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 11th Apr 2017, 11:32 pm   #41
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,801
Default Re: Marconi 2019 Signal Generator.

Great result!

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 12th Apr 2017, 3:47 pm   #42
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.

I've checked the modulation using my Racal Dana Modulation Meter and it's fine.

This leaves just one fault. On switch on a 'P' error is displayed indicating an EPROM fault, presumably a check sum error. Despite this the SG seems to work OK. I've reseated the EPROMs, but it made no difference.

Does anyone know if a binary/hex file is available of the EPROM contents? The firmware is issue 12. I don't have an EPROM Programmer anymore, but if a listing is available I'll try to find someone who can check my EPROMs and reprogram them as necessary.

Thanks.
__________________
Graham. Forum Moderator

Reach for your meter before you reach for your soldering iron.
Station X is offline  
Old 2nd Jul 2017, 3:20 pm   #43
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.

This SG is still displaying a 'P' error on switch on, but it appears to work OK.

I obtained .BIN files for the four EPROMs and correctly verified my own EPROMs against them, so it isn't an EPROM fault.

I need to think again, but I'd have thought a dodgy address or data line connection to an EPROM connection would be fatal?
__________________
Graham. Forum Moderator

Reach for your meter before you reach for your soldering iron.
Station X is offline  
Old 2nd Jul 2017, 6:01 pm   #44
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.

Possible progress, or possibly a red herring.

Pin 18 (Output Enable) is stuck high on one of the EPROMs. This is fed from pin 13 (Output 2) of an 74LS138 which acts as an address decoder for two address lines. There is continuity between the two pins.

On the other three EPROMs pin 18 is pulsing as expected.

I've established that the inputs from the two address lines to the address decoder are pulsing, but I don't know whether all 4 combinations of high/low are being presented. It could simply be that whatever firmware was running at the time I did my testing was only accessing three of the four EPROMS.

I'll try 'scoping the two address lines, but it may not be possible to tell anything from the traces.
__________________
Graham. Forum Moderator

Reach for your meter before you reach for your soldering iron.
Station X is offline  
Old 2nd Jul 2017, 6:08 pm   #45
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,190
Default Re: Marconi 2019 Signal Generator.

Well, you know both address lines can be high (output 3 of the '138 is being asserted) and can both be low (output 0 of the '138 is being asserted). So unless you can look at them both together you can't tell much. Even then, you would want to check that they are in the right state at the same time that the '138 is being enabled (and it has 3 enable inputs, 2 active-low and one active-high).

Can you reset the processor (e.g. by shorting the reset capacitor)? If it is checking the EPROMs at power-on (so as to be able to produce the 'P' error) then it must access all of them.

Does 'P' only mean an EPROM fault, or can it be something else wrong with the processor circuitry? What is the processor, and what (if anything) are the external RAMs used with it?
TonyDuell is offline  
Old 2nd Jul 2017, 6:14 pm   #46
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.

Error 'P' definitely means a ROM fault. There are other error messages for RAM and EAROM failure. In fact I cured an EAROM failure fault.

The outputs used to enable the four EPROMs are actually O0, O2, 04 and O6.

I'll try your suggestion of monitoring pin 18 then doing a uP reset.
__________________
Graham. Forum Moderator

Reach for your meter before you reach for your soldering iron.
Station X is offline  
Old 2nd Jul 2017, 6:44 pm   #47
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,190
Default Re: Marconi 2019 Signal Generator.

You also know that the 2 address lines to the '138 inputs B and C are not shorted together, given that output 4 is asserted sometimes.

You don't actually know what the self-test tests. The reason I mention this is that I have had such tests give confusing results (in other devices) in the past, claiming, say, EPROM failure when the problem was actually RAM-related (the RAM locations used to accumulate the ROM checksum were defective). So do watch out. And if it happens to use 2114 RAMs, suspect them on-sight. They are the IC equivalent of a Hunts capacitor
TonyDuell is offline  
Old 2nd Jul 2017, 9:50 pm   #48
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.

RAM is on a pair of 8155 RAM + IO +TIMER chips.
__________________
Graham. Forum Moderator

Reach for your meter before you reach for your soldering iron.
Station X is offline  
Old 2nd Jul 2017, 11:10 pm   #49
karesz*
Heptode
 
Join Date: Jun 2010
Location: Zurich, Switzerland
Posts: 538
Default Re: Marconi 2019 Signal Generator.

Hi Graham,

I dont know if you can use it?

Greetings, Karl
karesz* is offline  
Old 3rd Jul 2017, 5:29 am   #50
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,190
Default Re: Marconi 2019 Signal Generator.

The 8155 is normally reliable, although I have had them fail. This sounds like it uses an 8085 CPU, I've had those fail in strange ways (some instructions don't do what you expect!)
TonyDuell is offline  
Old 3rd Jul 2017, 9:57 am   #51
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.

Quote:
Originally Posted by karesz* View Post
Hi Graham,

I dont know if you can use it?

Greetings, Karl
Thanks for the link Karl.

Please see post #43.
__________________
Graham. Forum Moderator

Reach for your meter before you reach for your soldering iron.
Station X is offline  
Old 3rd Jul 2017, 10:05 am   #52
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.

Quote:
Originally Posted by TonyDuell View Post
The 8155 is normally reliable, although I have had them fail. This sounds like it uses an 8085 CPU, I've had those fail in strange ways (some instructions don't do what you expect!)
The 8085 and one of the 8155's had to be replaced to get the SG working at all. Fortunately, from my point of view, these three chips are socketed and I was able to put the original chips back in to prove they really were faulty before binning them.

I'll try swopping over the 8155's to see whether the symptoms change.

There were 8 or 9 faults in all on this SG spread over the digital and analogue sections. The EAROM was also faulty and I remember replacing another chip on the uP board, it may have been a latch.
__________________
Graham. Forum Moderator

Reach for your meter before you reach for your soldering iron.
Station X is offline  
Old 3rd Jul 2017, 9:17 pm   #53
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.

I've spent a couple of hours this evening working on this SG. I'm almost there but not quite.

The initialisation sequence of this SG is quite long, perhaps 3 or 4 seconds before the 'P' denoting an EPROM problem appears. I connected my logic probe to what appeared to be a stuck ENABLE pin of an EPROM and switched on the SG. I could see three distinct pulse bursts and then the pin went permanently high. So nothing wrong with the 74LS138 then. The apparently stuck pin was a red herring.

I then swopped over the 8155's and noticed a strange effect. Depending on which way round the 8155's were the SG did a long initialisation sequence during which the SG's relays were exercised, or a short initialisation sequence during which the SG's relays were not exercised.

I had quite a few 8155's to try, some of AMD manufacture and some of NEC manufacture. It took me some time to work out that if an AMD 8155 was inserted in the first position, the long initialisation sequence was used. If an NEC 8155 was inserted in the first position the short initialisation sequence was used. The type of 8155 in the second position didn't matter.

I cannot explain what's going on here, but if someone with one of these machines can tell me the correct initialisation sequence I'll fit the correct 8155.

Needless to say, regardless of which 8155's were used the 'P' error was still displayed.

I was clutching at straws now, but I noticed that after initialisation the frequency displayed was 10.7MHz and not the default 1040MHz for this SG. This was because I had 10.7Mhz stored in store location 10 and I'd selected an option to use store location 10 following initialisation. I deselected this option and on initialisation the SG displayed 1040MHz with no 'P' error. Well naturally I reselected the option. On initialisation 10.7MHz was displayed with no 'P' error. I repeated this sequence several times, but could not get the 'P' to reappear.

Another unexplained event. There's only one place options can be stored and that's in the EAROM. whenever this is written to the checksum is recalculated and stored in the EAROM itself. This is then checked on initialisation, but if it's wrong should not result in a 'P' error.

I noticed that the manual says that a 'P' error should be fatal, but in my case it wasn't with the SG apparently behaving normally.

Thanks everyone for your help with this. All I need now is for someone to tell me which is the correct initialisation sequence.
__________________
Graham. Forum Moderator

Reach for your meter before you reach for your soldering iron.
Station X is offline  
Old 4th Jul 2017, 7:28 am   #54
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.

I think I have a problem with memory access times. The 8085 has a 6.144MHz crystal, so it's being clocked at 3.122MHz. I need to check the suffixes of the 8085 and 8155's.
__________________
Graham. Forum Moderator

Reach for your meter before you reach for your soldering iron.
Station X is offline  
Old 4th Jul 2017, 11:29 am   #55
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.

The parts list specifies an 8085A uP. The circuit shows a nominal 6MHz crystal consistent with the 3Mhz speed of an 8085A. As far as I can see there is no WAIT circuit.

An Intel 8085A was originally fitted, but as this was faulty it was replaced with an 8085AH, the HMOS (High performance) version.

As I understand it both these uP's need 400nS memory?

The NEC chips are 8155HC-2 with an access time of 330nS

The AMD chips are 8155PC/p8155. I can't find a datasheet for these, but I assume these have an access time of 400nS and they're what I should fit?
__________________
Graham. Forum Moderator

Reach for your meter before you reach for your soldering iron.
Station X is offline  
Old 10th Jul 2017, 10:54 am   #56
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.

In the absence of further information I fitted the AMD 8155's. This gives a long initialization period with the relays being exercised. No error messages are being generated.

The SG is now all back in one piece and I've used it to check a set's alignment.

In view of the reported unreliability of these old SG's, which I can well believe, I'll now wait for the next fault to manifest itself.
__________________
Graham. Forum Moderator

Reach for your meter before you reach for your soldering iron.
Station X is offline  
Old 11th Jul 2017, 10:43 am   #57
chriswood1900
Octode
 
Join Date: Jan 2007
Location: Dorridge, West Midlands, UK.
Posts: 1,475
Default Re: Marconi 2019 Signal Generator.

Graham
Whilst some people say they are unreliable, I'm not sure my 2019A was reliable when I was using it regularly but developed a fault when not used. They are also still fixable with available parts and plenty of info which is more than can be said for some of the more recent ones or the ones from manufacturers which are less forthcoming such as R&S & Adret.
Keep those fingers crossed.
Chris
chriswood1900 is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 3:17 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.