RA17 AGC question - UK Vintage Radio Repair and Restoration Discussion Forum

Keith · 14th Jan 2008, 8:53 pm

Can anyone on the forum explain the intended function of the wire ended thermionic diode V24 in this receiver? It seems to be configured as a forward biased diode (for +ve voltages) across the AGC line but is not mentioned in any of the literature that I have. I found that the 20Mohm (2*10Mohm) resistor from its anode to the HT line was disconnected in my receiver. Although the resistor values were within tolerance, re-connecting them appears to positively offset that part of the AGC line rendering it largely ineffective. Any advice appreciated.

MichaelR · 14th Jan 2008, 9:29 pm

Keith,

That is the original AGC diode, I believe a lot of these were replaced in service with a silicon diode type IN16. It is 15 years ago since I rebuilt a RA117E so I am not familiar with the circuitry of the RA17L. At that time I read everything I could about the RA17 and RA117 before I made the decision as to what to buy.I can tell you that there were many variants of the RA17L and it appeared to me that there were changes made espcially to the AGC section to suit special needs of end users.

Regards
Mike

Mike

Skywave · 15th Jan 2008, 12:46 am

I had a good long think about this - and came up with the following analysis.

In the original RA-17, V24 doesn't appear: AGC is applied to the 1st. IF amp. (V14), 2nd. IF amp. (V16) and IF O/P (V17) only. There is no AGC voltage applied to the RF amp. (V3, E180F) at all. Consequently, the RF amp. could easily be overloaded - especially when the receiver is operated in close proximity to an associated transmitter.
The RA-17L - which uses a cascode RF amp. (ECC189) - however, does have AGC applied to this stage (

) - which, of course, circumvents this overload problem. However, to optimise the S/N ratio, the AGC voltage on this RF amp. is only required to 'come on' for large signals. Hence, by biassing the AGC line to the RF amp. only slightly positive - by the action of V24 and the two 10M-ohm resistors to HT+, the -ve AGC line will have to become significantly large to overcome this +ve bias. That is, it will only go -ve for really strong signals, so the RF amp. becomes the last stage to have its gain reduced under these conditions. Note that R91A (470K) substantially reduces this +ve bias from upsetting the -ve AGC voltage that is applied to the 100Kc/s IF amps.

As for as I know, the RA-17L was derived from the RA-17: a comparison of the two cct. diags. clearly reveals many refinements of the former over the latter. I can only conclude that the fitting of V24 is one of these - possibly for the reason suggested above.
Going way, way back to the mid-'60's, I do recall performing some dynamic range measurments on an RA-17L - and was truly amazed at the overload performance: something like 1-volt of aerial input signal before serious overloading took place!

HTH

Al / Skywave.

Keith · 15th Jan 2008, 12:18 pm

Mike/Al,

Thank you both for your replies, especially to you, Al, for your excellent detailed explanation.

I guessed that the use of three separate diodes in the AGC circuit meant that it was more complicated than I had thought. What I hadn't appreciated was that the line labelled "AGC line" on my marginally legible circuit was only used by the front end. Looking at it with the benefit of your description I can see that it's effectively a two stage delayed AVC system. My set is actually the RA17 C12, i.e. an American Mk3, but clearly has the RA17L improvements.

I think its time for me to open it up again and see if it will indeed work as intended with the +ve bias put back. When I tried it briefly before it seemed to upset the long time constant operation. I think I need to check out the values of the associated R's and C's.

I see on the recelectronics site that there seem to have been some rather heated arguments about the RA17 AGC

. Also I have a copy of another article suggesting that the circuit can be improved for SSB operation by decreasing the attack time so I might have a play around with that.

Having said all that I still think its a superb receiver - only problem is when you have to lift it

.

Cheers,

MichaelR · 15th Jan 2008, 3:14 pm

Beaten only by the R390A in my humble opinion Keith.

regards
Mike

Skywave · 15th Jan 2008, 5:39 pm

OTOH & IMHO, the RA-17 has a much more attractive front panel than the R390A .

Al / Skywave.

RobinBirch · 16th Jan 2008, 1:08 pm

And it's not that heavy when compared to an AR88/B40 etc etc

Robin

14th Jan 2008, 8:53 pm	#1
Keith Heptode Join Date: Jan 2003 Location: Tewkesbury, Gloucestershire, UK. Posts: 693	RA17 AGC question Can anyone on the forum explain the intended function of the wire ended thermionic diode V24 in this receiver? It seems to be configured as a forward biased diode (for +ve voltages) across the AGC line but is not mentioned in any of the literature that I have. I found that the 20Mohm (2*10Mohm) resistor from its anode to the HT line was disconnected in my receiver. Although the resistor values were within tolerance, re-connecting them appears to positively offset that part of the AGC line rendering it largely ineffective. Any advice appreciated. __________________ Keith Yates - G3XGW VMARS & BVWS member http://www.tibblestone.com/oldradios/Old_Radios.htm

14th Jan 2008, 9:29 pm	#2
MichaelR Retired Dormant Member Join Date: Mar 2006 Location: Sheffield, South Yorkshire, UK. Posts: 1,587	Re: RA17 AGC question Keith, That is the original AGC diode, I believe a lot of these were replaced in service with a silicon diode type IN16. It is 15 years ago since I rebuilt a RA117E so I am not familiar with the circuitry of the RA17L. At that time I read everything I could about the RA17 and RA117 before I made the decision as to what to buy.I can tell you that there were many variants of the RA17L and it appeared to me that there were changes made espcially to the AGC section to suit special needs of end users. Regards Mike Mike

15th Jan 2008, 12:46 am	#3
Skywave Rest in Peace Join Date: Jun 2006 Location: Chard, South Somerset, UK. Posts: 7,457	Re: RA17 AGC question I had a good long think about this - and came up with the following analysis. In the original RA-17, V24 doesn't appear: AGC is applied to the 1st. IF amp. (V14), 2nd. IF amp. (V16) and IF O/P (V17) only. There is no AGC voltage applied to the RF amp. (V3, E180F) at all. Consequently, the RF amp. could easily be overloaded - especially when the receiver is operated in close proximity to an associated transmitter. The RA-17L - which uses a cascode RF amp. (ECC189) - however, does have AGC applied to this stage () - which, of course, circumvents this overload problem. However, to optimise the S/N ratio, the AGC voltage on this RF amp. is only required to 'come on' for large signals. Hence, by biassing the AGC line to the RF amp. only slightly positive - by the action of V24 and the two 10M-ohm resistors to HT+, the -ve AGC line will have to become significantly large to overcome this +ve bias. That is, it will only go -ve for really strong signals, so the RF amp. becomes the last stage to have its gain reduced under these conditions. Note that R91A (470K) substantially reduces this +ve bias from upsetting the -ve AGC voltage that is applied to the 100Kc/s IF amps. As for as I know, the RA-17L was derived from the RA-17: a comparison of the two cct. diags. clearly reveals many refinements of the former over the latter. I can only conclude that the fitting of V24 is one of these - possibly for the reason suggested above. Going way, way back to the mid-'60's, I do recall performing some dynamic range measurments on an RA-17L - and was truly amazed at the overload performance: something like 1-volt of aerial input signal before serious overloading took place! HTH Al / Skywave. Last edited by Skywave; 15th Jan 2008 at 12:51 am. Reason: Grammar!

15th Jan 2008, 12:18 pm	#4
Keith Heptode Join Date: Jan 2003 Location: Tewkesbury, Gloucestershire, UK. Posts: 693	Re: RA17 AGC question Mike/Al, Thank you both for your replies, especially to you, Al, for your excellent detailed explanation. I guessed that the use of three separate diodes in the AGC circuit meant that it was more complicated than I had thought. What I hadn't appreciated was that the line labelled "AGC line" on my marginally legible circuit was only used by the front end. Looking at it with the benefit of your description I can see that it's effectively a two stage delayed AVC system. My set is actually the RA17 C12, i.e. an American Mk3, but clearly has the RA17L improvements. I think its time for me to open it up again and see if it will indeed work as intended with the +ve bias put back. When I tried it briefly before it seemed to upset the long time constant operation. I think I need to check out the values of the associated R's and C's. I see on the recelectronics site that there seem to have been some rather heated arguments about the RA17 AGC . Also I have a copy of another article suggesting that the circuit can be improved for SSB operation by decreasing the attack time so I might have a play around with that. Having said all that I still think its a superb receiver - only problem is when you have to lift it . Cheers, __________________ Keith Yates - G3XGW VMARS & BVWS member http://www.tibblestone.com/oldradios/Old_Radios.htm

15th Jan 2008, 3:14 pm	#5
MichaelR Retired Dormant Member Join Date: Mar 2006 Location: Sheffield, South Yorkshire, UK. Posts: 1,587	Re: RA17 AGC question Beaten only by the R390A in my humble opinion Keith. regards Mike

15th Jan 2008, 5:39 pm	#6
Skywave Rest in Peace Join Date: Jun 2006 Location: Chard, South Somerset, UK. Posts: 7,457	Re: RA17 AGC question OTOH & IMHO, the RA-17 has a much more attractive front panel than the R390A . Al / Skywave.

16th Jan 2008, 1:08 pm	#7
RobinBirch Hexode Join Date: Apr 2003 Location: Cotswolds, UK. Posts: 465	Re: RA17 AGC question And it's not that heavy when compared to an AR88/B40 etc etc Robin