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 26th Feb 2018, 7:15 am   #1
Radio1950
Hexode
 
Join Date: Apr 2017
Location: Buderim, Queensland, Australia.
Posts: 428
Default Renovation of Eddystone EC10 S/N 1103

Renovation of Eddystone EC10 S/N 1103 January 2018.

I bought this EC10 Receiver in an antiques shop quite by accident, and it was in extremely good cosmetic condition, so I carried out a mini-restoration to bring it back to spec, keeping all original parts where possible, but without being too pedantic.
On test after purchase, it performed reasonably well, but band 2, 8.5 – 18 MHz had a suspiciously low noise floor, and the AF volume could not be turned down to zero on any band.
Sensitivity on all bands was < 4uV, and IF sensitivity was 1.25uV, all measured as per manual.
It was in very good condition inside, and a previous owner had replaced most electros, and added a BNC socket for antenna connection to “A1”. He also added a 2.5mm DC socket to the rear plate of the battery box to use a 9 V DC plugpack.

I won’t go into too much detail with restoration, as this set has been covered many times elsewhere, but I did have one surprise; see later.

I added a AC/DC PSU in a little plastic box to fit inside the metal battery box, which use a diode bridge and 7909 regulator. This allows use of standard AC or DC plugpacks of any polarity and 9 to 12 V AC, and 12 V DC. It has a zener fed LED on the end of the plastic case, so I can see in the side of the EC10 and see if it is powered up OK.

The four pin McMurdo battery box to main chassis connector has a broken wing, but is functional, and I am sourcing a replacement, either an original, or a cut down version of a McMurdo 10 pin version. My receiver is the “battery only” type, not the mains version.


Disabled – antenna input protection diodes (I live close to a BC station).

Modified – I removed the detector diode D2 and detection capacitor C65 to be outside the IFT can, to allow assessment with a product detector. This is on hold, as the EC10 oscillator stability is possibly beyond improvement to allow sufficiently stable SSB use. (I’ll use my 1830).

Added a small piece of “1cm” open wire square mesh between speaker and chassis side to avoid accidental “fingers in the cone” whilst servicing.

Replaced

1- all RF and IF OC171 transistors with AF178 types. The old transistors had varying degrees of leakage and low gain, or both, but no whiskers. No issues with the new AF178s, which were from an eBay seller in Munich Germany – all tested OK on delivery. Lowest leakage devices used in RF.
2 – all resistors in RF, Mixer, Oscillator, BFO and IF stages, some in AF. Most had gone 10 to 40 % high, but none was OC. I used modern 1 watt carbon types which looked appropriate, and which I thought might improve stability.
3 - zener diode D3 with 1N4736 at 6V8, hoping to improve oscillator stability, but no improvement.
4 –DC supply line electrolytic capacitors, those not replaced by previous owner.
5 – AGC line components.
6 – Capacitor C30 in Mixer was leaky; replaced with ceramic.
7- Several other capacitors in the AF.
8 – Volume pot
9 – LES panel lamps (from an eBay seller in UK).
10 – Mixer Band 2 capacitor C26 50 pfd tubular ceramic, found to be near open circuit, big surprise, as these normally last forever. I didn’t believe it at first, but it tested low C on two bridges. This cured the low noise on band 2, and allowed correct tuning tracking. (The capacitor marked “50 pfd”, measured 6 pfd on a zeroed C bridge, and with a squeeze on one end with small piers, it jumped to approx 50 pfd. The end connection must be suspect. A bit rare).

Alignment

All OK as per manual steps and methodology.


Test Performance


RF sensitivity all in Eddystone spec, and typically less than 1 uV for 10 dB S/N.
BC Band less than 3 uV, which is still very good.
IF sensitivity 0.9 uV for 10 dB S/N injected as per manual.
AGC OK and in spec.
Image Response not the best on high bands, but in spec.
AF N+D in spec.

Observations

1 – lovely looking set, easy to work on, well made, no original dry joints. I was smitten.
2 - very good AM performance, for its day.
3 – oscillator stability OK for AM, but not able to be improved in its present form?

Suggestions

Next job is probably to add a Crystal Calibrator, switched ON tgether with the Panel Lamps mom. push switch.
I tried two types of product detector mods, a two transistor type and a DBBM four diode type, but was let down by oscillator stability and other factors. I have some MC1496’s on order, and may try an active product detector later.

No need for photos, mine is as ex factory from the front, just extra BNC and DC sockets on the back.
Inside is 99% as ex factory.

Now, where’s that 1830 service manual of mine?

mike

Last edited by Radio1950; 26th Feb 2018 at 7:23 am.
Radio1950 is offline  
Old 26th Feb 2018, 9:15 am   #2
HamishBoxer
Dekatron
 
HamishBoxer's Avatar
 
Join Date: Aug 2007
Location: W.Butterwick, near Doncaster UK.
Posts: 8,923
Default Re: Renovation of Eddystone EC10 S/N 1103

Great write up Mike.
__________________
G8JET BVWS Archivist and Member V.M.A.R.S
HamishBoxer is offline  
Old 26th Feb 2018, 12:52 pm   #3
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: Renovation of Eddystone EC10 S/N 1103

Mike,

Very good move using the AF178. It is particularly important in the EC-10 due to the frequency coverage that a suitable replacement like these is used. I have restored five EC-10's, starting on the first one in the 1980's and used AF178 for all.

There is an issue with the EC-10 relating to its alignment:

The coil array:

1) They used single peaked coils in the antenna input , RF stage output (prior to the mixer). There are TWO positions of the tuning slug that these coils go into resonance at the correct frequency, in each case the slug, geometrically speaking, pokes out the opposite side of the main coil winding. However, in one case it increases the coupling to the other winding. So there are in fact "two ways" to tune up this radio. In the slug positions where the coupling to the other winding is higher, the radio's overall gain is improved.

2) The radio, typical of most transistor radios, contains what I have come to call the "classic AGC mistake". (common to not all but most transistor radio designs).

With low signal levels and low signal output levels from the detector (AGC on) the AGC filter capacitor has a small voltage applied from the divider network of R20 and the RV1. You will notice if you look at the polarity(wiring) of C63, the AGC filter capacitor, it is correct under these conditions.

However, with a good signal from the detector, the voltage across C63 (10uF electro) reverses. (Reverse applied voltage to an electrolytic is never a wonderful idea).

So in all my transistor radios that suffer from this dilemma (many) I always fit a bipolar electrolytic for the AGC filter capacitor. It is odd that in over 60 years of transistor radio technology this seems to have gone largely unnoticed.

Often in vintage transistor radios, from the mid 1950's, the AGC capacitor is open circuit, probably because it spends most of its life under reverse polarity.
Argus25 is offline  
Old 27th Feb 2018, 6:37 am   #4
Radio1950
Hexode
 
Join Date: Apr 2017
Location: Buderim, Queensland, Australia.
Posts: 428
Default Re: Renovation of Eddystone EC10 S/N 1103

Thanks Argus. Good comments.

I didn’t realise that aspect of the AGC with this receiver.

I just checked your observations and found my EC10 reverses AGC polarity if I raise the input above approx -20 dBm (approx 20 mV), but is in constant reverse polarity if AGC is OFF. I then replaced my C63 to be bipolar.

I suppose we all wish that Eddystone had put FETs in the front end, and that would really make it perform and with good RF selectivity. A decent LO, SSB detector and Audio derived AGC would also have been nice. And ...


I do remember the dual peaking aspect of some coils in the front end.
But I seem to also remember that I found that if I peaked some coils with the ferrite core more fully in, and so maximised the coupling to maybe get better sensitivity at the low end, I then found that S/N was reduced at the high end.

So I think that, after I got a feel for the set alignment, I adjusted coils with the core fully out at first, then peaked on first peak inwards, peaked the capacitor at the high end, and checked S/N across a band. If it was satisfactory, I left it at that position, as in general, sensitivity and S/N were mostly very good anyway (less than 1.0uV for 10 dB across the bands).

I was always taught, (and mostly found) that this was how most designers and manufacturers built their sets, and how realignment was expected to be carried out, and I just do it naturally. “Cores out, first peak in, then check”. Not universal, but a good start.

And, my thinking was that I may have left it with “in spec” or slightly better image response, as the coupling would have been quite satisfactory for S/N but the circuits would not be over-coupled, and thereby maximising circuit Q and selectivity. I never did any comprehensive quantitative image measurements, just coms tech type listening tests, as image response on the high bands is inherently relatively poor.
Anyway, I was mindful that the EC10 was not designed to be a top of range coms receiver.

Having said all that, I just opened up my receiver now, and found that a few coils have cores more fully in. And so I must have decided that those positions were a best compromise, on the day.

I must admit also, that one or more of my coil peaking positions may now have been perhaps a little affected and determined by age degraded or faulty fixed capacitors across coils, as I did find one tubular ceramic cap C26 with low capacitance , viz nominally 50 pfd, and measured 6 pfd. My experience is that this is rare.

I decided that due to the practical difficulties of removing all such capacitors for test, and the possibility of associated damage, I would just leave it all alone, and investigate and fix on suspected fault.

Most of my listening would be at the high end of each band, except BC of course, (I will leave that for you to guess why, and in both cases), and my thinking is that a problem trimmer cap will affect the high end tracking more, and I will probably notice the loss of signal, then investigate.

I could go on ...

I only have one EC10, but I do have 65 English Record metal wood planes.
I am not a collector.

mike
Radio1950 is offline  
Old 27th Feb 2018, 7:09 am   #5
Radio1950
Hexode
 
Join Date: Apr 2017
Location: Buderim, Queensland, Australia.
Posts: 428
Default Re: Renovation of Eddystone EC10 S/N 1103

Some Selected Test Figures for my EC10, taken today.

AGC Performance

Setup as per manual.
Freq 2MHz, AGC ON, RF Gain at max, RF input start point 6uV.

AF lvl rises 11 dB for and RF lvl 80dB above 6uV.
This is slightly better than spec.


For interest

AF lvl change for
1uV to 10uV is 7dB
2uV to 20mV is 6dB

All of the above is quite good for this type of receiver.

Image Response.

Setup AGC OFF, RF Gain at max, RF LVL 2uV at nominal frequency

02 MHz 55 dB
10 MHz 41 dB
15 MHz 29 dB
20 MHz 25 dB
28 MHz 15 dB

This is a little better than spec.
Comparison with other more sophisticated receivers may be unfair.
I make no other comment!


mike
Radio1950 is offline  
Old 27th Feb 2018, 8:39 am   #6
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: Renovation of Eddystone EC10 S/N 1103

It is interesting, about comparing the EC-10 to more sophisticated receivers. I have more sophisticated comms receivers with supposed better specs than the EC-10, many with digital based controllers. But it is the simplicity and elegance of the whole EC-10 that makes it so appealing. Along with the EC-10's incredibly low battery current drain compared to modern receivers which can be 20 to 50 times or more higher due to all the additional electronics to support the tuning and display systems !

Plus the EC-10 is plenty sensitive for practical applications and is very appealing looking, it would look right at home on an episode of Thunderbirds or Joe 90.

If you would like it I have a VFO buffer circuit to export the VFO signal to a BNC connector on the rear panel, then you can feed it to a frequency counter with the IF offset applied. I would never commit the crime of cutting a hole in an EC-10's rear chassis panel myself, but one of my units had a hole cut by someone else, so I used that.

A great LED frequency counter module (with offset option) is the one I used in the H-field transanalyser I posted on the homebrew thread, if you need a module, they are better than most and less than $10. I've used LCD ones but by comparison to LED they are not as good and harder to read.

I built a whole range of VFO buffer versions for the EC-10 with the aim of zero VFO loading, one is an optical system where you can short its output with no detectable effect on the oscillator frequency.
Argus25 is offline  
Old 27th Feb 2018, 11:09 pm   #7
Radio1950
Hexode
 
Join Date: Apr 2017
Location: Buderim, Queensland, Australia.
Posts: 428
Default Re: Renovation of Eddystone EC10 S/N 1103

Hi Argus
I will keep all that in mind, thanks.
As you say, these sets have charm, and mine pulls in WWVH 15 MHz just as well as my Icom 7600.
Radio1950 is offline  
Old 28th Feb 2018, 9:23 am   #8
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: Renovation of Eddystone EC10 S/N 1103

Quote:
Originally Posted by Radio1950 View Post
mine pulls in WWVH 15 MHz just as well as my Icom 7600.
Which is amazing when you consider the small number of active devices in the radio frequency system in the EC-10(compared to the 7600) and the amazingly low energy consumption of the EC-10. It is almost as though some of the newer receivers went backwards, that is excluding fancy digital displays and memory etc.

Probably with the digital frequency counter added to the EC-10, a good one restored unit like yours with AF178's and in proper alignment as you have it, would be a difficult proposition to beat from the simplicity & performance aspect.

I have remembered another issue in some of my EC-10's...I see you replaced some resistors already. One I would definitely replace if not done, R40 the 4.7k resistor in the collector of the audio pre driver stage. This carbon composition resistor puts more noise into the audio than the transistor.
Argus25 is offline  
Old 28th Feb 2018, 11:36 am   #9
turretslug
Dekatron
 
turretslug's Avatar
 
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
Default Re: Renovation of Eddystone EC10 S/N 1103

This has been an interesting thread to follow- I'll own up and confess that I'd tended to regard the likes of the EC-10 as cutesy toys without actually trying one for real. There's a 750 and a 670A here, chosen for their utility rather than being specifically Eddystone products, but I'd felt that early transistor comms. receivers represented a stepping-stone as opposed to an achievement in their own right. I certainly agree with the comments about the excellent power efficiency of the products of this era- in an age when we're expected to be eco-aware, the simple battery-capable transistor receiver that was pioneered in the late '50s, refined in the '60s and polished in the early '70s must represent a benchmark.

It's also encouraging to hear how well it performs against high-faluting sets and formal measurement, the designers were evidently doing a good job with the minimal active device quota.

Agreed re. counter modules- I really don't like LCDs and the LED/VFD modules are something of a steal, granted that consumption and screening are more significant but hardly show-stopping.
turretslug is offline  
Old 1st Mar 2018, 1:05 am   #10
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: Renovation of Eddystone EC10 S/N 1103

Quote:
Originally Posted by turretslug View Post
Agreed re. counter modules- I really don't like LCDs and the LED/VFD modules are something of a steal, granted that consumption and screening are more significant but hardly show-stopping.
Here is something that surprised me recently. When I was looking for counter modules recently for my Transanalyser project (which I had considered initially would be battery operated), I first bought an LCD counter kit, it was moderately expensive, it had a back lit LCD display where the LCD characters were about 5mm tall. Its current consumption measured in at 45mA.

Then I got samples of the PLJ-6LED counter with a 2.5 ppm TCXO, they have adjustable LED brightness. They are less than $20 usually and all over ebay. With the brightness level set on 3 they are bright and the current consumption was less than the LCD counter at about 35mA on the same supply voltage !

Here is some info on them, they have to be the best value counter I have ever come across, compact easy to mount and with good sized LED digits:

http://www.minikits.com.au/doc/PLJ-6LED-counter.pdf
Argus25 is offline  
Old 1st Mar 2018, 1:58 am   #11
turretslug
Dekatron
 
turretslug's Avatar
 
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
Default Re: Renovation of Eddystone EC10 S/N 1103

Yes- they very much come into the "how do they do it for the money?" category, at the time of purchase 1GBP per digit! The sort of price where you think, got to be worth a punt- and what arrived was something of a revelation, competently designed and assembled on a decent GRP PCB. I also bought the VFD version for around double the price (still a bargain, IMHO, as a fan of VFD for both clarity and as a last outpost of thermionics) but these were much less widely featured and soon seemed to vanish, though I'm sure some of the 'net terriers here would do better than m

An early experimentation shot, pretty much as the padded bag was opened. Trying with an LF receiver showed that the reading does abruptly halt below 100kHz- but then, scale resolution hardly needs digital augmentation down there!
Attached Thumbnails
Click image for larger version

Name:	Counter 5.jpg
Views:	171
Size:	23.1 KB
ID:	158443   Click image for larger version

Name:	VFD Counter.jpg
Views:	165
Size:	48.3 KB
ID:	158444  

Last edited by turretslug; 1st Mar 2018 at 2:05 am.
turretslug is offline  
Old 1st Mar 2018, 6:57 am   #12
Radio1950
Hexode
 
Join Date: Apr 2017
Location: Buderim, Queensland, Australia.
Posts: 428
Default Re: Renovation of Eddystone EC10 S/N 1103

Argus

thanks for suggestion regarding R40 in AF.
It was replaced in my first wave, after reading an earlier Renovation Report by Gerry O'Hara.

Regarding LCD Freq readouts etc, I probably will leave my EC10 alone.
Its the charm aspect of a period piece.

If I add a removable 500KHz Crystal calibrator later on, that should be sufficient, and the RX is easily restoreable to factory if required.
I have "other" receivers available if required.

I will have a look tho' at that Mini Kits thingo for other projects.
Radio1950 is offline  
Old 1st Mar 2018, 8:58 am   #13
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,953
Default Re: Renovation of Eddystone EC10 S/N 1103

An interesting rstoration: I briefly had an EC10 in the late-1970s but just couldn't get on with it: image-reception [15MHz broadcast-stations in amongst the 14MHz amateurs I was trying to listen to] and LO drift/pulling - both slow thermal and much more problematic frequency-change-when-you-adjusted-the-RF-Gain meant SSB/CW operation was not a lot of fun.
Running it off a decent regulated power-supply didn't significantly help the frequency-change-by-the-RF-gain-control problem.

I'm not saying it was a *bad* receiver either in design or omplementation, just that it didn't do anything for me. I currently have an Eddystone 840A [which I suppose you could say was the decade-and-a-bit free-electron-device-based precursor of the EC10] and it has precisely the same issues!
G6Tanuki is offline  
Old 1st Mar 2018, 1:01 pm   #14
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: Renovation of Eddystone EC10 S/N 1103

For my EC-10 I would not consider modifying it to add a counter to its internals, only a counter in a separate box. The VFO buffers I have made are just a small pcb that mount on existing hole in the EC-10.

I have attached a photo of the one I use (home grown). The displays might look familiar, like the Texas TIL311...but they are not, even though they look exactly like they are and would work if you plugged one in to a TIL311's socket. The are a little known TIL311 substitute which is a low power CMOS version. They were made in Canada by Innocor, and have subsequently become unobtainium.

I built a range of wide band optical isolators for different applications, such as video signal earth isolators and VFO buffers for radios like the EC-10. I have attached a photo of one. At first glance it might look like dead bug construction, but it is not. The substrate is a brass plate, with small holes. These are soldered through from the rear to rectangles and squares of double sided pcb material. Once that is done the base is planed flat. It provides a way to connect everything up and have a good ground plane everywhere. The LED & phototransistor are in the machined brass cube. Unfortunately the photos are a bit over exposed.
Attached Thumbnails
Click image for larger version

Name:	PC030045.JPG
Views:	194
Size:	68.8 KB
ID:	158451   Click image for larger version

Name:	wackybuffer1.jpg
Views:	195
Size:	140.3 KB
ID:	158452   Click image for larger version

Name:	wackybuffer2.jpg
Views:	130
Size:	68.1 KB
ID:	158453  
Argus25 is offline  
Old 2nd Mar 2018, 4:20 am   #15
Radio1950
Hexode
 
Join Date: Apr 2017
Location: Buderim, Queensland, Australia.
Posts: 428
Default Re: Renovation of Eddystone EC10 S/N 1103

New Kitchen Radio on top of microwave.
With spare at top.
At least I think it adds a certain something to the kitchen decor.
Rod antenna hidden at back for ABC Radio National 792 KHz.
A bent paper clip works just as well.

Argus,
I know you weren't suggesting the physical modification of the radio to include a digital display, but rather to have an add-on.
I was just being a little too economical with my words to fully explain.
Attached Thumbnails
Click image for larger version

Name:	New Kitchen Radio DSC06470a.jpg
Views:	289
Size:	84.9 KB
ID:	158488  
Radio1950 is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 11:41 am.


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.