|
The receiver seems to work fine when the AGC switch is set to S. However, in the M position there is noticeable distortion on both noise and signals. In the F position, the distortion becomes significant. The distortion sounds like clipping, but none is seen on the audio output with a scope. The S meter shows no difference with the three settings. I am using the Sherwood audio amp and power supply. I set up an HP sig gen to provide a pulsed RF signal. What I see is the AGC pulse in response to that input signal increases in voltage by x3 in the M position and by x7 in the F position compared to the S position. The only changes to the AGC circuit that I see when the delay is changed to the M & F positions are the time constants (two RC networks). The resistors in those networks check ok and to the limited extent that I can check the two capacitors, they seem ok. What types of capacitors are they and do they have a history of going bad? Any suggestions would be appreciated.
-- Bill N0CU
|
Any capacitor (especially an old one, even more so where it's in a
very hot environment all its life) can easily go bad.
What are the values of the capacitors?
73, Jim W7RY
On 3/12/2023 3:13 PM, Bill Leonard N0CU
wrote:
The receiver
seems to work fine when the AGC switch is set to S. However, in
the M position there is noticeable distortion on both noise and
signals. In the F position, the distortion becomes significant.
The distortion sounds like clipping, but none is seen on the
audio output with a scope. The S meter shows no difference with
the three settings. I am using the Sherwood audio amp and power
supply. I set up an HP sig gen to provide a pulsed RF signal.
What I see is the AGC pulse in response to that input signal
increases in voltage by x3 in the M position and by x7 in the F
position compared to the S position. The only changes to the AGC
circuit that I see when the delay is changed to the M & F
positions are the time constants (two RC networks). The
resistors in those networks check ok and to the limited extent
that I can check the two capacitors, they seem ok. What types of
capacitors are they and do they have a history of going bad? Any
suggestions would be appreciated.
--
Bill N0CU
--
Thanks and 73, Jim W7RY
|
They are the two large caps on the AGC board. They are the size of small electrolytic capacitors, but they are not polarized, so I don't know what type they are:
C73: 0.22 uF
C74: 0.047 uF
--
Bill N0CU
|
I recommend these in the proper values:
Please don't order some chinese junk from amazon or ebay. Use these
from Mouser or Digikey.
73, Jim W7RY
On 3/12/2023 7:01 PM, Bill Leonard N0CU
wrote:
They are the two large caps on the AGC board. They are the size of
small electrolytic capacitors, but they are not polarized, so I
don't know what type they are:
C73: 0.22 uF
C74: 0.047 uF
--
Bill N0CU
--
Thanks and 73, Jim W7RY
|
Thanks Jim.
I'll get them on order from Mouser.
I am curious what performance feature about these capacitors requires their use in the AGC circuit.
--
Bill N0CU
|
? S=Slow AGC ?response time M=Medium AGC ?response time F=Fast AGC ?response time
Fast is only really suitable for use in CW and RTTY. SSB will get quite distorted in the Fast AGC position and less so in the Medium position but still not zero for strong signals. Slow AGC is always best for SSB.
From what you describe, your radio is behaving as it should. The S meter reads, meaning that the AGC is working. The change in S meter response to a pulsed signal is expected because the S mater is tracking what it can insofar as the AGC is holding things back due to the time constant.
Try injecting a steady signal into the radio and you should see that, after the slow time constant is passed, the S meter will read the same in the Slow position as it does in the Fast position.
Those old capacitors you describe are paper caps, which is why they do not show a polarization marking. You can replace them with Mylar caps. ?Most Mylar caps you buy from reputable sources are rated to 630 volts and this presents no problem in terms of function or lifetime.
Gary
W0DVN
toggle quoted message
Show quoted text
On Mar 12, 2023, at 9:10 PM, Jim W7RY via groups.io <jimw7ry@...> wrote:
?
I recommend these in the proper values:
Please don't order some chinese junk from amazon or ebay. Use these
from Mouser or Digikey.
73, Jim W7RY
On 3/12/2023 7:01 PM, Bill Leonard N0CU
wrote:
They are the two large caps on the AGC board. They are the size of
small electrolytic capacitors, but they are not polarized, so I
don't know what type they are:
C73: 0.22 uF
C74: 0.047 uF
--
Bill N0CU
--
Thanks and 73, Jim W7RY
|
Gary,
The S meter is behaving as you described with a steady signal.
The only issue that has me puzzled is how the audio distortion increases as the AGC time constant is reduced. I haven't seen this with any other receiver I have owned, analog or digital. I can see how distortion due to clipping could occur if the AGC risetime isn't fast enough to keep the signal from clipping, but my understanding is that the R4-C AGC switch only varies the decay time, and I wouldn't expect that to cause clipping. So, if the radio is behaving as it should, is this a characteristic of the Drake AGC design? If so, it isn't a problem because I don't operate CW, however, I would simply like to know if there is a problem with my R4-C that I need to track down.
--
Bill N0CU
|
They are non polarized and more reliable than an electrolytic.
See here:
73, Jim W7RY
On 3/12/2023 7:36 PM, Bill Leonard N0CU
wrote:
Thanks Jim.
I'll get them on order from Mouser.
I am curious what performance feature about these capacitors
requires their use in the AGC circuit.
--
Bill N0CU
--
Thanks and 73, Jim W7RY
|
Also here:
Jim W7RY
On 3/12/2023 9:05 PM, Jim W7RY wrote:
They are non polarized and more reliable than an electrolytic.
See here:
73, Jim W7RY
On 3/12/2023 7:36 PM, Bill Leonard
N0CU wrote:
Thanks Jim.
I'll get them on order from Mouser.
I am curious what performance feature about these capacitors
requires their use in the AGC circuit.
--
Bill N0CU
--
Thanks and 73, Jim W7RY
--
Thanks and 73, Jim W7RY
|
Be sure the Drake AGC circuit hasn’t been modified.? Having worked on many R-4Cs over a 40 year time period it wasn’t unusual to have radios come into the lab
with AGC issues.?? Also in that time period I never had to replace a standard Drake AGC capacitor.
?
Attached is a PDF that includes component listings for the stock AGC and the two additional speed I typically added.
?
Rob, NC0B
?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]]
On Behalf Of Bill Leonard N0CU
Sent: Sunday, March 12, 2023 7:02 PM
To: [email protected]
Subject: Re: [DRAKE-RADIO] Puzzling problem with the AGC in my R4-C
?
Gary,
The S meter is behaving as you described with a steady signal.
The only issue that has me puzzled is how the audio distortion increases as the AGC time constant is reduced. I haven't seen this with any other receiver I have owned, analog or digital. I can see how distortion due to clipping could occur if the AGC risetime
isn't fast enough to keep the signal from clipping, but my understanding is that the R4-C AGC switch only varies the decay time, and I wouldn't expect that to cause clipping. So, if the radio is behaving as it should, is this a characteristic of the Drake
AGC design? If so, it isn't a problem because I don't operate CW, however, I would simply like to know if there is a problem with my R4-C that I need to track down.
--
Bill N0CU
|
Rob,
Thanks for the info.
I have a couple of questions:
-Is the distortion I am experiencing on the M & F AGC settings normal? If not, have you seen this problem before?
-The only mod I have found is the addition of a 1 Meg resistor in series with a 0.01 uF capacitor that was recommended by VE3EFJ in his article "4.0 R4B and R4C Mods and Tech". It was added on the AGC board to improve AGC transients. I haven't tried removing it. Should I?
--
Bill N0CU
|
Hi Bill,
?
There are multiple issues to consider.? I always felt the Fast AGC decay speed was useless, and installing an extra C value (or R/C pair) across the Fast setting
is a good idea.? I would keep the 1 meg + 0.01 in the Fast position.? I still wouldn’t use that Fast decay on SSB, but maybe it is useful in rare instances on CW.? This really fast decay speed may go back to early Drake QSK operation. ?This is from memory
in the days of the Drake 2A and 2B so it could well be inaccurate. ?
??
The other problem is the dual-diode product detector in the late R-4As, all R-4Bs, all R-4Cs and early TR-7s is terrible.? The instantaneous IF level during
AGC attack exceeds the BFO injection, causing awful (from my perspective) AGC attack distortion.? The faster the decay the more often the AGC attack distortion occurs. ?A simple dual-diode product detector (which is a mixer) should have at least 5X the LO
injection level as the IF level to avoid distortion.? This isn’t even close to being the case in those stock Drake product detectors.
?
Another issue is the stock dual-diode product detector allows detected audio to leak back into the IF, modulating the AGC voltage.? This may explain why even
during a steady-state test with a signal generator it also shows distortion changing with decay speed selection.? A doubly balanced Gilbert Cell product detector doesn’t have either of these problem.?
?
My first R-4C replacement product detector used an MC1496 Gilbert Cell chip that had a very high parts count.? Later I changed to an Texas Instrument TL-442
Gilbert Cell chip, but when that went out of production the NE602 or SA602 was on the market with only a modest parts count. ?Attached is a PDF with the schematic of the 602 Gilbert Cell product detector.? Note: in later TR-7s Drake changed to an MC1496 product
detector.? Déjà Vu.?? ?
?
I always wondered why Drake engineers were “tone deaf”, seemingly unaware of AGC attack distortion.? The original R-4 and early R-4As had a tube product detector
which did not have this problem.? I also had to fix the AGC and product detector distortion in my 75S-3C, so Drake didn’t have a monopoly on distortion.
?
73, Rob, NC0B
?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]]
On Behalf Of Bill Leonard N0CU
Sent: Sunday, March 12, 2023 9:51 PM
To: [email protected]
Subject: Re: [DRAKE-RADIO] Puzzling problem with the AGC in my R4-C
?
Rob,
Thanks for the info.
I have a couple of questions:
-Is the distortion I am experiencing on the M & F AGC settings normal? If not, have you seen this problem before?
-The only mod I have found is the addition of a 1 Meg resistor in series with a 0.01 uF capacitor that was recommended by VE3EFJ in his article "4.0 R4B and R4C Mods and Tech". It was added on the AGC board to improve AGC transients. I haven't tried removing
it. Should I?
--
Bill N0CU
|
The only time I ever used the FAST agc position, is when listening to AM signals, like SWBC.....esp on the 41 and 49m SWBC? bands...both of which can experience rapid, fast fades.?
What might make more sense is to use something like a 1 megohm pot and one cap. Then? (on paper) you should be able to tweak the agc from fast...to slow...and everything in-between.? It's just an RC time constant.?
?
What ever happened to ...'hang agc' ?? If you had a 90-200 msec? 'hang time', before the decay started, imo, it would sound a lot smoother, both on CW and also SSB.?
|
Rob,
Thanks for sharing your insights on this.
What I think you are saying is that AGC attack distortion is a “feature” of some of the designs in the receivers by Collins and Drake, and that this distortion increases as the decay time is reduced (which is what I am seeing). Also, the time constants used for the fast decays are off from what you recommend.
I don’t operate CW so the distortion I am seeing with the M and F settings is not an operational issue for me. However, the engineer in me wants to fix the AGC if something is broken and it is not operating as designed. What I have concluded from your comments is that this is not the case.
What I failed to mention is that I have installed your NE602 product detector. I didn’t mention it because I didn’t suspect it could be a contributor to an AGC problem. So, did this detector mod eliminate the AGC attack distortion, or only reduce it.
As an aside, before I built and installed your NE602 product detector I tried a 1496 mixer circuit that I had laying around from a previous project. The conversion gain was so much lower than the original design that it was not a viable option.
-- Bill N0CU
|
Hi Bill,
?
It was possible to get the MC1496 mixer to work as far as conversion gain.? The later chip options were simpler to implement.
?
Assuming the Drake AGC circuit hasn’t been changed, most of the attack distortion has been minimized with the NE602 mixer.? That said, I always wanted a slower
AGC decay than Drake offered, so my custom AGC switch implementation had 5 choices (including the almost useless Fast option).?? Unless dealing with QRN, I almost always used the new slowest decay speed on SSB.? To my picky ears the R-4C audio was fine, assuming
the terrible stock audio amp had also been replaced. (Note: I have retired and am not trying to sell a Sherwood AMP-4 module!)
?
I suppose there may still be some distortion during AGC attack since by definition the gain of the radio is changing in a few milliseconds.? One could define
this as distortion.? In a similar way Dynamic Noise Reduction in modern radios is always a tradeoff between reducing noise and living with DSP artifacts.? With my Icom I never run noise reduction on SSB above 2 out of 10 since the digital artifacts annoy me
more than the noise.? That of course is a subjective statement while others run NR as high as 8 out of 10.?
?
I have operated 20 out of the top 25 transceivers on my website table in contests, and only Icom and possibly Kenwood offers a slow enough AGC decay to suite
my operating preferences.? Those brands that are still too fast on SLOW are Apache, Elecraft, Flex and Yaesu. This isn’t necessarily an AGC attack distortion issue as simply me not liking band noise to be as loud as the station I am listening to if the person
speaking pauses for a fraction of a second.
?
On the other hand with at least the Yaesu FT-710, if the AGC is on AUTO, on CW the AGC attack distortion is absolutely terrible.? Are design engineers deaf??
I just don’t get it, or do OEMs only listen to their radios in the lab with a signal generator and not on the air?? Why didn’t Drake notice that there wasn’t AGC attack distortion in the original R-4 but in later R-4As and the follow-on ?products the audio
had seriously degraded??
?
Maybe radio OEMs should have a Hi Fi person with “golden ears” on their staff !?? Hi Hi
?
73, Rob, NC0B ?
?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]]
On Behalf Of Bill Leonard N0CU
Sent: Monday, March 13, 2023 9:48 AM
To: [email protected]
Subject: Re: [DRAKE-RADIO] Puzzling problem with the AGC in my R4-C
?
Rob,
Thanks for sharing your insights on this.
What I think you are saying is that AGC attack distortion is a “feature” of some of the designs in the receivers by Collins and Drake, and that this distortion increases as the decay time is reduced (which is what I am seeing). Also, the time constants used
for the fast decays are off from what you recommend.
I don’t operate CW so the distortion I am seeing with the M and F settings is not an operational issue for me. However, the engineer in me wants to fix the AGC if something is broken and it is not operating as designed. What I have concluded from your comments
is that this is not the case.
What I failed to mention is that I have installed your NE602 product detector. I didn’t mention it because I didn’t suspect it could be a contributor to an AGC problem. So, did this detector mod eliminate the AGC attack distortion, or only reduce it.
As an aside, before I built and installed your NE602 product detector I tried a 1496 mixer circuit that I had laying around from a previous project. The conversion gain was so much lower
than the original design that it was not a viable option.
--
Bill N0CU
|
Rob Sherwood