locked #Q65 Latest Performance Results for Q65 with Settings #Q65


Hasan Schiers N0AN
 

Months of testing every morning with KB7IJ, NM3G, WB4HIE, and K5GZR have pretty much resulted in the following conclusions. (Path lengths 600 to 900 miles)

Averaging should be left on all the time now.
Decoding while monitoring works much better than it did.
Decoding while monitoring still works better if the signal you are looking for is inside the F Tol setting. Once you are in QSO, F Tol can be as narrow as you like and your equipment permits (stability)
For schedules, always put the Call and Grid in, before starting the QSO. Don't call CQ, call the person you are skeding with.  This invokes AP decoding immediately.

Reports like q0 indicate AP decoding is not running because you don't have call/grid entered)
Reports like q3 indicate AP decoding is working
Reports like q32 indicate AP and Averaging are working the 2 means 2 sequences average)

Q65 Settings for 15A, 30B, 120E
Always hit F4 key to clear collected junk before starting a performance run
Always hit Clear Avg button before starting a new QSO if evaluating performance
F Tol 50 Hz (can be set lower for schedules)
Decode: Fast, Enable Averaging, Auto-Clear Average After Decode

Notice: at bottom of Decode Screen (next to Current Mode Listing) two numerical boxes. These show skipped frames (number of frames since last decode). This is a VERY good indication of decode efficiency when manipulating power levels) Which number increments or goes to 0 with a decode depends on whether you are on 1st or 2nd ...


Q65 30A (30 sec sequences with Submode A) is still the standard on 50.275 (USA)
Q65 120E (120 sec sequences with Submode E) is on 50.235, and most sensitive for qrp on 6m (not EME)

Latest testing shows: (we have been testing 15A for several weeks every morning for about an hour)

Q65-15A works quite well for schedules. We have been working every morning at 666 miles with the far end at 5-10 watts and me at 20-30 watts (he has noise), and getting about 85% decodes on my end and he is getting about 40 to 50 percent decodes on his. He is using 7 EL @ 50', I am using 5 EL at 60', both fed with hardline. Both are LFAs  

Q65-30B works better than 30A, and very obviously so. (not night and day, but easy to see) 
     1 to 3  watts on the far end produces > 60% decodes 
Q65-30A works marginally, but not dramatically better than Q65-15A

Q65-15A works better than MSK144 during non-shower times (testing from pre-sunrise to about an hour after sunrise) It is clearly superior. We have repeatedly compared how much power and how much time it takes to complete a qso or to get a series of decodes using MSK144 vs Q65-15A.

Q65-15A requires MUCH less power to generate decodes than MSK144...by well over 5 dB. 
    It's message averaging works extremely well.

Q65-15A will complete the QSO in fewer sequences than MSK144
Q65-15A will generate  more sequences with decodes than MSK144 (by that we mean, 15 second sequences where no decode appears)
If there are a LOT (like in a shower), of meteors, then MSK144 is still faster to complete than Q65-15A or 30B

For typical early morning conditions, both Q65-30A and 15A are clearly superior to MSK144 in that they require significantly less power to generate more sequences that produce decodes. 

One might wonder WHY is Q65 better under most conditions than MSK144 on 6m terrestrial?
Here is our theory, but it is speculation.

MSK144 is optimized for meteors. It will respond to other modes of propagation, but for some reason, not as efficiently as Q65. (efficiency defined by low power performance)  Sustained sporadic E is an exception, as it acts like a perfect mirror and MSK144 loves it...but since MSK144 as we use it relies on time division multiplexing, a common frequency becomes unusable very quickly.

Q65 seems to make use of all propagation modes, it does not care, but is particularly effective for scatter modes. It acts like a funnel. All of these propagation modes efficiently contribute to a decode in Q65:

Ionoscatter
Troposcatter
Tropoducting
Meteor Scatter
Sporadic E
Airplane Scatter

In summary: Do Not Run 120E or 30B on 50.275!

120E is magical, but way long and hard on equipment.
30B works nearly as well as 120E
30A is the 'standard' on 275 and is most likely to generate a response to a CQ

15A works surprisingly well. So well, we have stopped running tests on 30B and only rarely test on 120E for special super qrp stuff.
We have been running 30A and 15A on 275 if it is not busy, trying to generate more 15A activity.
We run 120E and 30B on 50.235

Hopefully, this will help people get the most out of Q65 on 6m for terrestrial work.

73, N0AN
Hasan


Don Melcher
 

On Wed, Jun 2, 2021 at 10:05 AM, Hasan Schiers N0AN wrote:120E is magical, but way long and hard on equip
Can you expand on this comment. I haven’t tried the mode yet but I have 6m antennas up. 
 
--
Don
W6CZ/W6ZB
DM07bk


Dwight Jones
 

Equipment tends to overheat with two minute transmissions. It also tends to be rather boring to have QSOs that take that long.

Dwight
Ko6fe
CN71


 

Hi Dwight
in times of CW (what was this?) we had 2,5 minutes TX periods in EME or 5 minutes in high speed CW with full power. A QSO easily lasted up to one our ... Once a 70cm meteorscatter QSO I could complete after 4 hours and I was a very happy guy. But nowadays we live in sort of a "highspeed society" ....
--
---

Christoph Petermann DF9CY

JO54al (Baltic Sea Coast)
7 MHz Dipole @9,5m - 400w

50MHz 6 ele OP-DES Yagi - 270w

144MHz 9 ele DK7ZB - 330w
432MHz 23 ele - 35w

https://www.df9cy.de


Charles Suckling
 

Christoph

The technical issue with longer periods is that unlike CW,  WSJT modes are essentially 100% duty cycle while transmitting, so poorly designed rigs or PAs may overheat with the longer period modes.  Better heatsinks, bigger fans etc should be able to deal with this.  I tested my small 130W 1296 PA with up to 300s periods and it reaches thermal equilibrium after a couple of minutes, and ran cool enough.

Re time to complete a QSO, I agree with you that if longer periods will enable a difficult QSO then who cares how long it takes. I have run 300s periods with KA1GT on 1296 EME  using small antennas (eg a helix).  In these extreme cases having the longer periods available is very useful.

73

Charlie G3WDG

On Fri, 4 Jun 2021 at 06:21, Christoph Petermann DF9CY <mail01@...> wrote:
Hi Dwight
in times of CW (what was this?) we had 2,5 minutes TX periods in EME or 5 minutes in high speed CW with full power. A QSO easily lasted up to one our ... Once a 70cm meteorscatter QSO I could complete after 4 hours and I was a very happy guy. But nowadays we live in sort of a "highspeed society" ....
--
---

Christoph Petermann DF9CY

JO54al (Baltic Sea Coast)
7 MHz Dipole @9,5m - 400w

50MHz 6 ele OP-DES Yagi - 270w

144MHz 9 ele DK7ZB - 330w
432MHz 23 ele - 35w

https://www.df9cy.de





 

Yes Charlie, that is correct. After an EME QSO on 70cm one could literally "feel" the energy that went through the cables at 1.5kW ... they were warm. For testing the system I once transmitted 15 min at full power straight up into the sky. The rig, the transverter and the driver with a 2C39 tube stayed coool and the homemade PA was a monster with a 300 watts blower ... I never ever broke that stuff. Now using SSPAs even with good cooling is more a thing I must have an eye on.
--
---

Christoph Petermann DF9CY

JO54al (Baltic Sea Coast)
7 MHz Dipole @9,5m - 400w

50MHz 6 ele OP-DES Yagi - 270w

144MHz 9 ele DK7ZB - 330w
432MHz 23 ele - 35w

https://www.df9cy.de