Location
Our 2m QTH is situated at the hill top in the N/W part of 9A, near town Cakovec. Hill is only 335 m high, but there are no higher hills towards N and NE all the way to OM and YO. Situation is different to other directions, we have mountains relative close to our QTH (in OE and S5, as well as in the middle part of 9A and T9). The worst takeoff is direction S, we have over 1000 m high mountain only 40 km away from our QTH.

Tropo experience from JN86EL is giving a little bit surprising results. Takeoff to N is the best, as we could expect from position described. Contrary to expected, takeoff to N/W is much better as takeoff to N/E. For that is probably responsible ground configuration around QTH, terrain is slowly falling under the constant angle in direction N and N/W. Ground effect is giving us a few (but significant) dB over our antenna gain.

We are using 4 x 17el F9FT antennas (2 x F9FT before 1999) and 800 W output power(about 500 W in the past). In favourized directions our signal is better as signal of some neighbouring stations, located much higher as we are and using more ERP-s.
 
Tropo conditions in JN86EL
On average tropo conditions our limits are JO92, JO82, JO72, and JO62 on the north, JO51, JO40, JO30, JN39 on the N/W, KO01 on the N/E, KN2x on the E, JN70, JN80, JN90 to S (the worst takeoff), JN33 to S/W, and JN47, JN38 to W. It is hard to define "average" tropo conditions, let's say we can work these squares in the 80% of cases during summer / autumn season.

However, QRB is depending on takeoff and gain of the station on another side. QSO with "big guns" or stations with extraordinary takeoff, as well as troposcatter QSO-s are excluded from the list above. There are some stations in JO52, JO42, JO31, JO93, KN32 etc. we can work almost regardless propagations with surprisingly good signals.

On good tropo conditions our working area is much bigger. Good conds direction north, JO73 or even JO74, JO75 are not rare, unfortunately there is not to much activity in that areas. Interesting, it is not valid for JO63 and JO64 squares. Almost same QTF, but conds are "cut off" after JO62. We had more QSO-s with JO65 as JO63 or JO64. Anyway, there are only 2 squares in DL we never reach tropo: JO34 and JO44 (more than 500 DL DOK we have cnfrmd by QSL).

Lot of nice openings we had to N/W, northern part of PA and N/W DL (JO33, JO32). Openings in that direction are more common as openings towards southern part of PA or ON, because it is in the shadow of Alps.

Alps also are reason why we rarely work F stations. QSO-s with F stations near DL border are common, as well stations from the eastern part of Mediterranean coast (JN33), but QSO-s with other parts of France are rare (with exception of "Trans Alps" conditions - more about that later).

Conds to G are rare too. Only about 10% of nice openings from G toward southern DL "jumps" over Alps. Most of the G stations we worked are from JO02. QSO-s with other parts of G are exceptional event.

We are the "shortest" toward south, few times we reach to IT9 and we have only one Italian JM square worked. There is mountain 40 km from us in that direction, and the rest of the path is over high mountains in T9 as well.
 
Ducts
Duct in JN86 is very rare event. Generally, ducting over the continent area is much more rare as ducts over sea surface. Duct to N/W and W almost never reach JN86, it is broken by Alps. Sometimes we can catch some short-living conditions to G when south DL is enjoyng nice propagation to Great Britain. Duct over Mediterranean sea is not rare during summer season, but it usually ends at mountain line along 9A Adriatic coast. In the most if cases we can work EA6 big guns in deep QSB.

We had only 3 strong and long-duration duct openings here in the period of the last 10 years: one to EA3 / EA6 / EA5, one to LY, and one to SM.

For ducts are typical strong signals (sometimes like ES). During duct to LY 26th of February 1998 (who said winter season is not for TR dx-ing, hi ?) we were QRV using 25 W only and 2 x 17 el F9FT (List of stations worked: LY3ED ko14, LY2MW ko24, LY2BH ko25, LY2SA ko14, LY2PU ko14, LY2WR ko24, SP5CCC ko02, hrd: EU5MS - only one stn missed).

Duct to SM in the October 2001 deserves a few words more, click here if you want to read event description.

Our tropo ODX, GM4AFF from IO86ST (1726 km) is not typical duct QSO. There were 2 days beautiful duct from G to SP/LY, and no any sign of good tropo in JN86. At the evening 20th of September 1998, north DL stations were reporting extraordinary signal from GM4AFF (50+40), and I decided to wait hoping that "something" of that signal will be refracted to my side too. Finally, after few hours of waiting at GM4AFF QRG, I got nice 51 signal. Opening was short, but more than enough for full QSO, including QTH loc exchange. After him I heard GD4GNH calling, but signal disappeared before QSO was completed.
 
"Trans Alps" propagation
Contrary to other types of tropo propagations, for QSO-s to the other side of Alp mountains we need low-level pressure field (usually before arriving of frontal system from Atlantic). Using that propagation type we are able to work HB9, F and some stations on south-western part of DL we never heard on "normal" tropo conditions. Such signals are sometimes very strong, but always with deep QSB (Typical is QSO wih F6HVK, JN27LH, 872km). This is almost only way how we can work with JN26, JN27, JN36, JN37, JN46. We also have "Trans Alp" propagations to some other squares (JN38, JN47 etc) we could reach with "normal" tropo too.
 
Troposcatter
Troposcatter QSO-s are not so much propagation dependent ("2m DX any time"), but, more power is required. We were testing troposcatter lot of times with some stations at 900 to 1000 km distance, and result was successful in 90% of tests. (DK1KO JO53CT, PA0PVW JO22VA, and many others). By example, we were using 2 x 17ek F9FT in the past, and tests with DK1KO were unsucessful in 50% of cases. Later we doubled antenna gain, and success was almost every test (DK1KO is also using 4 x 17el F9FT). At the distances over 1000 km percentage of successful QSO-s is reduced. We were testing conds with SK7MW lot of times, success was abt 60% when there was no outstanding tropo conds to the north. Of course, we had lot of random troposcatter QSO-s, but all using high power and high gain antennas. Except power and antenna gain, good takeoff and quiet QTH is required, signals are between S1 to S3 in the most of cases.