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 25th Feb 2018, 12:45 pm   #1
G4WYZ MikeyP
Triode
 
G4WYZ MikeyP's Avatar
 
Join Date: Nov 2012
Location: Chorley, Lancashire, UK.
Posts: 21
Default Racal 9081 phase detector problems.

Phase Detector -again!

The original fault on the Phase Detector was no varactor voltage, Q16 U898E was replaced with a 2N3819 resulted in the correct voltage tuning range of 3.6 – 14.8 volts.

Then the flaky PSU folded again it had folded a number of times, although the 2N3055 always worked when tested, it was replaced with a new one and all the issued ceased.

Unfortunately the -6volt supply failed and took this circuit out. all the DC coupled bi-polars transistors were changed except the dual n-channel JFET.

Now works, but the tuning voltage is to high, the first thoughts to increase the value of R63 which is a voltage dropper, since the current drawn by the two varactors is 1.82µA the resulting resister would be large.

The dual-gate n-channel JFET E412/U412 (Q1) is effectively being used as a fast analog switch, does anyone have any ideas how to alter the tuning voltage?

All suggestion would be very appreciated.

73’s
Mike G4WYZ
NB This has been a looooooong term project.
G4WYZ MikeyP is offline  
Old 25th Feb 2018, 10:48 pm   #2
PETERg0rsq
Heptode
 
PETERg0rsq's Avatar
 
Join Date: Jun 2009
Location: St Helens, Lancashire, UK.
Posts: 641
Default Re: Racal 9081 repair questions

Sorry

not sure which circuits you have repaired here. Do you have board numbers and components?

Not sure I understand what you mean by "tunning voltage too high"!

It is a PLL, so the phase detectors will "adjust" the varactor voltage until it comes into lock. If it cant come into lock then the varactor voltage will be too high or too low!
__________________
SPECIALIST.....Knows everything about nothing
EXPERT..........Knows nothing about everything
PETERg0rsq is offline  
Old 8th Mar 2018, 6:22 pm   #3
G4WYZ MikeyP
Triode
 
G4WYZ MikeyP's Avatar
 
Join Date: Nov 2012
Location: Chorley, Lancashire, UK.
Posts: 21
Default Re: Racal 9081 repair questions

Hello.

The Phase Detector Assembly is 19-0846.
The tuning voltage I have miss-stated it should have been the varactor voltage.

It's in lock now, it's when I rotate the 'Tune' control even at a rapid rate, I don't get the required varactor range. The ROM which controls the upper and lower frequency stops, for some reason mine is 496 to 560MHz.

Had a further look this afternoon with a logic analyser on the Blanking circuits, as when I applied an external supply to the varactor, monitoring TP9 in the time domain 828us (256MHz) to 480us (520MHz)
G4WYZ MikeyP is offline  
Old 9th Mar 2018, 9:51 pm   #4
PETERg0rsq
Heptode
 
PETERg0rsq's Avatar
 
Join Date: Jun 2009
Location: St Helens, Lancashire, UK.
Posts: 641
Default Re: Racal 9081 phase detector problems.

The prom "reads" the divider lines from the counter to the divide by N and sets the limits by pulsing the "tune up" or "tune down" lines, which simulate the operation of the spin-wheel, until the out of range frequency comes back into range.

Each range and channel spacing setting will have a different "lookup table" in the PROM

You can eliminate the PROM by disconnecting the two lines from it to the digital board.

I have reverse engineered the PROM code, but first it is best to "read" the address lines going to the PROM and check the counter circuits are working correctly. Sometimes a divider chip fails and doesn't pass on the carry bit to the next divider, so you might need to decode all the dividers, which is easier than it sounds if you have a logic analyser.

Start with the basics, make sure you have the correct pulses coming into the counter chain and check the pulses propagate through the dividers correctly (the division factor of each chip is indicated in the manual), and can then be read at the PROM

Then check the blanking circuits (examples of how it works in the service manual) and finally check the divide by N section is working
__________________
SPECIALIST.....Knows everything about nothing
EXPERT..........Knows nothing about everything
PETERg0rsq is offline  
Old 10th Mar 2018, 3:41 pm   #5
G4WYZ MikeyP
Triode
 
G4WYZ MikeyP's Avatar
 
Join Date: Nov 2012
Location: Chorley, Lancashire, UK.
Posts: 21
Default Re: Racal 9081 phase detector problems.

Thank you for the info, I've found the fault.

On the Highest Frequency range when the frequency reaches 492MHz there is a slight shift in voltage IC11/7 causes the Q18/IC17 to oscillate at 3.9kHz. Removed using a 100nF decoupling capacitor now tunes down to 256MHz but with no end bottom stop.
This tends to point that the ROM is U/S, as I had to change IC11, which is a dual 4 channel multiplex, one channel was faulty.
G4WYZ MikeyP is offline  
Old 10th Mar 2018, 8:09 pm   #6
PETERg0rsq
Heptode
 
PETERg0rsq's Avatar
 
Join Date: Jun 2009
Location: St Helens, Lancashire, UK.
Posts: 641
Default Re: Racal 9081 phase detector problems.

The function of the PROM is quite complex, but easy to check.

On each band it monitors the counter outputs (A1 to A8 NOT including A0 to confuse matters!). It then triggers the outputs Q0 to Q5 depending on the band edge detection and if in High Octave or Low octave.

The band edges are also dependent on the channel spacing, and this information is fed into the multiplexer IC11 which selects the correct outputs from the ROM (depending on channel spacing selection) and feeds them into the Level converters (Q18, Q19 and Q22).

The PROM and IC11 supplies and outputs are floating due to the PROM supply requirements, so be careful when measuring what is going on!

Once level shifted the outputs can then be used to determine if the frequency selection is in band or not, and if it is too high, or too low:-

Q0 is used for band edge detection on 5, 10, 20 kHz channels
Q1 is used for band edge detection on 6.25, 12.5, 25, 50 kHz channels
Q2 is used for band edge detection on 15, 30, 60kHz channels

Output is LOW when in band, high when outside of band.


Q3 is used for high octave indication for 5, 10, 20 kHz channels
Q4 is used for high octave indication for 6.25, 12.5, 25, 50 kHz channels
Q5 is used for high octave indication for 15, 30, 60kHz channels

Outputs are high when in HIGH half of octave.


The high octave output is used to control some oscillators in the 9081 AND used to determine if the band edge detection is the LOW edge or the high edge using IC17.

Bands 2-5 are identical in operation, but band 1 has a different set of parameters in the ROM and is detected by A0

With the above information, and a logic analyser it is straightforward to determine if the PROM is bad.

If it is bad I have some "repair boards" made that can use an EPROM to replace the PROM, though it is not difficult to fabricate something.


However the PROM supplies are prone to failing, so check the voltages and Q20 and Q21.
__________________
SPECIALIST.....Knows everything about nothing
EXPERT..........Knows nothing about everything
PETERg0rsq is offline  
Old 10th Mar 2018, 10:13 pm   #7
PETERg0rsq
Heptode
 
PETERg0rsq's Avatar
 
Join Date: Jun 2009
Location: St Helens, Lancashire, UK.
Posts: 641
Default Re: Racal 9081 phase detector problems.

Just a further thought on what you saud

The 3.9kHz oscillation is probably correct, and by removing it with a capacitor to ground has disabled the function of the ROM!

The pulses enter Q11 on pin 1 and are used to pulse the outputs to the level shifters.

This pulsing is needed to generate tunning pulses on the Step up or step down lines, which are sent to the digital board to simulate tunning pulses to steer the counter in the correct direction to bring the divider to the correct value to bring the output back into the band.

Have you checked what is happening on STEP UP and STEP DOWN lines. If the N counter is too high then you should get pulses on the step down line, and if the n-Counter is too low you should get pulses on the step up line.

One of the first diagnostic steps I take with the 9081 and tunning issues is to disconnect these lines and see what the frequency coverage of the generator is, as if one of the N-Counters fails, or one of the N dividers fail, then the PROM detects an out of band signal and tries to tune back into band constantly.

This sounds like what you are experiencing when not able to tune down below 492 MHZ
__________________
SPECIALIST.....Knows everything about nothing
EXPERT..........Knows nothing about everything
PETERg0rsq is offline  
Old 12th Mar 2018, 1:26 pm   #8
G4WYZ MikeyP
Triode
 
G4WYZ MikeyP's Avatar
 
Join Date: Nov 2012
Location: Chorley, Lancashire, UK.
Posts: 21
Default Re: Racal 9081 phase detector problems.

Thank you for the very comprehensive info Peter.

Results
Once level shifted the outputs can then be used to determine if the frequency selection is in band or not, and if it is too high, or too low:-

Q0 is used for band edge detection on 5, 10, 20 kHz channels
Q1 is used for band edge detection on 6.25, 12.5, 25, 50 kHz channels
Q2 is used for band edge detection on 15, 30, 60kHz channels

Output is LOW when in band, high when outside of band.

Mine are still LOW even out of the band.

On reflection I've had an issue with the 1/2 octave oscillator before. This tends to confirm the ROM is faulty.

Can't really decide to continue with this as I just discovered another fault the 5MHz hi-stability oven is 2kHz off frequency adjustable to adjust as well an issues with the modulation.

Just a matter of interest is the repair board available?

Mike
G4WYZ MikeyP is offline  
Old 12th Mar 2018, 8:50 pm   #9
G4WYZ MikeyP
Triode
 
G4WYZ MikeyP's Avatar
 
Join Date: Nov 2012
Location: Chorley, Lancashire, UK.
Posts: 21
Default Re: Racal 9081 phase detector problems.

Update

The 5MHz oven is OK, the 300kHz I wrongly assumed this was derived from the standard (shouldn't work on this late in the evening), the 300kHz is derived in the 120MHz loop.

Mike
G4WYZ MikeyP is offline  
Old 12th Mar 2018, 10:37 pm   #10
PETERg0rsq
Heptode
 
PETERg0rsq's Avatar
 
Join Date: Jun 2009
Location: St Helens, Lancashire, UK.
Posts: 641
Default Re: Racal 9081 phase detector problems.

Have you checked the PROM inputs are all switching and confirmed if the PROM outputs are switching?

PSU for PROM can fail.

Is your PROM on-board, or on the digital board? If on the digital board the connecting cable could be a problem.


I have only had two PROMs fail in about 15 units.

I had manufactured some EPROM boards a few years ago. They were only intended for my own use though. I will need to review them before I could offer one.

My board replaces the PROM, the multiplexer, the level converters and the logic gates. I cant remember if I re-programmed the EPROM to just fire three outputs High, LOW and MIDDLE directly from the inputs, so I need to re-visit what I actually did.

My MK1 replacement board was a PIC PROM simulator which also worked well.

I have also just moved from Windows machine to Linux, so have to see if I can get my EPROM programmer to work.

Also need to get KiCad working again!
Attached Thumbnails
Click image for larger version

Name:	EPROM-2.jpg
Views:	92
Size:	121.8 KB
ID:	159130   Click image for larger version

Name:	EPROM-1.jpg
Views:	67
Size:	129.9 KB
ID:	159131  
__________________
SPECIALIST.....Knows everything about nothing
EXPERT..........Knows nothing about everything
PETERg0rsq is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 7:26 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.