+++++++++++++++++++++ See also KAF2_Notes +++++++++++++++++++++ Date: Sat, 15 Sep 2001 23:30:55 -0700 From: "N7SG K7FD" Subject: [Elecraft] KAF2 RTC - Is this normal? When I tap display to look at the time, I get an incorrect reading...then one second later, the correct time is shown. This only happens if I haven't looked at the time for about an hour or so. If I immediately go back and check the time, it's OK. If I wait awhile and go back in to check it, it reads goofy for about a second, then it's OK. Maybe it's getting its bearings or something... ...wonder if anyone else has noticed this erroneous readout periodically when they first tap display and look at the time... I've been playing w/ the AFIL/AF2 BPF tonight during the Washington Salmon Run and it really is super duper. It definitely has a 'sweet spot' based upon your chosen sidetone frequency; the received signal really peaks up nicely when you're zerobeat... 73 John K7FD +++++++++++++++ Date: Sun, 16 Sep 2001 08:13:59 -0400 (EDT) From: "John L. Sielke" Subject: RE: [Elecraft] KAF2 RTC - Is this normal? Yes, mine does exactly the same. No big deal, but for a second I thought my other clock was wrong! John W2AGN On 16-Sep-2001 N7SG K7FD wrote: > This only happens if I haven't looked at the time for about an hour or so. > If I immediately go back and check the time, it's OK. If I wait awhile and > go back in to check it, it reads goofy for about a second, then it's OK. > Maybe it's getting its bearings or something... > ++++++++++++++++++ Date: Sun, 16 Sep 2001 11:19:39 -0400 From: "Tony Fegan VE3QF" Subject: RE: [Elecraft] KAF2 RTC - Is this normal? Hi John and group, Mine works the same way. Looks like it is corrected once the first update takes place. I have also noted that my clock seems to be sampled roughly every 1.1 seconds. This results in approximately every ninth second being bypassed. The clock in itself keeps excellent time and is well within 1 second per day so I won't have to do any fiddling with the capacitors. CPU time is valuable and I am sure that these very minor quirks are quite acceptable. Needless to say, the AF filter is excellent on CW and it also adds a little more quality to SSB signal reception. 73 Tony Fegan VE3QF K2 # 490 +++++++++++++++++++ Date: Sun, 16 Sep 2001 11:47:01 -0700 From: "N7SG K7FD" Subject: RE: [Elecraft] KAF2 RTC - Is this normal? Tony, right you are...the cw filter is tops...I've been listening to it all day here on 20m with no fatigue. It really performs... 73 John K7FD >From: "Tony Fegan VE3QF" > Needless to say, the AF filter is excellent on CW and it also adds a >little >more quality to SSB signal reception. +++++++++++++++++++++ Date: Mon, 24 Sep 2001 18:19:04 -0700 From: Gary L Surrency Subject: [Elecraft] RTC runs fast That sounds like the 32kHz clock oscillator is running much faster than 32kHz. You may have received a 100kHz or higher freq. C-2 package crystal. It should say "S093" on the side of X1. If possible, use a scope, counter, or some other means to check the freq. at X1 or U2 pins 2 or 3. We will send you a new 32kHz crystal if this is the case. Gary Surrency AB7MY Elecraft Technical Support support at elecraft.com (please do not send support issues to gsurrency at juno.com) From: Brian Wingert Date: Mon, 24 Sep 2001 08:40:15 -0700 Subject: [Elecraft] AF works great - RTC doesn't - help appreciated. Message-ID: Finished the AF/RTC this morning before work. The audio filter works FB. "AF2" is simply amazing. It was easy to adjust without any ringing. Alas, I'm having problems with the RTC clock: 1 - It seems to run very fast (4x) - approximately 2 hours elasped in only 30 minutes, and 2 - Can't set minutes/hours. I haven't done any trouble shooting yet. The date set OK. After setting the date, the clock started. Pushing the band +/- buttons does not adjust the hours or minutes or reset the seconds to zero. I'm expect to replace the timing capacitor tonight and check or resolder the IC and crystal solder pads. Any other suggestions? Please e-mail me directly. Thanks. - Brian Wingert N7RVD +++++++++++++++ Date: Tue, 25 Sep 2001 07:23:02 -0700 From: Brian Wingert Subject: [Elecraft] Pilot error (ooops). AF and RTC both work flawlessly! The RTC works as specified. I re-read the manual and discovered the error of my ways. Thanks for your helpful comments. - Brian N7RVD +++++++++++++++++ Date: Sat, 06 Oct 2001 16:14:35 -0400 From: "Philip L. Carter" Subject: Re: [Elecraft] Clock fucntion with KAF2 Hi Don: It it CRITICAL that you follow the diections exactly. I had a similar problem the first time. Read the directions, and then read them again. If you do anything after you enter the clock mode that isn't "set the year" nothing else will work. Power off and then wait about 10 seconds and try again. You cannot set anything if you exit the clock display mode and then reenter that mode. A little bother, but I found that I was trying to do something else in the process, and the firmware just won't let you. Maybe there is a better way. But for now it's a bit tricky. Phil #370 ++++++++++++++++++++ Date: Sat, 06 Oct 2001 15:26:39 -0500 From: Dave Gingrich Subject: Re: [Elecraft] Clock fucntion with KAF2 At 11:36 10/6/2001 -0700, you wrote: >I just completed my KAF2 and everything works OK except the clock >function. I cannot advance the date or clock. > >I do not have a voltage at pin 1 of U3. Everything else checks out >including U3 orientation, cap C19 and XTAL X1. I turned on the RTC >function from the menu and can switch to clock mode but thats all. > >Any ideas out trhere for those that might have had the same problem? >PIC programming? Bad U3? I assume you get the clock display okay? It may be working fine. The clock must be set in exactly the steps indicated in the instructions. If I recall correctly, you must power cycle the radio and then set the date first or none of the other sets will work. It was specifically designed that way to avoid accidental sets. Took me a few times to figure it out too. ===================== Dave Gingrich, K9DC ++++++++++++++++++++ Date: Sun, 07 Oct 2001 18:35:51 -0700 From: Wayne Burdick Subject: Re: [Elecraft] Couple of wonderments Hi Steve, -- snip -- > And - RTC seems to gain about 2 seconds per day. Fix, if any? Make sure you have the 33 pF capacitor installed on the bottom of the board as specified in the KAF2 instructions. The clock will run too fast if this cap is missing. If the cap is in place and you're still gaining 2 seconds per day, you can go to a larger cap. I'd try 47 pF, for starters. The RTC chip's accuracy is specified at +/- 1.7 seconds per day, so you're right on the edge. 73, Wayne N6KR ++++++++++++++++++++ Date: Mon, 08 Oct 2001 07:47:20 -0400 From: "Mark J. Dulcey" Subject: Re: [Elecraft] Setting the KAF2 Clock T.J. \"SKIP\" Arey N2EI wrote: > Okay I'm probably not the first person to ask this but once you have > initially set the clock, how can to re-set it??? Do you need to remove > the battery??? Am I missing something in the directions??? I follow the > same procedures as I did to set the clock up initially but when I go to > try to change the clock the Band+ and Band - buttons don;t seem to take > any action. Otherwise the whole new board is a great addition. 1. Turn off the rig for at least a few minutes. 2. Turn rig back on. 3. Switch to date display. 4. Change the date. (You can change it, and then change it right back to the original value. Just so you actually change something.) This step is important - you can't change the time unless you change the date first. (They did that on purpose to make it hard to change the time by accident, as explained in the KAF2 manual.) 5. Switch back to time display. 6. Change the time. 7. To make sure you don't make any more accidental changes, switch to date display again, then back to time. The K2 will now not let you change the time or date until you power-cycle again. +++++++++++++++++ Date: Tue, 09 Oct 2001 20:49:34 -0600 From: Larry East Subject: [Elecraft] Praise for the KAF2 I assembled the KAF2 AF Filter / Real Time Clock board last week. It went together quickly and worked on the first try! -- snip -- The RT Clock is also a very nice addition. Just for kicks, I replaced C19 with a 50pF trimmer cap (mounted so that it can be accessed from the top of the rig when the KAF2 board is installed) so that I can "tweak" the clock frequency. After a couple of days of carefully adjusting the trimmer, looks like the clock is now stable to within plus or minus one second per day. (Measured over a two day period -- it'll be interesting to see where it is after a week or so. :-) The KAF2 is best addition to my K2 since the automatic ATU! 72/73, Larry W1HUE/7 ++++++++++++++ Date: Wed, 10 Oct 2001 13:24:14 EDT From: WA9GQT at aol.com Subject: [Elecraft] KAF2 Clock Speed Hello to All! Just tried slowing KAF2 clock down by putting a 10 pF cap in parallel with C19 thus giving a total value of 43 pF. This did slow the clock down & now is within the spec's. ++++++++++++++++++++ Date: Mon, 15 Oct 2001 11:07:20 -0400 From: Gilbert J Cross Subject: [Elecraft] KAF2 RTC I know I should be doing more important things , but I have been adding capacitance to C19 ( 33pF ) and now have 51 pF total. It now seems to be keeping very good time , however I have just now noticed that the time tone falls on either 58 or 59 sec.( 2 sec slow to begin with ). This is 72 hours later. Then in more watching I noticed that the clock was occasionally skipping the next sec coming up and going to the next. Wonder if the 51 puff is causing the oscillator to run a bit erratic? Gil K8EAG ++++++++++++++++++ Date: Wed, 17 Oct 2001 18:31:36 -0400 From: "David A. Belsley" Subject: Re: [Elecraft] RTC losing time slowly Jerry: If you're losing time, decrease the value of C19 on the KAF2 board somewhat. One minute, however, in two weeks seems excessive. Double check that figure; the loss ought be no more than 1/2 minute in that time. If the loss seems stable, you might try decreasing C19 to 27pf or even 20pf. Keep good records and proceed by interpolation or extrapolation. best wishes, dave belsley - --On Wednesday, October 17, 2001 7:01 PM +0100 Jerry Bliss wrote: > After installing the AF module (which includes Real Time Close - RTC), I > find that my RTC is losing several seconds per day. Over a period of a > couple of weeks, I find that I am about a minute behind WWV. > Resyncing is somewhat tedious so I just don't bother doing it very often. > > Anyone else observing this or do I have a defective component? > > Really do like the Audio Filter mod. > > Jerry/k6iii > San Jose, CA > k2 #500 ++++++++++++++++++++++ Date: Wed, 17 Oct 2001 15:37:38 -0700 From: Wayne Burdick Subject: Re: [Elecraft] RTC losing time slowly The crystal used for the RTC chip is specified at rougly +/- 1.7 seconds per day, based on the 33-pF cap we supplied for C19 (soldered on the bottom of the board). This is adequate for most applications. If you need better accuracy: If the clock is running fast, increase the size of C19; if it's slow, decrease it. You can optimize the value of C19 iteratively by keeping track of the error over a long period of time, or you can use a frequency counter as described in the KAF2 manual. 73, Wayne N6KR Ed Juge/NM wrote: > > Mine is about 25 seconds off in a month. Not bad, really. > > Ed, W5EJ ++++++++++++++++++++++ Date: Wed, 24 Oct 2001 09:38:24 -0600 From: Larry East Subject: [Elecraft] KAF2 Clock Now that most of use are accustomed to $25 digital watches that are accurate to within a few tens of seconds a month, it might be a little disconcerting to some that the KAF2 clock can gain or lose almost one minute per month and still be within spec (+/- 1.7 sec./day). However, the KAF2 clock is capable of doing much better -- it just needs to be "tweaked" by changing the value of C19. To make adjustment easier, I replaced C19 with a 50pF trimmer in parallel with a 10pF NPO cap. The trimmer is accessable from the top of the K2 when the KAF2 is installed. It has been 6 days since I last "tweaked" the trimmer and the KAF2's clock has remained within 0.5 sec. with WWV. Just thought some fellow nit-pickers might like to know... :-) 72/73, Larry W1HUE/7 ++++++++++++++++++++++ Date: Wed, 24 Oct 2001 13:53:12 -0400 From: "David A. Belsley" Subject: Re: [Elecraft] KAF2 Clock I'm in the middle of these clock-precision experiments with s/n 744. I started with the stock 33pf. This was about 1.3sec/day fast. I increased to a 39pf for about .8s/d. 47pf gave about .5s/d. A rough projection indicated something around 54pf for 0s/d. I now have a small trimmer (2-50pf) in parallel with the original 33pf set to give about 53.8pf, but I only have one day's reading (and that looks quite good -- although the display jog makes precise reading difficult). Once I get a decent general value, I'll move to a smaller trimmer (maybe 3-10pf) in parallel with an appropriate (maybe 47pf) fixed cap to allow easier tweeking. I recall someone else indicating good results with 52pf. I'm not sure what the variation in these clocks is, but we now have a couple of postings that indicate something a bit above 50pf is preferable to the stock 33pf. I'd be interested in other's results. best wishes, dave belsley, w1euy - --On Wednesday, October 24, 2001 9:38 AM -0600 Larry East wrote: > Now that most of use are accustomed to $25 digital watches that are > accurate to within a few tens of seconds a month, it might be a little --snip-- - ---------------------------- David A. Belsley Professor of Economics +++++++++++++++++++++ Date: Thu, 01 Nov 2001 07:48:53 -0800 From: Wayne Burdick Subject: [Elecraft] Re: RTC discussion > > I have an Oregon Scientific radio controlled clock, a PC and a watch, so I > > don't really need the RTC either... > > There's an old Chinese proverb that says something like, "Man with more than > one clock never knows what time it is!" Just to add my two cents: the KAF2's RTC is not intended to replace clocks at home. It lets you *leave* them there when you take the K2 on the road! 73, Wayne N6KR ++++++++++++++++++++ Date: Thu, 01 Nov 2001 07:53:25 -0800 From: Wayne Burdick Subject: [Elecraft] Re: KAF2 without RTC Hi Ingo, You can leave the battery out and/or turn off the RTC in the menu. The clock's primary function is to eliminate the need for an external clock when travelling. So you might want to set the clock and leave the battery in, then turn it off in the menu when you're at home. Even if you have the clock turned on, however, you don't need to use it. Tapping DISPLAY twice skips the clock display and goes to the voltage/current display, as before. 73, Wayne N6KR ++++++++++++++++++++ Date: Sat, 10 Nov 2001 11:03:10 -0500 From: "David A. Belsley" Subject: Re: [Elecraft] KAF2 RTC Accuracy Frank: I have finally been able to tame my RTC and it's taken about 64pf to do the job. I started out with the 33pf (32.26pf) stock cap, and the clock ran about 1 to 1.2 sec/day fast. I'm not exactly sure of this speed because I had not then developed an accurate way to read the clock's delayed display refresh. In several steps, I determined the clock rate was responding to capacitance at a rate of 31.74pf/second/day, and the relation seemed quite linear. This projected to 64pf to bring the clock to proper speed, and this seems to be very close to the case. It has gained a 'just-perceptible-fraction' (the ol' jpf) of one second over one week. I think it is still running fast -- but, to attest to how good things currently seem, I will have to wait perhaps another week to see for sure. I got a 3-50pf mini padder from Mouser to use for this job; it is no bigger than the original 33pf ceramic. My initial projections suggested the 50pf ought be enough, but I was soon shown wrong there. So I ended up simply putting that padder in parallel with a 47pf ceramic. That combination is now at 64pf. You seem to be starting from a faster speed, so your finding that a 75pf capacitor doesn't do the job is not out of line. If you really were 2 seconds/day fast with the original 33pf capacitor, and your clock chip responds to capacitance the same as mine, you might need as much as 90-100pf to calm things down. Note, by the way, that the inherent capacitance between pins 2 and 4 (ground) of U3 is (at least for mine) 23.4pf. Thus, whatever capacitance you are tacking between these pins is going on top of that. best wishes, dave belsley, w1euy (and best wishes particularly to K1JD and N2JZ who, I am sure, are both extremely amused by this posting). - --On Friday, November 9, 2001 5:46 PM -0500 W4FMS at aol.com wrote: > Hi Gang: > > I've been trying to slow down the RTC in my KAF2. With the stock 33pf > cap, the RTC started out at least 2sec/day fast (which might be to spec) > but I found this unacceptable. Knowing it was off, I would be adjusting > the time monthly. Since I didn't have a small variable cap to fit > behind the board with C19, I've been using fixed value caps to adjust > the clock speed. > > I'm up to 75pf and the clock still seems to being gaining almost > 1second/day. This is the highest value that I've heard and I'm suprised > it still appears to need more. For the others on the list that have > done this experiment, does this seem correct or am I overlooking > something? Will I eventually hit a value where the clock will run too > slow? I'm wondering if my clock xtal is way out of tolerance? Any info > appreciated. > > Frank, W4FMS ++++++++++++++++++++ Date: Wed, 21 Nov 2001 05:21:41 -0600 From: ken-k5id at juno.com Subject: [Elecraft] Real Time Clock Correction I recently decided to make correction to the clock speed on my K2 (it was gaining about two seconds per day). It took 56 pF in parallel with the existing 33 pF to tame it. It is now within about a half second per month, based on the first month's results. Your results may vary.... Regards, Ken, K5I ++++++++++++++++++++ Date: Thu, 13 Dec 2001 13:12:59 -0500 From: Mychael Morohovich Subject: [Elecraft] RE:Odd KAF2 RTC behavior. David- I believe the disparity you are seeing comes from the fact that the RTC does not poll the MCU each and every second. The exact figure for this escapes me, but I believe it can be found somewhere in the manual. Consequently, what you are seeing represents normal behavior for the RTC. Something worth noting here is that when I first built my KAF2 module in late August as past of the FT program, my RTC lost two seconds per 24 hour period. Since that time I have left it entirely alone, never adjusting it even once, and at present it reads 40 seconds fast. Curious, I quizzed Ron (AC7AC) about this, and was told that electronic components such as crystals and capacitors age during the first few hundred or so hours of initial use. Another factor he mentioned, that probably is the main contributor to this inconsistent behavior, is that "the oscillator is not temperature-compensated, according to Elecraft, so changes in room temperature will cause small changes in the oscillation frequency of the RTC circuit." So, for what it is worth, you might consider allowing some time and use before fiddling much with any cap swapping. Hope that this was of some help. Mychael AA3WF K2 # 1025 ++++++++++++++++++