Phil,
There seem to be a number of variables at play here that make apples to apples comparisons difficult. Your numbers do sound low, but using TAS will make a difference and altitude will make a difference in both fuel flow and TAS. Another issue is statute versus nautical...some are using NM, and some are using SM...and the SMPG number will always look bigger, just as MPH looks faster than KTS. But some here have given NMPG (and Pete, man can you lean!!) Just don't forget to convert if you are comparing to a MPH result.
First TAS:
For your scenario, 135 KIAS at 4K is 143 KTAS (at ISA), so your up to almost 18 nmpg off the bat...still low, right? If its ISA +20 (summer day), your KTAS is 148, and you're over 18.5...still low.
Next KTS vs MPH:
Converting the above results to MPH/SMPG, you get:
ISA = TAS 164 MPH = 20.5 SMPG
ISA +20 = 170 MPH = 21.3 SMPG
Still too low, right?
Now start climbing.
In planning for Airventure Cup, several of us used a loss of speed (due to loss of HP in the climb) of 1 knot per 1000' of climb (I think that is the answer to the question you were asking). I think that is conservative, as I don't think I actually lose quite that much. Granted I'm doing max speed tests, and am pushing a lot of air with my hershey bars, but my testing and racing have shown an average of 214 KTAS at SL down to about 205 KTAS at 17,500 (9 knots per 17K feet, or a little over .5 knot per 1K. Those are many different flights on different days, and temps varied, so its way into ball-parking...so I stick with 1 kt per 1K' for estimates.
My testing on fuel flow versus altitude is also WOT/2700 RPM, leaning for 100 ROP (mostly up high...I lean gingerly at high power at low alt...if at all...I'm running hard in these tests). Not sure if it would relate to LOP ops closely or not, but what I've found is that my FF at those settings drops 1 gph per 2,000 feet of climb, or .5 gph per 1000'. Its pretty linear too, as far as I can tell. My guess is that the change per 1000' at LOP ops would be somewhat less.
Say you climbed from 4K to 8K. If you leaned to LOP at each altitude, the difference may or may not be 2 gph (my gradient above). But if it was close, and you were running 6 gph at 8K, but you lost 2 kts due to the climb, that 135 KIAS would be 151 KTAS (ISA), and your NMPG would be a tad over 25 NMPG...getting better!!
OK, there are a LOT of assumptions, rules of thumb and SWAGs in there, and testing is really the only way to see for sure. The more X-Cs you take as you stretch your legs, the more you will be able to sit in cruise and play with this.
It would be interesting to hear from you as you record results, and to hear from others on airspeed loss gradients in climbs and fuel flow decrease gradients in climbs. I think Web Willmott did some work on this a while back, but haven't heard from him lately...Web, you out there?
Finally, on the discussion of best range speed. Take a look at this article on
"Maximum Endurance, Maximum Range, and Optimum Cruise Speeds" by an RV-6 pilot. The gist is that:
Max Range occurs at minimum drag (L/D max = Best Glide Speed)
Max Endurance occurs at minimum HP required (Best Glide Speed / 1.316)
Optimum Cruise Speed occurs at minimum fuel flow per knot (Best Glide Speed X 1.316)
The article goes into the math, shows the graphs and discusses it in much greater detail, and calls the Optimum Cruise Speed "Carson's Speed"...never heard that before tonight (in this thread and in the article).
For me Max Range would be 87 KTS, Max Endurance would be 66 KTS, and Optimum Cruise Speed would be 114 KTS. Not gonna do that, though..."Speed Costs, How Fast Do You Want To Go" is a good axiom, but so is "Time is Money"...I want to get there!
If you have to stay airborne to let that fog lift, fly Max End (or use the RV speed to fly to a VFR airport!). If you want to stretch that flight to the limts of range to make the island, fly Max Range (a little faster in a headwind, a little slower in a tailwind...another rule of thumb...there is another article showing that and comparing Max R and Max E
here).
Good luck in your testing...have fun!!
Whoa...that was longer than I intended...hope it adds value!
Cheers,
Bob