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 8th Jun 2017, 5:56 pm   #1
chriswood1900
Octode
 
Join Date: Jan 2007
Location: Dorridge, West Midlands, UK.
Posts: 1,475
Default Marconi 2018A RF signal generator no output.

Following from my earlier post on the Marconi 2019A http://www.vintage-radio.net/forum/s...d.php?t=136010 I am ready to tackle the 2018. This was bought as spares with a report of no output.
When I first tried it we had a boot up with a scrambled frequency display and as expected no output. Tried some key presses there was some response and a clicking of the attenuator. So lid off and start with checking the voltages on the PSU which were all in order so on to the display/keyboard, the common reason for problems in this area are poor connections from the keyboard to the CPU in the RF box. Off with the lid and on to board AA2. Unplugging all the multi way connectors on that and the display board brought back the display so I could now see what was happening.
As part of my diagnostics I thought I should take a look at the output from the RF box at the back SKAV where the connecting pipe connects via a SMC which needs removing to tilt the box up for repairs. I was surprised to find I had an SMC to bnc adaptor in the test lead box, (I have no idea where it came from) but it allowed me to connect both a frequency counter and a Scope to examine the output which was present at close to the frequency shown on the front panel. However I was disturbed at the shape of the wave form which at the lower end looked like a poor square wave so I thought I had a problem.
Next was to work through the various boards and checking each oscillator on AB3 but it was all working so I thought I would take a look at the RF box output on my 2019A, it turned out the wave shape was exactly the same, a square wave at lower frequencies turning into a sign wave above 260 Mhz. All the time the the main output was a good sign wave at all frequencies so I had been chasing a non existent problem oh well!
So it was RF box back together and lids on ready to check the lower box RF2. In this box I worked through AC2 AC3 AC4 AC the signal from RF Box 1 entered and was cleaned up and processed via the filter board AC3, this lead on to the output amp where the lower frequencies from the BFO and the higher frequencies were present up to AC4 TR8, at this point I discover the TR9 and TR10 the output amp are too hot to touch and the signal ceased.

This is where I need some help with the output amp AC4.

This circuit uses a BFQ34 ceramic dual emitter device something I have no experience of. I have attached the circuit of that area along with the voltages measured with a DVM, the numbers in brackets are the voltages from my working 2019A I have checked out all the passive components R36 to R40 the diodes replaced TR9 and C25 & 26 R39 but had no improvement some I am assuming I have a leaky BFQ34 drawing too much current. Interestingly the values of R40 & 41 are 8ohm not 16 as shown and this is true of both the 2019 and 2018. Obviously with the voltages shown R39, TR9 and TR10 were running very hot.
Can those with more experience than me take a look and check my diagnosis, do I need to bite the bullet and buy a new BFQ34 or is there a way of testing it?

Thanks Chris
Attached Thumbnails
Click image for larger version

Name:	img20170608_Marconi 2018.jpg
Views:	498
Size:	80.7 KB
ID:	144155  
Attached Files
File Type: pdf marconi-2019a-service-manual p219.pdf (547.3 KB, 197 views)
chriswood1900 is offline  
Old 8th Jun 2017, 7:12 pm   #2
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: Marconi 2018A RF signal generator no output.

TR9 sets the bias for TR10 so that the collector current is about 100mA. D7 is temperature compensation for TR9 Vbe.

TR10 is running far too much collector current:
1. TR9 is faulty and is supplying too much base current to TR10 even though it should be switched off according to your measurements. Measure the voltage across R38 which will be non-zero if TR9 is faulty.
2. C22 is leaking. Not sure how easy this would be but the easiest test would be to disconnect it.
3. If its none of these then TR10 is probably faulty.

Based on TR9 being hot when it should be cut-off, I would guess that is most likely the fault.
PJL is offline  
Old 8th Jun 2017, 7:23 pm   #3
Dickie
Octode
 
Join Date: Sep 2004
Location: St. Albans, Hertfordshire, UK.
Posts: 1,477
Default Re: Marconi 2018A RF signal generator no output.

You could try temporarily grounding the base of TR10 to see if that turns it off. If it does that suggests TR10 is probably OK.
__________________
Regards,

Richard, BVWS member
Dickie is offline  
Old 8th Jun 2017, 9:41 pm   #4
chriswood1900
Octode
 
Join Date: Jan 2007
Location: Dorridge, West Midlands, UK.
Posts: 1,475
Default Re: Marconi 2018A RF signal generator no output.

Thanks PJL and Richard I will try both suggestions tomorrow..
chriswood1900 is offline  
Old 9th Jun 2017, 5:59 pm   #5
Biggles
Rest in Peace
 
Join Date: Feb 2004
Location: Hexham, Northumberland, UK.
Posts: 2,234
Default Re: Marconi 2018A RF signal generator no output.

I don't know this machine but if you spot any OM345 chips I would start with them.
Alan.
Biggles is offline  
Old 9th Jun 2017, 6:03 pm   #6
chriswood1900
Octode
 
Join Date: Jan 2007
Location: Dorridge, West Midlands, UK.
Posts: 1,475
Default Re: Marconi 2018A RF signal generator no output.

Fortunately it has not used any of those modules, it was the next range that had those. 2022 etc. We are still on discrete components in the RF section 2018/9 range.
Chris
chriswood1900 is offline  
Old 9th Jun 2017, 7:19 pm   #7
Biggles
Rest in Peace
 
Join Date: Feb 2004
Location: Hexham, Northumberland, UK.
Posts: 2,234
Default Re: Marconi 2018A RF signal generator no output.

That's definitely a good thing.
Biggles is offline  
Old 9th Jun 2017, 9:11 pm   #8
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Marconi 2018A RF signal generator no output.

Just for the record I think that the OM345 is used in all variants of the 2018/9. Eg for AM. The 2022 used lots of them and they are a known weak spot in the 2022.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 10th Jun 2017, 7:11 am   #9
chriswood1900
Octode
 
Join Date: Jan 2007
Location: Dorridge, West Midlands, UK.
Posts: 1,475
Default Re: Marconi 2018A RF signal generator no output.

Jeremy
You are correct it is denoted as IC1 on board AC3 the AM modulator, fortunately it has not gone wrong so I hadn't noticed it.
Chris
chriswood1900 is offline  
Old 10th Jun 2017, 7:51 am   #10
Dickie
Octode
 
Join Date: Sep 2004
Location: St. Albans, Hertfordshire, UK.
Posts: 1,477
Default Re: Marconi 2018A RF signal generator no output.

At least with OM345s it is easy to diagnose if they're faulty, replacements are available at a reasonable price, and it's straightforward to replace them. Not so with exotic wideband RF power transistors!
__________________
Regards,

Richard, BVWS member
Dickie is offline  
Old 10th Jun 2017, 9:10 am   #11
Alistair D
Nonode
 
Alistair D's Avatar
 
Join Date: Jan 2010
Location: Edinburgh, UK.
Posts: 2,008
Default Re: Marconi 2018A RF signal generator no output.

Did you replace TR9 before C26? It is possible that leakage in C26 caused TR9s B-E junction to be reverse biased enough that the replacement TR9 failed at power up. A bit of a long shot but..

Al
__________________
I won't tell you how I discovered that.
Alistair D is offline  
Old 10th Jun 2017, 4:15 pm   #12
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: Marconi 2018A RF signal generator no output.

I had not read your posting to the end and not realised you had already replaced TR9 and C26. Richards suggestion for testing TR10 sounds a good idea. It is difficult to understand how TR9 could be killed as it will stand 5V reverse Vbe. Have you put a scope to it to see if it is unstable?
PJL is offline  
Old 13th Jun 2017, 8:40 am   #13
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: Marconi 2018A RF signal generator no output.

That is actually a very elegant circuit. Its a DC servo system to control T10's collector current. That current generates a voltage across the 10R resistor which is compared with the base-emitter voltage of T9. The temperature dependence of T9's base-emitter voltage (about -2.1mV/deg C) is largely cancelled by the diode in the top leg of the resistor divider for the base of T9.

(T9 also has heavy filter capacitance around it, so it only has a response at DC and there is no negative feedback at the operating frequency. If those capacitors went open, then the T10 stage would lose most of its signal gain as it would acquire AC negative feedback too).

The circuit is such that if T10's average collector current increases, T9's base and collector current decrease, reducing T10's base current, so it (t10's collector current) settles to a temperature stable equilibrium, nice.

However, your voltages show that T9 is cut off and the DC negative feedback loop is not working. In fact the current in R38 is reversed and T10 cannot be acquiring any base current via R38. So, assuming there is definitely no leakage of the coupling capacitor into T10's base, T10 is faulty/leaky sadly.

These medium power dual emitter transistors, or very similar to this type (2N5947) were also used by Tektronix in their SG503 sine wave generator, but their bias circuit wasn't as elegant.

Last edited by Argus25; 13th Jun 2017 at 8:55 am.
Argus25 is offline  
Old 13th Jun 2017, 12:20 pm   #14
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: Marconi 2018A RF signal generator no output.

I couldn't see the readings but have copied and pasted to a program I can zoom and it does show 5V at TR9 collector and 5.1V at TR10 base. This would suggest no current is flowing through R38.

C22 leakage is another contender but based on the 0.73/7.18V readings on TR8, the current in to the collector seems to match the current out of the emitter.

My only concern is the comment 'TR9... are too hot too touch'. I guess TR9 may be hot due to the proximity of other overheating components? The only other possibility is there is an AC component due to instability that we are not accounting for.
PJL is offline  
Old 13th Jun 2017, 3:53 pm   #15
chriswood1900
Octode
 
Join Date: Jan 2007
Location: Dorridge, West Midlands, UK.
Posts: 1,475
Default Re: Marconi 2018A RF signal generator no output.

Feedback on from the comments so far,
TR10 is running far too much collector current: ……….Agreed
1. TR9 is faulty and is supplying too much base current to TR10 even though it should be switched off according to your measurements. Measure the voltage across R38 which will be non-zero if TR9 is faulty. ………………….It was zero
2. C22 is leaking. ……………………..This is a surface mount component, fiddly but removable but it made no difference
3. If its none of these then TR10 is probably faulty. You could try temporarily grounding the base of TR10 to see if that turns it off. If it does that suggests TR10 is probably OK.
a) I grounded the base and TR10 stopped conducting.
b) I did some more measurements and the only conclusion I came to was TR10 was faulty.

I took the board out and decided that whilst I had it out to replace or check all the components between C22 and C30
I got out the de-soldering gun and removed D7-10, R39 (Which was distressed) TR9 C22,C25, C26, C29, C30 , and TR10. I then checked all the resistors and except for R39 they were all OK. TR10 shows a direct short between the 2 emitters so when I checked R40 and 41 they were each 16 Ohm contrary to my comment in post 1.
The 3 Electrolytic caps. C25,C26,C29 all were below par the worst being C25 which had dropped to 2uf and an ESR of < 20.Ohms. C22 and 30 were checked for leakage and showed OK but as they got quite hot in removing them I thought I would replace them anyway.
TR10 and its surrounding components are surface mounted on the top of the board and given there were some fairly large pads and the heatsink/lead of TR10 it was fairly tricky but patience and some solder-wick soon allowed me to lift the components around TR10, I could then lift the TAGs that connected it to the board using the end of a small scalpel whilst heating the tab.
In the manual and on stickers on the internal boxes were warnings that TR10 BFQ34 was mounted on Beryllium which is hazardous to health so I wore a face mask whilst removing TR10 and its mounting. The washer referred seems to be the cup shaped heatsink about 25mm diameter on the back of the board bolted to TR10 that fits into another circular hole on the RF box It looked like aluminium but was even lighter, not sure why Marconi chose Beryllium, perhaps someone knows!
I carefully re-soldered all the replacement parts, TR10 needed its tabs trimmed to fit, they are made from soft copper so it could be done with sharp small snips. As it was expensive almost £20 I took extra care over fitting it and ensuring I took proper care with static precautions. Once TR10 was in I could finish by fitting the remaining surface mounted parts that fit over the tabs.

Success fitted the board back in and it was now working, not overheating and giving around 4v PP varying with the selected output. measured at the SMC socket on RF box 2.
I coupled up the coax to RF 2 and there was very little output at the front panel so it is time to look at the Attenuator and its associated cables. Later this afternoon.
chriswood1900 is offline  
Old 13th Jun 2017, 4:09 pm   #16
chriswood1900
Octode
 
Join Date: Jan 2007
Location: Dorridge, West Midlands, UK.
Posts: 1,475
Default Re: Marconi 2018A RF signal generator no output.

A couple of further thoughts, I meant to include a picture of the area, it looks a bit messy but that was how it looks on the original. For electrolytic capacitors I used Panasonic 105°replacing the Elna 85°parts, given their condition I might have to do the others on this board but I thought I should get it working before risking more self-inflicted problems.
Attached Thumbnails
Click image for larger version

Name:	Marconi 2018A Output.jpg
Views:	620
Size:	106.1 KB
ID:	144427  
chriswood1900 is offline  
Old 13th Jun 2017, 5:08 pm   #17
dave cox
Nonode
 
dave cox's Avatar
 
Join Date: Jan 2009
Location: Bristol, UK.
Posts: 2,059
Default Re: Marconi 2018A RF signal generator no output.

I believe the beryllium is in the form of its oxide, which is a good insulator.
dave cox is online now  
Old 13th Jun 2017, 7:27 pm   #18
chriswood1900
Octode
 
Join Date: Jan 2007
Location: Dorridge, West Midlands, UK.
Posts: 1,475
Default Re: Marconi 2018A RF signal generator no output.

Attenuator repair
I disconnected the connectors on attenuator assembly set the frequency to 10Mhz and measured the 4V PP on the input the losses meant there was not much to see on my scope at the output.
I had already checked that the relays operated as you stepped down through the output range.
I then removed the bottom cover over the resistors (2nd picture) whilst balancing the whole thing on top of RFbox1 the leads are just long enough. I started to trace the signal which disappeared at the 30db attenuator C which did not seem to be switching.
I decided I needed to get further into the unit, it appears the switches may be micro switches buried deep inside the boxes which is very well soldered in place, --not looking good.
So I stripped down the unit as per the picture, the relays operate tiny plastic plungers which operate the switches so I used electrical tape to hold them all down so I could trace a signal from the input to just before the output. It won’t trace to the output because without power the output protection relay won’t work and the output stays open circuit. However you can see a track which you can use, in this layout I was getting continuity and using a signal from my function generator I was getting no losses.
A bit of head scratching later and the essential cup of tea, I thought that maybe one of the plungers was not being depressed far enough (shown in 1st picture) by the relays. In the manual it gives warning about the difficulty of adjusting the attenuator but I thought I can’t make it any worse, I noted the position of the adjusting screws for the plungers and tweaked the RH one on attenuator C ½ a turn clockwise this was just enough to depress the plunger and allow the switch to engage, hey presto I had a working unit.
This was followed be a fairly lengthy reassembly I now had a decent output and I could step through a range of from 2v (13db) down to 2mv where I ran out of measuring capability.

I then proceeded to check FM and AM modulation and a range of frequencies along with tweaking the internal oscillator. All well so far then I connected an external standard of 10Mhz, up pops error ------15--- Oh dear, I tried various thing to no avail until I remembered that the std input is also the std output on these. So I coupled up a frequency counter and was only getting 1mhz out, this is odd as every other unit I have every touched used 10Mhz, another cup of tea and I remembered a lot of Ex MOD stuff seems to use 1Mhz so I tried that as an input and it worked with no error . So I thumbed through the User manual and found that under the protected mode options the std frequency is switchable from 1 to 10Mhz. It is a bit fiddly but after a couple of attempts the key press combination of Second Function/0 and MOD ALC held with AF on/off for 15secs get you into protected mode. You can then select option 14 which allows you to switch it to 10Mhz.. After this I could use the 10Mhz standard.
It is now on a soak test, let’s see what that throws up.
Finally I think I will make up some crowbar circuits to protect the 5v rail as suggested by another form member..
I hope my ramblings will be useful to others trying to repair these units.
Chris
Attached Thumbnails
Click image for larger version

Name:	Marconi 2018A Atten Top.jpg
Views:	265
Size:	48.2 KB
ID:	144434   Click image for larger version

Name:	Marconi 2018A Atten Bottom.jpg
Views:	389
Size:	75.3 KB
ID:	144435   Click image for larger version

Name:	Marconi 2018A Atten Circuit.jpg
Views:	310
Size:	82.4 KB
ID:	144436  
chriswood1900 is offline  
Old 13th Jun 2017, 7:46 pm   #19
Station X
Moderator
 
Station X's Avatar
 
Join Date: Jan 2003
Location: Ipswich, Suffolk, IP4, UK.
Posts: 21,192
Default Re: Marconi 2018A RF signal generator no output.

Well done Chris.

I had to repair a 2019A with multiple faults and I know just how difficult it can be. It's easy to get disheartened.

I too will have to look at installing crowbar circuits.
__________________
Graham. Forum Moderator

Reach for your meter before you reach for your soldering iron.
Station X is offline  
Old 14th Jun 2017, 9:55 pm   #20
Radio1950
Hexode
 
Join Date: Apr 2017
Location: Buderim, Queensland, Australia.
Posts: 428
Default Re: Marconi 2018A RF signal generator no output.

Thanks for the info, and well done.

There is also a lot of other information on the Yahoo Group for Marconi Test Intruments.
https://groups.yahoo.com/neo/groups/...struments/info

.
Radio1950 is online now  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 10:39 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.