It can be any way to get a bearing. Carlisle RWY25 NDB/DME uses an NDB to define the entry and exits from the arc.
of course the tolerances are different but it is just an arc. Any decent roll steering autopilot can do it easily as can a hand flown pilot assuming they have a FD (and probably without). And whether it is an arc to an aid or a GPS waypoint is irrelevant to a modern system.
On the speeds, a GA aircraft will be flying it far slower than the design speeds for a light jet or bigger. Hence bank angles should not be a problem.
All very true, but only a relatively small percentage of GA aircraft are equipped with roll steering and or a FD or a modern system. I was just pointing out that the cues are different for guidance on a DME arc and an RF turn and that accuracy requirements are much more demanding for the RF turn, not that they can’t both be flown. Heck, we still have to bring Peter’s equipment into the 21st Century. 
It can be any way to get a bearing. Carlisle RWY25 NDB/DME uses an NDB to define the entry and exits from the arc.
As I understand it, TERPS in the US only permits a VOR. It states:
DME arc courses must be predicated on DME collocated with a facility providing omnidirectional course information.
It’s funny how the UK prohibits really useful stuff like LPV approaches which are offset, or which have a higher minima than a nonprecision approach to the same runway, but it does give you a priceless gem like this
(a DME arc around an NDB – hey how’s that for precision navigation)
I would have thought that anybody having to fly a “DME” arc depicted on his GPS (I can’t see them because the KLN94 doesn’t have any arcs in the database) would just fly it in HDG mode, following the magenta line with the heading bug, all the way to the localiser 
Then PANS-OPS must have it differently. To second JasonC; the two DME-arcs I have flown (the only ones I know about in Denmark) have nothing collocated with the DME:
The EKSB Rwy32 ILS-1 uses a VOR placed 7-8 NM from the DME, and a Locator on the centerline, 2-3 NM from threshold.
The EKSD Rwy26 ILS/DME uses a Locator placed 3-4 NM after the departure end of the runway. There is no VOR.
Here, it is a rarely used procedure with a bit of mystique around it. I have seen pilots use elaborate ways or bad excuses to avoid flying them. But the only problem with them is probably lack of familiarity and currency. They are not that difficult.
a DME arc around an NDB – hey how’s that for precision navigation
Actually, it’s a very convenient combination for those who don’t have an RMI with a VHF bearing capability: you enter the arc and steer to keep the ADF needle horizontal, checking your DME from time to time just in case.
It is difficult to shake that feeling that we are seeing with this whole satellite based navigation business in combination with computing power and the ability to create good user interfaces with very intuitive visualization for much better situational awareness something that scares a lot of people who worked hard to master flying on raw data.
Peter said it before nicely and I’m quoting him a second time:
I can’t help thinking there is a lot of job creation in all this, since GPS did to the navigation business what the CD did to the vinyl record business
One of my instructors in the US told me something along the lines of: the FAA was doubtful of GPS for a long time and now they have finally accepted it. That is why everything is now being done with GPS high and low.
Are we seeing another version of “job killer computer” once again? That stuff works and as long as the accuracy of the data can be assured, the same way as all the time with other means before, then where is the problem in flying whatever precision? Maybe the Avidyne person who has responded to my emails regarding the RF legs on R9 put “approved” in quotation marks due to similar thoughts.
What is the reasoning behind lowering approach minima of RNAV approaches in several steps? I fail to see something like a ripening process with that technology? The components of the system hardly become more accurate over time. So why are the CAAs feel compelled to start with a high DA and then lower it step by steps? Maybe the only reason is their own protection? As in ’we’ve not seen an accident in X time so we feel now it is safe’. That is nothing other than predicting the future based on the past, which can psychologically explained – keeps people within their comfort zone – but still doesn’t make a lot of sense.
I probably should have started this in the Hangar Talk section …
I fail to see something like a ripening process with that technology?
Then look closer :) There have been and still are steady improvements.
Remember selective availability? Now the hardware isn’t even installed anymore in Block IIF satellites.
Atomic Clocks on the space vehicles have been steadily getting better.
Satellite transmit power has been increased.
The L5 safety of life signal has been added. Even though it is located in an aviation frequency band, hardly any aviation receiver seems to be able to receive it, sadly.
There are talks about adding an additional frequency band, for even better ionosphere compensation.
The L1C signal is imminent (i.e. should be available this year).
During EGNOS testing, the results actually got less accurate when EGNOS was enabled in the receiver. This has now been improved, the issue apparently was clock accuracy.
