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 Test Gear and Workshop Equipment

Notices

Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders.

Closed Thread
 
Thread Tools
Old 7th Apr 2017, 10:47 pm   #21
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Marconi 2019 Signal Generator.

I think the 10R resistor and the 10pF cap in the emitter leg of TR5 will have a very big influence of the frequency response. So I'd be cautious about changing these as well. They look to be chip parts but I'd leave them alone unless you are certain they are duff.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 7th Apr 2017, 10:58 pm   #22
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Marconi 2019 Signal Generator.

Quote:
so the filter has to be switched to cover the range in two bites.... or more bites.
Yes, At first glance I thought it was varactor diode tuned but I can see in my manual here that it uses switched components. So the 4 lines to the filter are digital control lines not tuning lines as I had written earlier. So as long as those switched lines LD0-LD3have the right selector voltages on them and the diodes are all OK the filter should be OK unless there is a poor connection somewhere.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 8th Apr 2017, 2:18 am   #23
karesz*
Heptode
 
Join Date: Jun 2010
Location: Zurich, Switzerland
Posts: 538
Default Re: Marconi 2019 Signal Generator.

Hi Jeremy,
HTML Code:
I really don't know how much inductance (or what the package capacitance) is in those original CC resistors up at 1GHz.
The Beysclag HF-resistors, suggested by me, are practically equivalent in physical size/volumen 0207 and HF-parameters with an 1/8W CC element from AB...
There would developed for such GHz apps and does exist up to 470 Ohm!
In production there arn`t polished "in spiral"-are non-helical lasered, only in a long line, parallel of corpus axe (as CC too), only their noise is very better as the old CCs Important is: Rosa colored marking=HF version! For me there are the good modern replacement types for discussed resistors and as sayed an thickfilm chip can be a second alternative...

Regards, Karl
Attached Files
File Type: pdf Beyschlags (Mini)MELF HF-Resistors.pdf (145.3 KB, 94 views)

Last edited by karesz*; 8th Apr 2017 at 2:27 am.
karesz* is offline  
Old 8th Apr 2017, 7:39 am   #24
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Marconi 2019 Signal Generator.

With carbon comp or flat SMT resistors, the inductance in that feedback path is principally set by the path length. What needs avoiding are spirally cut resistors, even a cut for value trimming. The L-cut on an SMT part isn't worrying. They have to have length to get from C to B, so L70 is a way of accounting for it.

Twin resistors provide decoupling of the inevitable length up around the Ft of the transistor where it could be seen as a resonant stub.

I think there is some margin here and non-inductive SMT resistors will be OK.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 8th Apr 2017, 7:43 am   #25
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Marconi 2019 Signal Generator.

I too first wondered about varactor tuning, but I think those diodes are cheapie PINs sold for band-switching in TV tuners.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 8th Apr 2017, 8:23 am   #26
Dickie
Octode
 
Join Date: Sep 2004
Location: St. Albans, Hertfordshire, UK.
Posts: 1,477
Default Re: Marconi 2019 Signal Generator.

I think Graham (the OP) has a problem here. If he finds a real "hard" fault, such as an o/c track, dry joint, or incorrect filter selection issues, then all's well. If he ends up substituting components that may be out of tolerance then without suitable test gear (e.g. a spectrum analyser) he won't know whether the gain has increased to the point that instability creeps in, maybe at temperature extremes, or else the gain has fallen such that the ALC can't cope at high frequencies and the RF output level accuracy is compromised.
__________________
Regards,

Richard, BVWS member
Dickie is online now  
Old 8th Apr 2017, 11:00 am   #27
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Marconi 2019 Signal Generator.

Indeed. What can be done with a detector probe is limited, and as they can't tell fundamental apart from harmonics, they aren't much help in fixing doublers. Unfortunately Graham's a bit too far away for me to pop round with an analyser.

The fault sounds like total failure in the doubler band rather than just a bit of component degradation, there bay be something properly dead yet to find.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 9th Apr 2017, 11:27 am   #28
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Marconi 2019 Signal Generator.

Don't forget that these generators only operate on AM up to about +7dBm at the sig gen output so you should get a 6dB boost through the doubler circuit if you run CW or FM and you set the sig gen power to the region above the AM limit of about +7dBm. It might make the signals easier to detect or trace.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 9th Apr 2017, 2:24 pm   #29
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Marconi 2019 Signal Generator.

Quote:
The Beysclag HF-resistors, suggested by me, are practically equivalent in physical size/volumen 0207 and HF-parameters with an 1/8W CC element from AB...
Hi Karl those resistors look very sexy from an RF performance point of view
I think the resistors in the sig gen are leaded so I guess you would have to use a link wire or something? I don't know anything about the resistors used in the sig gen other than they are CC and I have no experience of this resistor type. I know they are low inductance and presumably have low capacitance. Are the originals still available to buy anywhere online?

I still think the first doubler section could be tested at a lower frequency. Looking at the system block diagram and the circuits I think it would be possible to set the sig gen to 555MHz and then inject in 20MHz at the input to AC13. Most easily done at the coax connection from the AM modulator as this would remove the 260.5MHz drive signal to prevent confusion. Set the 20MHz level to maybe -10dBm and then look at the waveforms in the TR4 doubler section to see it double the 20MHz to 40MHz. Also TR5 should show some useful voltage gain at the output at 40MHz.

The doubler has a 100R resistor across one half of the diodes which seems odd as this will upset the drive balance but they must have done this for a reason. The waveforms could be viewed on a regular scope with a x10 probe and you wouldn't have to fit the 56R termination resistor that I showed in my first post. Having seen the PCB images in my manual I don't think it would be wise to use a soldering iron anywhere near TR4, TR5 or TR6 as there are a lot of chip components there that could be thermally stressed by repeated applications of a soldering iron.

I think the doubler section could be tested at 20MHz with a scope. If the scope is a fairly modern digital scope it may have an FFT function and so the spectrum of the 20 to 40MHz doubling could also be viewed o the scope. The probe could also be used here to look for instability anywhere in the whole doubler path up to TR6 because any RF detected when the 20MHz is turned off must be due to instability/self oscillation somewhere.


Once this is complete and the sig gen connections are restored to normal, a basic RF probe could be used to see how far (any) RF gets through the doubler, the post doubler amp, the switched filter and TR6 and the diode switch after TR6 when running with the regular 260.5MHz drive with the sig gen configured and running back as normal. A sudden loss in RF could indicate a break in the signal path.

An RF probe isn't as good as a spectrum analyser but it is a lot better than nothing. By putting a simple tuned circuit at the input of the RF probe you could make it into a form of mini 520MHz wavemeter/detector to give it some frequency discrimination but I doubt this would be necessary. Hopefully the fault will be a glaring one.
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 9th Apr 2017 at 2:36 pm.
G0HZU_JMR is offline  
Old 10th Apr 2017, 2:43 am   #30
karesz*
Heptode
 
Join Date: Jun 2010
Location: Zurich, Switzerland
Posts: 538
Default Re: Marconi 2019 Signal Generator.

Hello Jeremy,
I found some description of Beyschlags HF-Resistors and there are orderable form such online firms as CCs (Mouser ie has only0,25--2W) too. Here is a description of CC-system; the whole corpus as carbon cylinder is an resistor. There are good for impulse loads and RF applications, but have relative high tolerances (5/10/20%) and are solderings temperatur sensitive -at ca 2%... Unfortunately are the most buyable types nowadays from china. I think that original ABs (coded as RC05) are only as surplus to find, but Westflorida Components does it offer too...
Regards, Karl
Attached Files
File Type: pdf ABs Carbon Composition Resistors_CID001-4.pdf (1,012.1 KB, 64 views)
karesz* is offline  
Old 10th Apr 2017, 1:49 pm   #31
Station X
Moderator
 
Station X's Avatar
 
Join Date: Jan 2003
Location: Ipswich, Suffolk, IP4, UK.
Posts: 21,192
Default Re: Marconi 2019 Signal Generator.

Wow!

I'm overwhelmed by the responses to my post.

I hope to do some more work on this today.

Rest assured I won't be changing any components unless I'm absolutely sure they're the cause of the fault.
__________________
Graham. Forum Moderator

Reach for your meter before you reach for your soldering iron.
Station X is offline  
Old 10th Apr 2017, 6:05 pm   #32
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Marconi 2019 Signal Generator.

I've got an old 2019 here and also a newer 2019A and yesterday I took the bottom covers off the 2019 to reveal this AC13 module because I was curious to see the various components used.

I've had a few surprises looking in there. I don't think I've been inside this module on this 2019 before and mine looks like it has had some work done on it in the past. It's definitely had some work done around the doubler as there is some flux residue around the diodes and a couple of the SMD blocking caps don't look original and they are a bit wonky.
The output blocking cap by TR6 has definitely been removed and refitted at some point. I've had these two generators over 20 years but I don't think this was done by me! The two reworked caps look to be the same family package but are a different colour to all the other 39nF caps. i.e. the same odd long and skinny shape but almost certainly replacements. Not sure what package these caps are as they aren't 1206 as they are too skinny.

So sadly, my 2019 isn't a perfect example. However the two series feedback resistors in TR5 have aged a lot. The 100R resistor is 114 ohm and the 51R resistor is 78 ohm. So that's a combined change of about 25% in my generator too.

However, the biggest surprise of all was the feedback arrangement in other amplifiers that are on display once the cover is removed. The output module is one example. The resistors are fitted way up in the air using what looks like the full resistor leg length! They typically use a single 200R resistor here and to make things even stranger the feedback resistor legs look like they have been deliberately bent sideways. Maybe this is to prevent interstage coupling and instability as these amplifiers are sometimes right next to each other. What a strange design! The manual shows these resistors on very long legs so I think it is meant to look like this and they must need lots of inductance in the feedback.

I did try feeding 20MHz into AC13 at -10dBm (with the generator set to 550MHz) and I can confirm that the doubler still 'doubles' although the 40MHz doubled signal only gets as far as the switched BPF obviously. But it is possible to see the waveforms in TR4 and TR5 on a regular scope and prove that they are what would be expected from a healthy doubler.

Obviously, this isn't a thorough test of the doubler because you should really test it at 260.5MHz but I can post up the scope waveforms if you are interested? My scope can see the doubler working at both 20MHz and 260.5MHz using a suitable active probe and it can display the spectrum as well. The only issue with driving the doubler at AC13 is that you would need an SMB male connector to mate with the coax cable that feeds to AC13. I've got SMB connectors here but you would have to buy one from ebay if you don't have one. The cheapest BNC to SMB male cable I could see was £4 from Hong Kong but these would take a month to arrive (if ever).
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 10th Apr 2017 at 6:11 pm.
G0HZU_JMR is offline  
Old 10th Apr 2017, 6:37 pm   #33
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Marconi 2019 Signal Generator.

The 10uH chokes used in the doubler look to be hand wound chokes wound on a miniature binocular core on my 2019. Have you checked these for continuity? That could be a weak spot.
Also, if you measure the anode and cathode voltages of all of the diodes in the switched filter this should perform a fairly good continuity check through the switched filter path itself. A poor connection somewhere might affect the control voltage arriving at a diode? As I mentioned in my previous post, two of the 39nF caps in my doubler path look to have been replaced as they are a different colour to the rest. Maybe these caps can fail over time? MLCC caps that have been stressed can develop micro cracks internally and can work fine for a while but fail (years?) later. However, they would probably still pass 260MHz waveforms fairly well. But they might become leaky or S/C at DC.
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 10th Apr 2017 at 6:43 pm.
G0HZU_JMR is offline  
Old 10th Apr 2017, 9:41 pm   #34
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Marconi 2019 Signal Generator.

There should be current-seting resistors with the PIN diodes. If you select each band in turn and run along the resistors to check the volage drop on them, it performs a continuity test for the whole CD bias arrangement through the diodes, chokes and all (so long as you've checked the resistor resistance first!)

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 10th Apr 2017, 9:45 pm   #35
Station X
Moderator
 
Station X's Avatar
 
Join Date: Jan 2003
Location: Ipswich, Suffolk, IP4, UK.
Posts: 21,192
Default Re: Marconi 2019 Signal Generator.

Thanks for the additional updates Jeremy. I read them after I'd done today's testing.

TR6 Voltages:-

C 14.96V
B 2.776V
E 0.1V

These look OK to me.

After carrying out some more FUNCTIONAL tests I fear I must redefine the problem.

I had my frequency counter connected to the RF output with the wick wound up.

Doubler is supposed to double frequencies into the range >520 to 1040 MHz.

It works at 519 and 520 MHz (doubler not used).
It does not work from 521 to 529 MHz.
It works from 530 to 660 MHz.
It does not work from 661 MHz to 2040 MHz.

When I say it does not work I mean that the output frequency is unstable and wanders all over the place.

The interesting thing here is the break at 660 MHz, which is one of the filter switching points.

I have checked the filter switching diodes. when turned on they have between 0.75 and 0.8V across them. When turned off they have 7V or 2.150V across them depending on whether the series resistor is 2k4 or 3k9 ohms.

In some cases this indicates that inductors are OK, as current passes through them.

I can check the L's for continuity, but checking the C's will be more of a problem.
__________________
Graham. Forum Moderator

Reach for your meter before you reach for your soldering iron.
Station X is offline  
Old 10th Apr 2017, 9:47 pm   #36
Station X
Moderator
 
Station X's Avatar
 
Join Date: Jan 2003
Location: Ipswich, Suffolk, IP4, UK.
Posts: 21,192
Default Re: Marconi 2019 Signal Generator.

David posted while I was composing my post.
__________________
Graham. Forum Moderator

Reach for your meter before you reach for your soldering iron.
Station X is offline  
Old 10th Apr 2017, 10:28 pm   #37
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Marconi 2019 Signal Generator.

I think your voltages for TR6 look wrong. Typo? ...or could the meter be getting affected by RF? Or is your circuit different to mine?

A simulator says TR6 :

C 8.3V
B 2.65V
E 1.9V

My 2019 measures as follows with no RF at TR6 to confuse the meter. I did this by setting the generator to 20MHz and then measured TR6:

C 9.6V
B 2.6V
E 1.8V

I'm not sure what to make of your latest symptom report yet. My head is still full of tonight's football game but there could be things wrong with the synthesiser/VCO module or maybe some instability somewhere in the complete signal path. ... or maybe your counter is simply getting confused by multiple signals from the faulty signal path.
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 10th Apr 2017 at 10:34 pm.
G0HZU_JMR is offline  
Old 10th Apr 2017, 10:50 pm   #38
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Marconi 2019 Signal Generator.

If TR6 turns out to be duff, don't rush into desoldering it straight away. The components around TR6 will be fragile and easily stressed or damaged by soldering especially if you try and use the soldering iron as a lever to pry components free of each other.

Sadly, I am not the right person to advise the right way to do it. All I can do is advise not to do it the wrong way and trash the parts around TR6 with the iron. The two big resistors on the base look quite rugged but the little leaded cap and the chip parts and the feedback resistors and the 39nF caps could all be stressed or damaged easily.

Maybe post up a photo so someone can advise how to get that BFR91 out neatly and safely.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 11th Apr 2017, 6:39 pm   #39
Station X
Moderator
 
Station X's Avatar
 
Join Date: Jan 2003
Location: Ipswich, Suffolk, IP4, UK.
Posts: 21,192
Default Re: Marconi 2019 Signal Generator.

Here's an update which I'm sure Jeremy for one will be pleased to read!

TR6 was indeed faulty. Many thanks for pointing out the dodgy voltage readings to me Jeremy. There was no voltage across the collector or emitter resistors.

I had changed surface mount transistors in this SG before to cure other faults, so I had no hesitation in changing this one. Having done so I got much more sensible voltage readings.

So I went for the nuclear option, selecting an output frequency of 1040 MHz. Hardly daring to breathe I connected my frequency counter to the output and saw 1040 MHz displayed. A selection of frequencies in the three filter ranges used by the doubler displayed correctly too and the display remained correct when the output voltage was reduced.

I'm pleased that the fault, which could have been diagnosed with nothing more than a DMM, has been fixed.

There are a couple of what I hope will be simple faults remaining, but I'll post about those later.

Thanks for everyone's advice.
__________________
Graham. Forum Moderator

Reach for your meter before you reach for your soldering iron.
Station X is offline  
Old 11th Apr 2017, 10:20 pm   #40
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Marconi 2019 Signal Generator.

Glad you got it working!
Looking at your test voltages for the failed amplifier (TR6) it looks like something bad happened to the B-E junction.

The transistor that drives the doubler diodes TR4 (BFR96S) runs at a high temperature due to the high Pdiss of about 500mW so I'd expect this to be the least reliable device in this signal path but I've not seen TR4 fail in any 2019 that I've had experience of.

That includes all the ones we used to have at work. They had lots of other faults though. They became too unreliable and were declared BER in the late 1990s and scrapped. The two I have here were bought cheaply as non runners at a rally and like you I had to cure numerous niggly faults over the first few years to keep them running.

Even when you clear all the faults, you may well find you will be inside that 2019 again soon but I managed to get both of mine reliable eventually. I think it helps to run them in a well ventilated area to reduce thermal stress.
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 11th Apr 2017 at 10:30 pm.
G0HZU_JMR is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 10:51 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.