+++++++++++++++++++++ See also K2_Low_S-Meter_Sensitivity +++++++++++++++++++++ From: "Don Brown" To: "Elecraft" Subject: Re: [Elecraft] S-Meter Sensitivity Date: Tue, 4 Jun 2002 08:59:29 -0500 Hi Thanks to all who corrected my mistake (I think it was the only one yeste= rday) Yes the db attenuation is 20 X the log of the voltage ratio. Howeve= r I set the 191 with a HP 3400C true RMS voltmeter. This meter is accurat= e to 10 mhz so using 7.100 mhz as the test frequency it works out fine (I= have cross checked it with a scope). I set the 191 to 7.100 mhz and 5000= uV out with the 3400C terminated with 50 ohms. Then removing the termina= tor connect to the step attenuator set to 40 db and then out to the K2. I= adjusted the CAL S LO per the manual and adjust the CAL S HI to get S-9 = on the S-meter. All settings made with the K2 attenuator and preamp off. I just checked four K2 radios I have in my lab and the CAL S HI needs to = be reduced to around 10 to 12 to get a reading of S9 with the corrected s= etup above. This means the S-meter does read a little low on these radios= with the manual calibration. Don Brown KD5NDB ----- Original Message ----- From: k6se at juno.com Sent: Monday, June 03, 2002 11:27 PM To: DolfinDon at msn.com; elecraft at mailman.qth.net Subject: Re: [Elecraft] S-Meter Sensitivity Don, KD5NDB wrote: "I have a Tektronix 191 leveled RF signal generator and a HP step attenuator that goes down to 100 db. I set the RF level to 0.5 Volts and the attenuator to 40 db giving 50 UV. With this connected to the K2 I adjust the Cal S HI so the meter reads S9. This will usually come out +or- a little from the manual set point." =3D=3D=3D=3D=3D=3D=3D=3D It sounds like Don was using the power ratio for dB, where a reduction of 10 dB is a reduction in power to 1/10, therefore a reduction of 40 dB is a reduction in power of 1/10000. However, it takes 20 dB of attenuation to reduce the voltage by a factor of 10, therefore to reduce 500,000 uV to 50 uV would require 80 dB of attenuation. His 40 dB only reduced the signal into the receiver to 5000 uV. Someone else commented that the Tek 191 is calibrated in peak-to-peak voltage, while S-meter sensitivity measurments are in RMS voltage. So I guess Don should redo his S-meter calibration by setting his Tek 191 output to 14,142.27125 uV P-P (which is 5,000 uV RMS). Then 40 dB of attenuation would reduce the signal generator output to 50 uV. If it took 5000 uV to make his S-meter read S9, his S-meter is stingier than mine was, by far. 73, de Earl, K6SE 66 countries on K2 #2622 and counting..... ++++++++++++++++++ To: DolfinDon at msn.com, elecraft at mailman.qth.net Date: Tue, 4 Jun 2002 08:30:37 -0700 Subject: Re: [Elecraft] S-Meter Sensitivity From: k6se at juno.com Hi, Don, You wrote: "I just checked four K2 radios I have in my lab and the CAL S HI needs to be reduced to around 10 to 12 to get a reading of S9 with the corrected setup above. This means the S-meter does read a little low on these radios with the manual calibration." ========== Reducing CAL S HI by a count of 10 to 12 sounds like quite a bit. If those four K2s had their S-meters calibrated to S9 as per the procedure in the manual, how many uV would it take for them to read S9? This would give us all a pretty good idea of the S-meter sensitivity of the average K2. I'm curious to know, with your S-meter calibrated as above, how many uV does it take for the your S-meter to read S5, 20 over 9, and 40 over 9 (the other front panel markings for the S-meter)? Does it retain reasonable accuracy (where S5 is approx 4 uV, 20 over 9 = 500 uV and 40 over 9 = 5000 uV)? For S5 I'm assuming that each S-unit is 6 dB (which means a reading of S1 would require a signal 48 dB below 50 uV). 73, de Earl, K6SE 67 countries with K2 #2622 and counting..... ++++++++++++++++++ Subject: Re: [Elecraft] K2 S-Meter Sensitivity Date: Tue, 4 Jun 2002 13:14:03 -0400 To: , From: Bill Coleman On 6/3/02 6:39 PM, k6se at juno.com at k6se at juno.com wrote: >The S-meter on my K2 (with pre-amp on) seemed to be quite stingy compared >to my FT-1000MP, so I searched the archives and found a wealth of >information on this subject. I had this problem with my K2. I would tune to SWBC stations on 40m, which read S9+40dB on my TS-430S, and the K2 would register them as S5. >Most commonly, D1 on the control board was blamed for this problem and >selecting a diode with a higher forward voltage drop was the fix. I did this, but it didn't have much effect in my unit. >In the archives, Gary Surrency said that if the output of the 8-volt >regulator on the control board was on the high side (mine reads 8.2 >volts), it could also contribute to the problem because it would elevate >the AGC threshhold voltage. He went on to say that if you notice a large >drop in IF hiss when you turn the AGC off that it would be beneficial to >reduce the value of R1 on the control board from 51k to 33k. I did this mod, and it didn't change much of anything. I do think he meant a large increase in hiss with AGC off (eg more IF stage gain). I haven't done any measurements, but I did find a solution. The transistor isolation amp in the AGC, which I think is Q12, was bad in my unit. Since I didn't have any PN2222As, I replaced it with a 2N2222, which is in a TO-18 (metal case) package. (You have to be sure to get the leads in the right holes, too) After this, S-meter response seemed much more lively. Those SWBC stations on 40m are typically S9+10-20dB. Bill Coleman, AA4LR, PP-ASEL Mail: aa4lr at arrl.net Quote: "Not within a thousand years will man ever fly!" -- Wilbur Wright, 1901 +++++++++++++++++++ From: "Don Brown" To: , "Elecraft" Subject: Re: [Elecraft] S-Meter Sensitivity Date: Tue, 4 Jun 2002 12:28:41 -0500 Hi OK here are the numbers with the K2 calibrated per the manual and a S-9 i= ndication K2 #1 550uV K2 #2 370uV K2 #3 380uV K2 #4 540uV K2 #4 S+40 54000uV S+20 5400uV S9 540uV S5 65uV S1 .82uV As you can see the k2's are off by a factor of about 10 that's 20 db so I= don't think it would hurt to bump the CAL S HI down a little into the 10= to 15 range. Also this is at 7.100 mHz I have not done a check on other = bands but I am sure it will vary some because the receiver gain varies fr= om band to band. The S-meter does track fairly well as you can see from t= he numbers. One way that will get you close is set the RF gain to 9 o'clo= ck rather than fully counter clockwise and then do the CAL S HI. =20 Don Brown KD5NDB ----- Original Message ----- From: k6se at juno.com Sent: Tuesday, June 04, 2002 10:41 AM To: DolfinDon at msn.com; elecraft at mailman.qth.net Subject: Re: [Elecraft] S-Meter Sensitivity Hi, Don, You wrote: "I just checked four K2 radios I have in my lab and the CAL S HI needs to be reduced to around 10 to 12 to get a reading of S9 with the corrected setup above. This means the S-meter does read a little low on these radios with the manual calibration." -- snip -- ++++++++++++++++++++ From: "Don Brown" To: "Elecraft" Subject: Re: [Elecraft] S-Meter Sensitivity Date: Tue, 4 Jun 2002 13:02:26 -0500 Preamp and attenuator is off as stated in the earlier post. Don ----- Original Message ----- From: Bill Coleman Sent: Tuesday, June 04, 2002 12:56 PM To: Don Brown Subject: Re: [Elecraft] S-Meter Sensitivity On 6/4/02 1:28 PM, Don Brown at DolfinDon at msn.com wrote: >K2 #4 > >S+40 54000uV >S+20 5400uV >S9 540uV >S5 65uV >S1 .82uV > >As you can see the k2's are off by a factor of about 10 that's 20 db... Pre-amp on or off? I'd suggest you calibrate it with the pre-amp off. Bill Coleman, AA4LR, PP-ASEL Mail: aa4lr at arrl.net Quote: "Not within a thousand years will man ever fly!" -- Wilbur Wright, 1901 +++++++++++++++++++ To: DolfinDon at msn.com, elecraft at mailman.qth.net Date: Tue, 4 Jun 2002 11:21:23 -0700 Subject: Re: [Elecraft] S-Meter Sensitivity From: k6se at juno.com Don, KD5NDB wrote: "OK here are the numbers with the K2 calibrated per the manual and a S-9 indication K2 #1 550uV K2 #2 370uV K2 #3 380uV K2 #4 540uV K2 #4 S+40 54000uV S+20 5400uV S9 540uV S5 65uV S1 .82uV " ========== Assuming this is with the preamp on, the K2's S-meter is indeed stingy. Above S9, K2 #4 is quite accurate (as far as the 20 dB steps above S9 go). Below S9, it is obviously not linear. In the ARRL review on the K2, they said that on 14 MHz the K2's meter would read S9 with only 19 uV with the preamp on and 115 uV with the preamp off. Apparently all K2s are not created equal. My K2s meter seems to be close to the Collins standard of S9 = 100 uV I know S-meters are only a relative thing, but it would be nice if all manufacturers would try to adhere to a set standard. Kenwood uses a neat standard (or used to) where S9 = 100 uV and S1 = 2.5 uV, with each S point equal to 4 dB. Other manufacturers prefer to make the S-meter more generous and use the 50 uV standard, apparently because some hams may be naive enough to believe that the higher an S-meter reads, the better the receiver. Kudos to Elecraft for having an S-meter on the stingy side rather than trying to dupe the ham public. 73, de Earl, K6SE 67 countries with K2 #2622 and counting..... +++++++++++++++++++ From: "Dave Johnson" To: Cc: Subject: Re: [Elecraft] S-Meter Sensitivity Date: Tue, 4 Jun 2002 20:42:50 +0100 >I just checked four K2 radios I have in my lab and the CAL S HI needs to >be reduced to around 10 to 12 to get a reading of S9 with the corrected >setup above. This means the S-meter does read a little low on these radios >with the manual calibration. I've just checked my CAL S HI and it's showing 016 for S9 = 50 uV on my Marconi signal generator. The original setting (as per the Elecraft manual) left the S-meter rather on the mean side. Dave, G4AON K2 #1892 ++++++++++++++++ From: "Don Brown" To: "Elecraft" Subject: Re: [Elecraft] S-Meter Sensitivity Date: Tue, 4 Jun 2002 15:02:44 -0500 Hi I Know four K2's is a small sample but it does appear the CAL S HI needs = to be around 10 to 16 rather than around 20 to 22 when done by the manual= method. If you don't have the equipment to set it up, it looks like you = can lower the CAL S HI 5 to 8 points. You must set CAL S LO first by the = manual then try several settings of CAL S HI until it is to your liking. = If you do have access to the equipment then it is simple to set it up rig= ht. =20 Don Brown KD5NDB +++++++++++++++++++ From: "Don Brown" To: "Elecraft" Subject: Re: [Elecraft] S-Meter Sensitivity Date: Tue, 4 Jun 2002 23:03:23 -0500 Hi The S-meter sensitivity has nothing to do with the gain of the radio. The= S-meter reads the AGC voltage and displays it on the bar graph. The gain= and offset of the of the S-meter amplifiers is adjustable and that is wh= at you are setting when calibrating the S-meter. If you want to set it fo= r 50 uv or 100 uv at S9 with or without the preamp or any where else it i= s up to you. As has been said in many posts the S-meter is quite subjecti= ve and open to a lot of different interpretations. Even the major manufac= turers can't agree on a standard. A lot of K2 builders have complained ab= out the S-meter being stingy so I decided to make some measurements becau= se I just happened to have 4 K2's in my lab. I posted the results of my t= esting. Based on the measurements if you want to set the K2 up to read S9= at 50 uv the CAL S HI needs to be set 5 or 6 points lower than the manua= l setting. Your K2 may be different but if you feel the S-meter is stingy= then setting CAL S HI a little lower will make it better. BTY I checked = the S-meter calibrated at 50 uv and S9 and it is accurate and reads the s= ame on all of the other bands. This surprised me. I don't think there is = anything wrong with your K2. Set the S-meter to your liking. This is why = the S-meter is not a very good indication of signal strength when giving = reports. Seems the S-meter is about as controversial as baluns and 450 ohm ladder = line ;-) Don Brown KD5ND ++++++++++++++++ To: DolfinDon at msn.com, elecraft at mailman.qth.net Date: Wed, 5 Jun 2002 02:17:08 -0700 Subject: Re: [Elecraft] S-Meter Sensitivity From: k6se at juno.com Don, KD5NDB wrote: "The S-meter sensitivity has nothing to do with the gain of the radio." ========== I believe any bozo realizes that. BUT, the gain of the radio DOES affect the S-meter reading. If ARRL says that it takes only 19 uV into the K2 with its preamp on to make the meter read S9 (and 115 uV with the preamp off), and you tested those four K2's with the preamp turned OFF which looks like they might be close to the K2 that ARRL tested. I estimate that my K2 requires about 100 uV to read S9 with the preamp ON, so the logical conclusion is that there's a loss of about 14 or 15 dB of gain in the receive chain somewhere before the IF amplifier U12 (where the AGC voltage which drives the S-meter is developed). 73, de Earl, K6SE 67 countries with K2 #2622 and counting..... +++++++++++++++++++ From: "Don Brown" To: "Elecraft" Subject: Re: [Elecraft] S-Meter Sensitivity Date: Wed, 5 Jun 2002 07:38:12 -0500 Hi Yes you are correct, but the S-meter reading of the AGC voltage is modifi= ed in gain and offset in software by the MPU. Maybe a better piece of inf= ormation is the actual AGC voltage at the output of U2A with 50 and 100 u= V in with the preamp on and off. I could make this measurement if you thi= nk it would help. If the readings are about the same as your K2 then you = may just need to adjust the S-meter gain to get S9 at 100 uV or 50 uv you= r choice. Don Brown +++++++++++++++++++ To: DolfinDon at msn.com, elecraft at mailman.qth.net Date: Wed, 5 Jun 2002 08:01:11 -0700 Subject: Re: [Elecraft] S-Meter Sensitivity From: k6se at juno.com K6SE wrote: "I estimate that my K2 requires about 100 uV to read S9 with the preamp ON, so the logical conclusion is that there's a loss of about 14 or 15 dB of gain in the receive chain somewhere before the IF amplifier U12 (where the AGC voltage which drives the S-meter is developed)." ========== My error here -- The AGC is deleloped on the control board and is fed back to the IF amp at pin 5 of U12 on the RF board. ========== Don, KD5NDB wrote: "Maybe a better piece of information is the actual AGC voltage at the output of U2A with 50 and 100 uV in with the preamp on and off. I could make this measurement if you think it would help." ---------- Please do. It may be that my receiver gain is okay and the problem is in the AGC circuitry instead. Measurements at pins 1, 3, 6 and 7 of U2 on the control board with 0 and 100 uV would be appreciated. ========== "If the readings are about the same as your K2 then you may just need to adjust the S-meter gain to get S9 at 100 uV or 50 uv your choice." ---------- I'm not at all interested in setting my S-meter to one or the other of the "standrards". It's just that my K2's S-meter does not even closely agree with the run-of-the-mill K2, and this tells me there's a problem somewhere that I'd like to find and cure. Tnx, 73, de Earl, K6SE 68 countries with K2 #2622 and counting..... ++++++++++++++++++ From: "Don Brown" To: , "Elecraft" Subject: Re: [Elecraft] S-Meter Sensitivity Date: Wed, 5 Jun 2002 15:00:23 -0500 Hi Here are the readings I have on U2 =20 100 uV in Pin 1 4.28v pin 3 4.28v pin 7 3.92 pin 6 3.92 0 uV in Pin 1 3.96 pin 3 3.96 pin 7 3.92 pin 6 3.92 U2B is AC coupled so I would not expect to see a change in DC voltages. T= his is a 150 khz signal Hope this helps Don Brown KD5NDB ----- Original Message ----- From: k6se at juno.com Sent: Wednesday, June 05, 2002 10:14 AM To: DolfinDon at msn.com; elecraft at mailman.qth.net Subject: Re: [Elecraft] S-Meter Sensitivity K6SE wrote: "I estimate that my K2 requires about 100 uV to read S9 with the preamp ON, so the logical conclusion is that there's a loss of about 14 or 15 dB of gain in the receive chain somewhere before the IF amplifier U12 (where the AGC voltage which drives the S-meter is developed)." =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D My error here -- The AGC is deleloped on the control board and is fed back to the IF amp at pin 5 of U12 on the RF board. =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D Don, KD5NDB wrote: "Maybe a better piece of information is the actual AGC voltage at the output of U2A with 50 and 100 uV in with the preamp on and off. I could make this measurement if you think it would help." ---------- Please do. It may be that my receiver gain is okay and the problem is in the AGC circuitry instead. Measurements at pins 1, 3, 6 and 7 of U2 on the control board with 0 and 100 uV would be appreciated. +++++++++++++++++++