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 > General Vintage Technology > Components and Circuits

Notices

Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets.

Closed Thread
 
Thread Tools
Old 20th Nov 2022, 4:32 pm   #1
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default AD603 IF Amp.

Hi peeps. Complicated circuit analysys is not my strongest point, and I usually put faith into the manufacturers application data, assuming they know what they are doing.
See attached circuit. It is billed as an AGC amplifier ... which was duly laid out on a pcb and tested. Unfortunately iit has no AGC function. The AGC line stubbornly sits at 5 v whatever the RF input. It makes a good attenuator.

The output only rises with large amounts of RF in. There is no sign of the 84 db gain it is supposed to give. In desperation I disconnected the AGC line and fitted a 10K pot from + to Gnd. and then I can get gain, but not the 84 db promised. Has A D put in one of their "errors", that I am not clever enough to work out ?
No doubt Jeremy and David will be rolling their eyes upward, but no matter
Your comments would be gratefully received.
Oh yes incidentally some of the very accurate resistors in the original circuit had to be trimmed, as R12 = 820R and the pin 4 bias has 2K7 resistors.
Attached Thumbnails
Click image for larger version

Name:	AD603 IF Strip ver 4.jpg
Views:	167
Size:	48.5 KB
ID:	268407  
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 21st Nov 2022, 12:40 am   #2
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: AD603 IF Amp.

I can suggest some basic checks if that helps?

The first thing to do would be to make sure the circuit isn't unstable and causing self AGC effects. I guess you have done this already.

If you have a 9V supply, then the voltage at the base of the PNP Q2 should be about 8V if the potential divider network is OK. There should also be about 4.6V at the base of Q1 if the divider is OK.

If the base of PNP Q2 is at (say) 8V then the emitter of Q2 will be one (very weakly) biased diode drop higher or about 0.6V higher. This means there will be 0.4V across R18 (9V - 8.6V = 0.4V). So Q2 will act as a constant current source, and this current will try and charge C9 and C10 with a current of 0.4V/R18 = 0.4V/1600 = 250uA. This effectively charges up the AGC line for max gain.

If Q1 (2N3904) and the AD603 weren't there, then Q2 would be able to charge C9 and C10 up to about 8V or so.

I think the way the AGC is meant to be biased is is that Q1 (2N3904) has to be biased to ensure that it does not constantly take some of the 250uA from Q2 and deny C9 and C10 from charging up in the condition where there is little or no signal input. If it could constantly steal current like this, then the circuit would not be able to charge C9 and C10 very well and it would not achieve full gain. So, the next thing to do would be to measure the voltage across R12 820R with no signal input. If there is significant voltage drop across it, then it means that Q1 is already stealing a big chunk of the constant current from Q2, and this denies the system from producing full gain. This would not be good.

Probably the easiest way to imagine the role of Q1 is that it should be biased such that it hardly steals any current from Q2 in the no signal condition. When the AD603 produces a large RF output then this will cause the detector Q1 (2N3904) to demand a bigger share of the constant current from Q2. This will reduce the voltage at C9 and C10.

The detector Q1 biasing has to be set very carefully, a bit like a hair trigger. Maybe your circuit causes the DC level at the output of the AD603 pin 7 to be offset enough that it is causing the detector to be triggered all the time? What DC voltage do you see at pin7 of the final AD603 when there is no signal input? This would be a good starting point for the design checking of the Q1 biasing resistors.

I hope I've got this OK, I've not used the AD603 before. I have done design work with the AD600 and this is a similar chip. However, this was about 25 years ago, and I used a different AGC/detector circuit.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 21st Nov 2022, 12:57 am   #3
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: AD603 IF Amp.

A quick and crude experiment could be done by removing R13, R14, R15 and R16 and replacing them with a 10K trimmer with the trim tap at the base of Q1.

Keep the 100nF cap at the base of Q1.

Then set the trimmer such that the AGC triggering behaves as expected. However, it's probably better to measure the DC voltage at the second AD603 pin 7 in the 'no signal' condition and use this to design the resistor values instead of fitting the trim pot.

I think what is meant to happen at AGC equilibrium is that Q2 puts in 250uA to C9 and C10 but the Q1 detector transistor will be stealing 250uA. Therefore, the AGC voltage stays constant. What you don't want to happen is for Q1 to be incorrectly biased such that it is already stealing more than 250uA in the 'no signal' condition.
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 21st Nov 2022 at 1:04 am.
G0HZU_JMR is offline  
Old 21st Nov 2022, 1:51 am   #4
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: AD603 IF Amp.

This all assumes that I understand the circuit correctly, of course...

However, I think it's possible to predict at what DC voltage (at pin 7 of the second AD603) the circuit will start to misbehave. If the base of Q1 is at about 4.6V then for Q1 to conduct weakly there will be maybe 0.5V dropped across its Vbe junction. So the emitter would be at 4.1V DC in this case.

The problems will occur if 250uA is drawn by Q1 in the 'no-signal' condition. 250uA through your 820R resistor R12 drops another 205mV.

So, I think your circuit could have problems if the DC voltage at pin 7 of the second AD603 is close to (or lower) than about 4.6 - 0.5 - 0.205 = 3.895V DC in the 'no-signal' condition.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 21st Nov 2022, 1:05 pm   #5
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: AD603 IF Amp.

Hi Jeremy. WOW.. thanks for the in depth analyses. I have printed out all the comments and take this PM to work through them. I will report back later. Looking at the very exact component values, I guessed this may be a theoretical circuit. I wonder if AD has actually built it. If I cannot get it to work, I have an RF AGC circuit that can be shoved in
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 23rd Nov 2022, 11:27 pm   #6
Herald1360
Dekatron
 
Herald1360's Avatar
 
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,527
Default Re: AD603 IF Amp.

The resistors look like standard E24 values except for that odd 800R.
__________________
....__________
....|____||__|__\_____
.=.| _---\__|__|_---_|.
.........O..Chris....O
Herald1360 is offline  
Old 24th Nov 2022, 9:22 pm   #7
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: AD603 IF Amp.

Hi Chris. Re resistors. I only stock the normal E12 range. However I do have a "Sample" folder with lots of intermediate values. If you look at the original Application notes, there are some odd values.
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 25th Nov 2022, 1:47 am   #8
Herald1360
Dekatron
 
Herald1360's Avatar
 
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,527
Default Re: AD603 IF Amp.

Ah, just done that, mystery solved- they're all standard E96 (1%) series values .
__________________
....__________
....|____||__|__\_____
.=.| _---\__|__|_---_|.
.........O..Chris....O
Herald1360 is offline  
Old 27th Nov 2022, 2:06 pm   #9
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: AD603 IF Amp.

Hi Chris. Im not in the mindset to buy E96 range SMD's. That particular idea is shelved..
I have "nicked" part of the schematic, as in the pin 2 bias chain. The Pin 1 will be controlled from a Plessey SL623 AM/SSB device which has the AGC pin (3). However this pin has the AGC voltage in the wrong sense. I.e I need the AGC Voltage to decrease as the signal gets larger, not increase, thus an inverting OP amp, suitably adjusted for gain is utilised. Back soon
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 2nd Dec 2022, 11:43 am   #10
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: AD603 IF Amp.

Just an additional bit of info..... DO NOT run this IC with the ground pin missing. Stupidly my new pcb layout omitted the ground pin in the second IC, I didnt realise till the second one was removed. I checked the artwork and found, when tidying up the zero level tracks, I omitted pin 6 to ground. Fortunately I have replacements. Back up and running, but costly.
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 2nd Dec 2022, 12:23 pm   #11
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,801
Default Re: AD603 IF Amp.

Barrie Gilbert was a technical bod at Plessey and involved in that series of radio mixer/IF/demod ICs. James Bryant, G4CLF, was also there and that's how the G4CLF SSB transceiver board using them came about.

Barrie was also at Tektronix for some time, and his non-linear techniques and multiplier architecture went into DC controlled gain in wideband amps in scopes.

Analog Devices inc. poached him. They created a new division in his area just to get hold of him. As a result, ADI got some new mixer and variable-gain devices and his input into all sorts of things. So the plessey SLxxx series, Tektronix, and some of the funkier ADI parts all are connected.

Oh, and James moved to ADI as marketing manager for Europe. We used to have a good natter whenever he came to the 'ferry.

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 2nd Dec 2022, 3:33 pm   #12
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: AD603 IF Amp.

Its nice to know the "relations" and "relatives" in the industry. Not all newbie graduates then.. Still "Old Skool" about
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 3:40 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.