Racal RA-17 AGC - UK Vintage Radio Repair and Restoration Discussion Forum

dtvmcdonald · 22nd Jan 2018, 6:24 pm

I have a Racal RA17-C12, the US model, which works very well.

But it has a very unusual "feature" which I recently noticed. It seems to have 27 volts of AGC delay. As best I can tell from both a simulation based on the schematic and actual tests , the IF voltage driving the AGC diode has to overcome 27 volts of bias before it begins conducting. This makes for a very large range of input levels before the AG cuts in and then an extremely steep
curve up to as much as a volt of RF input.

As a byproduct, the S meter present on the US version simply does not
read S units. My physical meter is a replacement, but I found one on ebay right now so I can copy the scale. My meter is the right current full scale but I don't know if its the correct resistance.

This behavior is quite different from any other radio I have, be it expensive boatanchor/military, regular domestic AM-SW radio, or really cheap ham radio set.

Comments? Could the US version be different from the schematic I have?
I don't recall what values were in fact in the set; this was in an
area that had been heavily attacked by the "mad snipper" that had got at my set before I restored it, so I just put in what the schematic said.

Silicon · 22nd Jan 2018, 11:31 pm

The English version of the RA17L has an AGC circuit that connects to the -25V and the +220V HT supply.

My version has a miniature thermionic diode on the AGC bus.

amornummi · 23rd Jan 2018, 12:14 am

The AGC line usually operates at a few volts.
About 14 volts needed for total cutoff.
The only place for - 25 volts to get in is via the RF gain control, does this do anything?
Check R120+R121 , and R122

amornummi · 23rd Jan 2018, 12:18 am

Failed to attach simplified circuit , here it is

https://drive.google.com/file/d/0B0E...ew?usp=sharing

dtvmcdonald · 23rd Jan 2018, 10:02 pm

More tests indicate that its working right, the adjustments are just hopelessly
sensitive. The calibration depends on the exact level of AGC delay added via
the nominally 100K resistor to B+, the exact series resistance in the AGC circuit, and
the exact setting of the S-meter zero control.

And the calibration changes for 30 minutes while it warms up.

The "knee" in the curve is about s8 or s9 and is way too sharp to
make the thing (that is, the meter) stably quantitative. The up side of this is that once
the AGC kicks in it hold the level really constant.

Keith · 25th Jan 2018, 7:46 pm

I run an RA17-C12 most days and the S meter seems to work fine.I did have to replace the original Earnest Turner movement which had its hairsprings completely corroded away. Calibration is correct at around 50uV for S9 and it correlates fairly well over the range against an SDR receiver running PowerSDR software.

dtvmcdonald · 26th Jan 2018, 6:58 pm

Quote:

Originally Posted by Keith

I run an RA17-C12 most days and the S meter seems to work fine.I did have to replace the original Earnest Turner movement which had its hairsprings completely corroded away. Calibration is correct at around 50uV for S9 and it correlates fairly well over the range against an SDR receiver running PowerSDR software.

My meter reads zero uA at about S2.5. It then goes smoothly up
to S8.5 at about 145 uA. Then it goes up to 60dB over S9 at about
154uA. Then it goes up to 90 dB over S9 at 195 uA.

That's with the set's attentuator full CW (not attenuation).

I think that the AGC feed to the RF amp needs a bit of reduction
(like cutting it to 2/3 the value fed to the two IF amps.

Keith · 26th Jan 2018, 7:23 pm

Not sure where you are measuring your uA. The meter on my receiver is as shown in the picture.

dtvmcdonald · 26th Jan 2018, 11:29 pm

Those values are read on my meter in that spot. Its a replacement 200 uA meter,
which tests to be right on the spot 200 uA. It has a plain uA scale. I have no idea
whether the meter has the right resistance to read correctly in all three switch positions.
With no resistor is parallel its wildly offscale on the S-meter scale and I have a 270
ohm resistor in parallel. This has very little effect on the audio position and
a very modest effect on the RF level position but a very large effect on the S-meter
position, especially as I have to rezero a lot when adding that resistor.

Note that at the moment it reads about right at 60 over S9.

22nd Jan 2018, 6:24 pm	#1
dtvmcdonald Pentode Join Date: Mar 2016 Location: Champaign, Illinois, USA. Posts: 227	Racal RA-17 AGC I have a Racal RA17-C12, the US model, which works very well. But it has a very unusual "feature" which I recently noticed. It seems to have 27 volts of AGC delay. As best I can tell from both a simulation based on the schematic and actual tests , the IF voltage driving the AGC diode has to overcome 27 volts of bias before it begins conducting. This makes for a very large range of input levels before the AG cuts in and then an extremely steep curve up to as much as a volt of RF input. As a byproduct, the S meter present on the US version simply does not read S units. My physical meter is a replacement, but I found one on ebay right now so I can copy the scale. My meter is the right current full scale but I don't know if its the correct resistance. This behavior is quite different from any other radio I have, be it expensive boatanchor/military, regular domestic AM-SW radio, or really cheap ham radio set. Comments? Could the US version be different from the schematic I have? I don't recall what values were in fact in the set; this was in an area that had been heavily attacked by the "mad snipper" that had got at my set before I restored it, so I just put in what the schematic said.

22nd Jan 2018, 11:31 pm	#2
Silicon Nonode Join Date: Jun 2007 Location: Coulsdon, London, UK. Posts: 2,163	Re: Racal RA-17 AGC The English version of the RA17L has an AGC circuit that connects to the -25V and the +220V HT supply. My version has a miniature thermionic diode on the AGC bus.

23rd Jan 2018, 12:14 am	#3
amornummi Triode Join Date: Mar 2008 Location: Huddersfield, West Yorkshire, UK. Posts: 45	Re: Racal RA-17 AGC The AGC line usually operates at a few volts. About 14 volts needed for total cutoff. The only place for - 25 volts to get in is via the RF gain control, does this do anything? Check R120+R121 , and R122

23rd Jan 2018, 12:18 am	#4
amornummi Triode Join Date: Mar 2008 Location: Huddersfield, West Yorkshire, UK. Posts: 45	Re: Racal RA-17 AGC Failed to attach simplified circuit , here it is https://drive.google.com/file/d/0B0E...ew?usp=sharing

23rd Jan 2018, 10:02 pm	#5
dtvmcdonald Pentode Join Date: Mar 2016 Location: Champaign, Illinois, USA. Posts: 227	Re: Racal RA-17 AGC More tests indicate that its working right, the adjustments are just hopelessly sensitive. The calibration depends on the exact level of AGC delay added via the nominally 100K resistor to B+, the exact series resistance in the AGC circuit, and the exact setting of the S-meter zero control. And the calibration changes for 30 minutes while it warms up. The "knee" in the curve is about s8 or s9 and is way too sharp to make the thing (that is, the meter) stably quantitative. The up side of this is that once the AGC kicks in it hold the level really constant.

25th Jan 2018, 7:46 pm	#6
Keith Heptode Join Date: Jan 2003 Location: Tewkesbury, Gloucestershire, UK. Posts: 693	Re: Racal RA-17 AGC I run an RA17-C12 most days and the S meter seems to work fine.I did have to replace the original Earnest Turner movement which had its hairsprings completely corroded away. Calibration is correct at around 50uV for S9 and it correlates fairly well over the range against an SDR receiver running PowerSDR software. __________________ Keith Yates - G3XGW VMARS & BVWS member http://www.tibblestone.com/oldradios/Old_Radios.htm

26th Jan 2018, 7:23 pm	#8
Keith Heptode Join Date: Jan 2003 Location: Tewkesbury, Gloucestershire, UK. Posts: 693	Re: Racal RA-17 AGC Not sure where you are measuring your uA. The meter on my receiver is as shown in the picture. Attached Thumbnails __________________ Keith Yates - G3XGW VMARS & BVWS member http://www.tibblestone.com/oldradios/Old_Radios.htm

26th Jan 2018, 11:29 pm	#9
dtvmcdonald Pentode Join Date: Mar 2016 Location: Champaign, Illinois, USA. Posts: 227	Re: Racal RA-17 AGC Those values are read on my meter in that spot. Its a replacement 200 uA meter, which tests to be right on the spot 200 uA. It has a plain uA scale. I have no idea whether the meter has the right resistance to read correctly in all three switch positions. With no resistor is parallel its wildly offscale on the S-meter scale and I have a 270 ohm resistor in parallel. This has very little effect on the audio position and a very modest effect on the RF level position but a very large effect on the S-meter position, especially as I have to rezero a lot when adding that resistor. Note that at the moment it reads about right at 60 over S9.