molevitch

More strange behaviour under RS485 Network Soooooo, I have a weird thing happening now..... My sketch for the ARK-9 works perfectly under serial_irq as a test. The dials work, the switches work, it all looks good. I change the code to set it as a slave, put the Max487 chip back in, plug the panel into the network board. Reset the board by unplugging the USB, power it down, re-power and replug the USB to the Master. I run DCS, instant action Mi8. Look at the ARK-9. The dials tell the right numbers. I turn the dials and the numbers change.... great! I flip some switches.... nothing. No switches work.:noexpression: I check the Sketch, all seems correct. I don’t understand... Help! Please! Mole.

Dodgy Max487 Hi Hans, Yes, it was a bad Max487 chip... All running now! Mole

Mi-24 pit build in progress I was building a Mi8 pit but once ED informed us that there will be a Mi-24 module I went a bit crazy and decided to switch direction. Having found a few sources for scrapped/decommissioned Russian aviation equipment, my hunt has gone into overdrive. I fly in VR, but like to use real switches and dials. I don’t need working gauges. I build easily replicated panels myself, such as most of the switch panels, and have acquired an ARK-15, a SPU-7, a R-8xx channel dialler, a Mi8/24 cyclic, a M8-8 collective and pedals, a seat and parachute harness from a ‘8, ARK-2 panel, Latitude setting unit, and lots of toggle and push switches.... I have become obsessed and will probably require medical intervention over the coming months. Permanent residency in an institution may be necessary, if they suddenly scrap the project.... The cyclic and collective have magnetic braking systems, and the cyclic has actuator controlled trim via the hat switch. Currently flying the 8 regularly and since its release, I now view it as a training module for the Mi-24. Lots of shared systems or comparable systems. My ARK-15 runs the ARK-9 under DCSBIOS along with the SPU-7 (couldn't find a SPU-8) and the R-8xx radio channel dialler. All the left wall panels/switch sets will go into DCSBIOS also, and the next complex build will be adding the pedals which will have foot switch activated trimming functions. And yes, I am a big show-off....

Thanks Hans, thanks guys, I will try running without DCs, see what output it gives me. The slave number is correct. I checked the Tx Rex pins also I will check pots, but I don’t think that’s the issue. I unplugged the slave mega and the network is running the other 2 boards no problem. So I need to start there. It worked as a serial irq panel, but not as a slave, or at least not well. Maybe a dodgy board. But something caused all 3 boards to slow done or in an unexpected way. Will do some more tests.... Molevich

Hi Outbaxx I will double check, but I don't think its that. My slaves are numbered 126, 125 and 124. I had 2 panels working on Nanos, and it became messed up when I added the third with a Mega. The Mega only has a green and a yellow light glowing... no reds. I wonder what that is telling me.

Hi DCS-BIOS Gurus, So I now have 3 panels all working on the network. Each was tested as a standalone under serial irq, before adjusting the code for slavery. I am running 9v 1a power into the network via vin. Apart from the requisite Mga Master, I am running a Nano for a Radio selector board (SPU-7), another Nano for a Freq Preset Dialler and a Mega which is for the ARK-9/15 for which Hans helped me write the Sketch. The strange result I am getting is... the input response is slow, and varies a lot. For example, I turn a dial and it can take 10 seconds for the dial to change in DCS. Some panels are slower than others. And some switches respond immediately. And then after a period of time the network just sort of..... stops. Any opinions on what is causing this will be happily received. Thanks, Molevich

This thread is about running DCS-BIOS over a RS485/Max487 Network... or am I sadly mistaken?

DCSBIOS Network running, but slow (?) Hi Hans, So I now have 3 panels all working on the network. Each was tested as a standalone under serial irq, before adjusting the code for slavery. I am running 9v 1a power into the network via vin. I am running a Nano for a Radio selector board (SPU-7), another Nano for a Freq Preset Dialler and a Mega which is for the ARK-9/15 for which you helped me write the Sketch. The strange result I am getting is... the input response is slow, and varies a lot. For example, I turn a dial and it can take 10 seconds for the dial to change in DCS. Some panels are slower than others. And some switches respond immediately. And then after a period of time the network just sort of..... stops. Any opinions on what is causing this will be happily received. Thanks, Molevich

Getting started with DCSBIOS Hi Michelip I know how you feel! When I first looked at moving from using Bodnar joystick boards to DCSBIOS, it took me many reads of the documentation to begin to understand. But suddenly it all fell into place. Ian (FSF) has done a fantastic job with DCSBIOS and the most recent iteration has made it even clearer and easier to set up. Start here, http://https://dcs-bios.readthedocs.io/en/latest/installation.html and I am sure you will soon have a single toggle running. After that it gets easier, but also harder as your ambitions increase. Also, all the help for this kind of work is found in the DCSBIOS discussion thread, http://forums.eagle.ru/showthread.php?p=2356932, rather than in this section. Good luck, and happy soldering! Molevich

Hi Ian, Well, I just turned on my pc to take a screenshot.... and by some magic, the plugins link is now there! So I guess I needed a restart. Thanks again for DCS-BIOS! Mole

DCS-BIOS ver 10.xx Hi Ian, hi guys I have just installed version 10, but I do not see the Plugins page link. Am I being dumb? I have reverted to 8.5 because that way I can see the control references for my modules. In 9 and above I see no plugins link. Ian, thanks for your new work on this great bit of software. The new interface is great. Mole

DCS-BIOS ver 10.xx Hi Ian, hi guys I have just installed version 10, but I do not see the Plugins page link. Am I being dumb? I have reverted to 8.5 because that way I can see the control references for my modules. In 9 and above I see no plugins link. Ian, thanks for your new work on this great bit of software. The new interface is great. Mole

Hi Hans, After leaving it all alone for a few months, I came back to the RS485 Network conundrum with fresh eyes. After 1 or 2 hours of new soldering, I am proud to say that now I have one button working on a Nano slave! Hurrah, (rubbing hands together), now the real work begins! Mole

Hover over water If you have limited visual references like landmarks etc then you have to fall back on those pesky instruments on your dashboard. :smilewink:

Thanks Hans, On holiday until Tuesday so will check when I get back. Mole

Hi DCS-BIOS Gurus! I have been fiddling with my pit, now that Summer has started to fade and the days are getting shorter.... I have still failed to get a RS485 network working, and now, even my serial coded modules are causing me strife! I have checked my Sketches and they compile/load without complaint. My ARC-9/15 Sketch which Hansolo helped with does the following now, (it used to work...). I run it, and after 3 seconds it just stalls. I can see from the lines that it is reading correct outputs from the hardware. But it also seems to interact with DCS correctly, just for a very short time. I have tried adding a 12v power supply to the Mega, no difference. The Mega is plugged into a USB extension port, but even when I tried it directly into the PC, same problem. Its like a syncing issue or a timing issue.... Any ideas/solutions gratefully received. This issue may even be contributing to my inability to get the network running. Thanks in advanmce, Mole

Cyclic Gaiter Here is my custom rubber gaiter for the base of my cyclic. My gimbal system was carefully measured and mocked up in plasticard and a piece of aluminium tubing. Then I sculped over that with oil based clay. I used a polyester moulding resin to make the mould and slush moulded the gaiter in black silicone. The shape is based on photos of Mi-24 cyclic gaiters, which have wider set convolutions on the left of the base, so asymetric deliberately. It serves to protect the Hall Sensor for the pitch axis as well as keeping crap out of the gimbal.

Thanks Trip, the feeling is mutual! Currently my build has become an octopus.... There are so many parallel projects within it now. I have had limited time to work on it for a few months as I am redecorating our house. I still have had no success with networking DCS BIOS units. So I have put that aside for when I have time to concentrate exclusively on it. Current project, make a cast silicone gaiter/boot for the cyclic to hide the gimbal and protect the electronics in there. I must post about how I set up the trim hat with linear actuators to make that function. Very neat and compact, and I already use it in the gazelle and the Mi-8 (though it does not really have it, the 24 does). I will post some new pics in a day or two. Hotter than hell here today! Mole

Hey Trip, yes, I think the brakes are very much off-on, but I wait to be corrected. I used a momentary PTB button switch, while Goblin is using a load cell. That button is talking to a Bodnar board. Because I am taking an almost “set-building” approach, I try to keep everything as authentic looking as possible, or even authentic vintage equipment back engineered. So for me, if I can I make bits as small and concealable as possible. M

No need to go as far as a scrapyard. Most lamp-posts or railings have an abandoned bicycle frame attached, no wheels or saddle, just a sad rusting frame and usually with the brake handles still attached! That's where I got my first version of my system....

Handbrake build Hi Trip, Here is my method. https://forums.eagle.ru/attachment.php?attachmentid=213967&stc=1&d=1563538438 I used a Push to Break momentary switch, so the brake is on by default. (Flying Mi-8 mostly but relevant to Ka50 and fixed wing aircraft too). So by pulling the brake lever, the brake is released as the pressure on the PTB switch is lifted. When the lever is let go, the compression spring reapplies pressure to the PTB switch, and the brake is on again. The compression spring and switch housing I attached to the base of the cyclic so it will go inside the boot/gaiter, using jubilee clips. Motorcycle brake lever off ebay, brake cable off ebay, scraps of 1mm/2mm aluminium sheet... off ebay, compression spring...off ebay. I think you get the idea! Need any more info, just ask. M

Shagrat, I totally agree. But of course from the "sim" perspective, one should be also be able to swap who is flying, ideally.

Surely the main issue with multicrew is not switches but controls. Only a very unhelpful co-pilot will flip switches on the pilot's dash/panels unbidden. But the RIO concept works if he is not trying to control the aircraft simultaneously with the pilot. But if both pilots have hands on sticks/cyclics and throttles/collectives, and they are not linked mechanically, and they push and pull in different directions, how does the sim grant authority? And if its a case of "You have it/I have it", what happens when control is swapped and respective sticks are in different positions/trimmings and throttle amounts/blade angles, pedal positions etc. ? That is the real issue I am sure for dual control airframes like helicopters and other 2-seated airframes, where the real world two sticks would be linked mechanically, while in the sim, the 2 pilots are unlikely to be sat next to each other sharing mechanically linked controls (much more likely to be in completely separate countries... and with completely different equipment). Syncing controls would be so hit and miss, or would require a great deal of skill and precision, or a very wooly lack of precision and hence a poor representation of flight. For one pilot to "hand over" control to another, they would need to both first match control inputs, angle of sticks, turn of pedals, throttle, trimming, etc in their physical sim-pit, so that when authority is swapped ("You have controls"), the aircraft does not jerk suddenly sideways or pitch forward, or nose up etc. I suppose there could be a short period of "autopilot control" during the hand over period, and then the new authority has to match that setting before authority can be granted. Even that would require perhaps a 10-20% allowance for imprecision, and a smoothing from the autopilot to the new pilot in control. Not ideal in a combat situation.... A great idea for multi-tasking, as in the Gazelle where one is basically the pilot, and the other weapons officer, (and similarly the impending Mi-24), or the Huey where one pilot is flying while the other operates the Flex guns... Of course, in real life, only one or other of the 2 pilots is flying the machine at any given time, even if both have hands resting on cyclic and collective, or FW equivalents. So yes, sitting as co-pilot and reading the map, operating the radios, arming the the weapons etc. that will be a great possibility for multicrew. But that transfer of stick control from one pilot to the other in the sim.... That is going to take some serious consideration of how. And I am sure that is the main reason it has not happened yet.

WOW, Hans! This post perfectly illustrates the generosity and kindness of some members of this DCS community... well, specifically, yourself! Thanks so much! I will give this my undivided consideration over the coming days. Also, a new set of Max487's arrived from a reliable supplier.... Mole

DCS-BIOS NVG Hi CptClownfish, For which DCS module are you trying to implement this? Mole