DOTs are good - extra space on CW?

Al_NN4ZZ

DOTs are good - extra space on CW? It looks like the missing DOTs is fixed. At first I thought it was still happening. What I am seeing is the insertion of a SPACE occasionally in the middle of a character. It has happened about 5 times in the last 24 hours. For example: V is S T P is E G It happens randomly and in all of the cases (so far) it happens when moving from the DOT paddle to the DASH paddle. I'm using a Begali Sclupture paddle and medium keying speed. Is anyone else seeing this? It could be me but I haven't had this problem in the past but need to see if anyone else is see this also. Regards, Al / NN4ZZ

Ed, K0KC

This sounds like the problem that I identified back in early July just after I received my 6700. I have not operated much CW with the new version (0.15.13), so I have not yet had a chance to determine if it is still a problem. It will be very disappointing if CW performance is still lacking with the release of v. 1.0. I may have a chance to do some testing this weekend.

Al_NN4ZZ

Tim or Steve, Do you have a test setup that can use to simulate sending V's and P's with different delays between the paddle presses? It would be ideal to have a set of relays you could program to simulate the paddles and be able to control the time between releasing the DOT paddle and pressing DASH paddle. With that running continuously and under controlled parameters you could see how the extra space issue is related to the time between paddle presses or whether it is a keyer code issue. (interrupt issue, key de-bounce issue, etc) A reason to suggest using relays (rather than solid state switching) is to see if the contact de-bounce is an issue. Regards, Al

Sergey R5AU

Al, I have the same issue with my Bencher, necessary to assess it additionally

Al_NN4ZZ

Sergey, Agree that it is not a paddle issue, and I don't think it's an operator issue since a few of us are reporting it now. Tim, it looks like a keyer issue. Should this be converted to a TROUBLE report? Regards, Al / NN4ZZ

K2CM

Most of the time the keying is good; but, periodically it messes up a long character such as an l or p. it sometimes takes several tries to get it right.

Carl/K5HK

Noted same problem here on 6500. Liveable but anouying.

Al_NN4ZZ

Tim / W4TME -- did this make it to the issues list as a defect in the developer's bug tracker for further investigation.? Steve / N5AC -- is this related to the "paddle latency" fix you mentioned in another thread? http://community.flexradio.com/flexradio/topics/internal_keyer_latency_paddle_press_to_sidetone It happens frequently and feels like an interrupt issues (i.e. another higher priority process interrupting the code). Regards, Al / NN4ZZ

Brad - N6SPM

I believe I've seen this a couple of times in release 0.15.13. Sending with a paddle using the internal keyer set to 18-20WPM, every once in a while characters like a "7" will be sent as T B. I've noticed no pattern as it has been infrequent. Tried two different paddles (Bencher and K8RA) and they both behave the same. 73 Brad - N6SPM

Al_NN4ZZ

Good to see V1.0 is coming along....From this weeks 20-Sep-2013 Insider, Gerald wrote: As we approach the target release date for SmartSDR™ v1.0 I would like to review the status. Most of the work we have been doing over the last few weeks has been focused on performance and stability enhancements. We are now down to a very, very short list of remaining items to be completed. --------------------------------------------------------------------------------------------------- I sure hope this KEYER Fix is included, it's annoying to use the keyer currently. --------------------------------------------------------------------------------------------------- Regards, Al / NN4ZZ

Brad - N6SPM

Al, Today calling in pileups I had a chance to observe the error you described. My call has a S P sequence and fairly frequently, the P was converted E G. My fist is not great, but I can do my callsign, signal reports and a few salutations reliably at 20WPM. This is definitely an internal keyer anomaly that makes CW a pain to use! I truly hope this fix made the next release that is in the tester's hands now. 73, Brad N6SPM (or N6SEGM depending on keyer)