Peter wrote:
There ought to be a nice retrofit market for LED replacements for those PCBs with the Minitron displays.
I would guess that by now 90% of these old avionics stacks have been replaced with dual GNS430 units or similar. There have been so many changes since the 1970ies (extension of the VHF COM spectrum, FM immunity issues, 8.33, mode S, B-RNAV capability, …) that at some point a drastical measure will have been the cheaper option. Additionally, replacing an old Sperry/ARC/Cessna 800 or 1000 series avionics stack with Garmins will save something like 30kg…
About the Bendix/King display technology I am not so sure. There seem to be seveal kinds. This one here is likely to be a gas discharge display (or “plasma” or “Panaplex”):
Here is a working Panaplex display for comparison (from a clock):
Other King displays are like this. It could be that this one is of the VFD type:
But I am not sure… Since years I collect pocket calculators, mainly pre-1980ies ones with luminescent displays (have in the order of 1000). So I am quite familiar with the display technologies in use then, but this one, I can’t really tell.
Peter wrote:
The fault changed to a missing decimal point, so I fixed that with a bit of stick tape
What happens if the distance goes above 99.9 NM? ;-)
That would be a very good DME!
I have seen this but very rarely. The EGKA DME can be picked up around overhead LFAT, at say 8000ft, just over 100nm. But only just, and you have no way of knowing (without identing it) whether it is that one or another one in say France.
I put the ADF on a central heating radiator for a couple of days and the display is OK now… till the next time.
LED replacements for King luminescent displays
With reference to this I enquired at Navcomm Avionics at EGHH who apparently offer a repair service and this is what I got back:
This is really interesting and obviously far cheaper than installing any alternative.
I am told the LED alternative is so expensive because the display driver has to be replaced, it is also likely the LED will last a long time.
Yes; I can think of several ways to do this, and none of them is trivial. One could replace the display driver (or pick off the signals on its pins and leave the chip in situ) or one could assume the display driver still works and emulate the original display by picking off the signals going to it.
I am not surprised somebody has done this because there are so many thousands of these boxes out there. What does surprise me is the KR87 mod; surely there is no market for this in the USA, so maybe this is a European product where there is still a lot of FTO business that uses the ADF.
I could not find anything on google and wonder how this mod is certified. It’s a really interesting area… modifications to crappy avionics. King sold some 10,000 of their crappy KFC225 servos in the last 15 years, to replace the burnt out ones, at a few k $ each, and an approved mod to prevent the burnouts (or even to convert it into a brushless unit) would be a great moneymaker.
The LEDs should last for ever. Well, centuries. They do decay but it is significant only at very high powers e.g. ones used to drive optical fibres. We make fibre modems at work and got around this by using very short pulses to encode just the transitions.
A few months ago, out of pure curiosity, I tried to evaluate how difficult it would be to replace the plasma displays in old King hardware with LED ones, and it was the KR87 schematics I actually looked at. I was hoping it would be as simple as replacing just the driver circuitry around the display itself, but it turned out to be a lot more than that. In particular, the decoder is an obscure chip designed specifically to drive plasma displays.
Assuming one can get the data sheet for the display driver, one could emulate it (with a fast micro, or an FPGA). I have done that sort of thing in years past.
It would be easier to assume the display driver is working and pick up the signals from the display connector, and then emulate them with an LED display. Probably a micro would be the easiest way to do that, but still a substantial project, and would it be worth it for a KR87, given that your market is limited by
IMHO emulating a whole KDI572 (remote DME indicator) would be more lucrative. Functionally it does very little but costs a fortune.