Whilst this work is straightforward it is worth noting, depending on the age of the 516F-2 colours of cable will vary and of course others may have been their first and changed colours or even components !

It is vital that you can read a circuit diagram and sometimes interpolate instructions vs what you are seeing in the power supply you have, also of course - voltages in this area are lethal so please make sure you know what your doing and if not get someone else who does. This post is to let you know what I did / information and if you do choose to do this yourself you do this entirely at your own risk - be warned ! 

I ordered two kits - which arrived very quickly from Barry Buelow, W0IY in the states, these are of excellent quality and come recommended from me 

Next (after reading and understanding the changes) was to check the actual power supply VS the instructions. now I'm very familiar with the 516F-2, but still took this vital step to check and ensure that the colours in the instructions matched my power supply, most did but some didn't (mine is a later version). I marked up the instructions as it applied to my supply so I didn't get confused.
The Kit can be used in many ways, keep the rectifier tubes, keep parallel resonant circuit of C1 or change it one on the PCB, use the relay on the PCB to switch on the power supply to reduce the stress on the radio's on off switch. I made the decision to do all of the mods on this particular power supply but you may decide otherwise. So to get started - build the PCB - not difficult, watch the orientation of all polarity sensitive components such as electrolytics, diodes etc. - I also like to align all resistors colour banding - it looks nice, but also helps should you need to read a value in a hurry!

After an hour so of pleasant assembly work, and checking - I had finished the replacement PCB.

So now onto the main event !

Follow the removal of components for the options you have decided to use, for me it was everything - although take care to label the transformer (T1) leads carefully and also to not remove anything you don't need to (obvious I know but when you get going with the side cutters and soldering iron it might get tempting !!).



The really observant will notice that C1 is not removed in the final picture above, simply I forgot to take a picture after I removed it ! 

You then need to use a drill a drill out the capacitor mounting hardware - take care not  to drill through the chassis just remove the rivet and the brackets. Some of these holes are used to mount the hex standoffs

You then need to slightly loosen the main transformer to enable you to mount the hex stand off's which the PCB mounts to - this can be a little tricky as a number of them are under the transformer (hence slightly loosening the transformer to get access) - using a small piece of card to hold the flat head works really well so it can fit into the holes left by removing the capacitor mounting hardware (picture left)

What you then need to do is follow the wiring instructions provide with the kit, I took the trouble to take the list and carefully (in word) annotate the table with my own power supplies colours of cable together with a box to allow me to tick off when I had completed that element of wiring. Take your time and check / recheck connections, colours and new wires that are needed. It is pretty straightforward but I suggest you have both the old circuit and the new one in front of you as sometimes it can be a little confusing, as it says in the instructions if you are not competent in reading circuit diagrams and applying it in the real world then don't attempt this !

When completed it will look something like this below

what remains is to make the connections to the mains power supply and ensure that some of the old windings are used to buck (reduce) the mains voltage to help with the higher mains supply voltages that are present in most homes now.

The way to do this is detailed in the new instructions - it essentially means applying a small dc voltage (9v) from a battery to the primary and monitoring the output of secondaries to work out which phase needs connecting in order to reduce the voltage.

once completed you use the supplied WAGO connectors (very good!) to wire the relevant wires in series with the mains input to reduce the voltage - when complete you should have something like below you can see the great egg shaped WAKO connector in the middle a left of the picture - also note that there are no connections to the old valve bases at all (and of course no C1!!)

This (unsurprisingly!) looked very much like the kits instruction pictures and so I was ready to test it out.

Now I have a modern electronic variac which allows me to raise the mains voltage and monitor the current very carefully - I use it a lot with old equipment I'm repairing as you do not wish to apply mains voltage - it can result in major power supply / capacitor failure - and that makes a bang, smells bad and results in more work usually !! - so I decided to use that to check that I'd not made any major errors or mistakes - i could also use it to check that I had got the phasing right on my WAKO connectors to reduce not increase the mains voltage !

This is where I diverged (Slightly) from the instructions - I decide not to use a radio to test it with (primarily this is to trigger the relay which turns the power supply on remotely). All this does is to switch the incoming mains voltage which is supplied by the power supply through the front panel switch on the radio (which causes problems as it carries all the inrush current generated by the transformer and why this modification changes that and uses it to switch a mains relay on the board instead - much lighter current - much longer switch life!). 

This connection is in pin 5 and pin 7 of the P1 plug - and so if these are shorted together instead of having to use a radio - you can test the power supply on its own. What I actually did was to apply mains directly to the board and not use the relay at all initially as at low voltage (< approx 145v) the 240v mains relay buzzes and will not pull in - which means you can't test it at lower voltages - which I wanted to do check that my wiring or PCB wasn't wrong.

So I started to apply mains voltage slowly going to 50 v an monitoring the HV output and noticing that it was climbing rapidly (~190V) then on through 125V (~450V) and finally at full mains where the HT was at just over 970v - obviously with no load.

Once other voltages were test (-75 line etc) and the heater voltages which were all present I used the shorting link to check that the relay part worked properly.

 The link (left) was made using some small connector pins that fitted the P1 plug nicely - together with some shrink sleeving) and is used as per the picture right - take care though there is mains voltages on this link and I took very great pains to ensure that it was properly insulated, and I know what I'm doing - if you do anything like this you do so at your own risk entirely - this is just my way of doing it - I' experienced and very careful!!

I'm very glad to report that all worked as expected and the power supply is now doing sterling service in my collins collection :) - now I have another one to do when I get around to it.

Summary - a truly excellent kit and strongly recommend it -  not for the beginner as the wiring and testing techniques require good knowledge and ability but for the 'average' collins enthusiast that does his / her own repairs testing etc. it is pretty straightforward to be honest - took me about 8 hours over two days to complete from start to finish with plenty of breaks and time to take photo's (!)  - I would be able to do it more quickly now though.

73s and good DX - Chris M1ABK