Grass's pit, a redo

[metalnwood]

looking really good. I am hoping to get the backlight plate before christmas and spend some time working on panels as well.

 

If it is still of any use to you I use a blower that is a bit over 500cfm for my laser which has a fairly large enclosure around its 60"x47" table. It works well but like Gadroc has said I wouldnt go for less. You really want to get rid of any smoke and vapor caused by the cutting/engraving process.

[agrasyuk]

ventilation is indeed crucial. my current solution is tolerable for engrave jobs, but when i cut i just leave and let it sit and vent for some time after job is complete. this obviously can not go on like this so so we have proper ventilation in the works - big fan and a chimney through the roof. the connection of exhaust to the machine itself is also limiting, that is going to be resolved as well.

[Duckling]

Hi Agrasyuk. Your SAS qauds looks great, a real neat and compact build :thumbup:

 

Have you found a way to trigger the release of the magswitches 'hold' from within the Sim ?

 

I read a note from Gadroc some time back that the function wasn't implemented in the 10C exports. I might missunderstood and haven't search for a solution yet.

 

Cheers

Gus

[Deadman]

you might want to back light the area for the yaw trim knob

[metalnwood]

Heh, how did you manage that when you first got the laser? I had lots of things to cut and a bit of trial and error. My patience wouldnt have let me go away for a while for the air to clear :)

 

I know it can be a pita getting things like this done correctly, it's all a bit of an additional expense and doesnt add the the quality of the work but worth it in the end. Between doing dust extraction right and doing fume extraction correctly it's hard to think about all the money that goes in to that when it could have gone in to materials or another machine.

 

You are rewarded with warm fuzzies when you realise it's all OK and your health is safe - until you chop off a finger on the tablesaw or similar :)

[Hansolo]

Hi Agrasyuk,

 

Very nice panel and excellent idea with the circuit breakers. May have to borrow that idea :music_whistling: Probably a lot cheaper than the Honeywell ET series which cost a small fortune :-(

 

 

cheers

Hans

Edited December 17, 2012 by Hansolo

[agrasyuk]

thanks to all who replied. one thing to note guys: those dual phase circuitbreakers are very wide. stacked that dencely they are still over 5 inches wide, which means that this particular device will not fit the standard rack opening. good thing i have it un-prototypically wider

 

Have you found a way to trigger the release of the magswitches 'hold' from within the Sim ?

I read a note from Gadroc some time back that the function wasn't implemented in the 10C exports. I might missunderstood and haven't search for a solution yet.

 

Cheers

Gus

 

Hi Gus,

perhaps i got a bit carried away when i said "tested". the current software inside the Arduino trips the switches on monitor test input (and lights the indicator uponbutton press). for now just i pretty much verified the functionality of hardware.

 

but there will be proper export :) there is no trigger to trip specific SAS switch, yes. DCS however does export the state of the switch itself (hence clicking on it in sim changes position in helios screen). I still need to build a serial bridge software that going to listen to datagrams DCS exports - i plan on watching the state of each switch and send on pulse on matching UNO output each time i detect transition from ON to OFF. me not being software developers makes this part challenging. but i already sucseeded in sending a command from Arduino IDE to the device. it all would be much better if i could convince that board that it is EOS device, but i just can't grasp how.

 

you might want to back light the area for the yaw trim knob

you right, as it is now the knob will not get back-lighted. but the front panel itself comes off without disassembly of everything else.

and yes, i still need to check in into your knob thread :)

 

Heh, how did you manage that when you first got the laser? I had lots of things to cut and a bit of trial and error. My patience wouldnt have let me go away for a while for the air to clear :)

 

I know it can be a pita getting things like this done correctly, it's all a bit of an additional expense and doesnt add the the quality of the work but worth it in the end. Between doing dust extraction right and doing fume extraction correctly it's hard to think about all the money that goes in to that when it could have gone in to materials or another machine.

 

You are rewarded with warm fuzzies when you realise it's all OK and your health is safe - until you chop off a finger on the tablesaw or similar :)

 

indeed. and it was hard to resist, yes, at first lack of patience was making me inhale some of those fumes. now i'm somehow don't want to do that anymore. :music_whistling: air seems to be clean 10-15 minutes after, and i just let the fan run overnight after evening of laser work. and there is always CAD work to do while i wait :)

 

Hi Agrasyuk,

 

Very nice panel and excellent idea with the circuit breakers. May have to borrow that idea :music_whistling: Probably a lot cheaper than the Honeywell ET series which cost a small fortune :-(

 

 

cheers

Hans

HI Hans. sure no prob:) keep in mind however that those doule phase switches are very wide. in order to fit 4 of these into prototipicaly sized panel you will need to use single phased ones. this makes things just a ted more complicated as you don't have separate phase for signalling and will need to add a relay or something to .

Edited December 17, 2012 by agrasyuk

[Hansolo]

Thanks for the heads up about the size, Anton. Do you happen to have a type for the ones you used. Thanks in advance.

 

 

All the best

 

Hans

[agrasyuk]

Whatever you can find for acceptable price, these breakers can get expensive as well , some people want crazy dollar for them. The most important part i think is for them to look like toggles and not having crazy amp ratings. I scored a 4 pack of Airpax 1A 2poles breakers, would have preffered single pole to fit better, but oh well. I think I seen a reasonably priced 7A 4pack of single poles on eBay the other day, not sure if it is to much A.

[Gadroc]

I read a note from Gadroc some time back that the function wasn't implemented in the 10C exports. I might missunderstood and haven't search for a solution yet.

 

I've been thinking on this and for certain situations it could be possible but not full-proof.

 

In DCS interfacing with the cockpit is essentially emulating mouse clicks. Each clickable item is given a click action and optionally a release action. For magnetic switches you have to fire both events or the magnetic portion doesn't work. Essentially if you just fire the click event DCS never thinks you've released the finger pressure on the switch. Think of it like the key in your car. It will stay in the start position as long as you hold pressure, but revert to on after you release pressure. The click event tells the sim you have moved it to the start position and the release event is telling the sim you released the pressure. If you never send the release event it will never revert to the on position as the sim thinks your hand is still turning the key.

 

You can see this in action with the anti-collision switch using Helios. Send a set position command and the game will leave the switch where you tell it to. If you then send the release command it will "do the right thing".

 

In order to get magnetic switches to work in the real pit you need to intercept moving into the magnet position. You'll need to send a "click command" and then after a pre defined wait period send a "release command" when the SAS switch is moved to the on position. Second you make the magnet trigger with in game switch position. If the switch is on make the magnet on. This will work as long as the user holds the switch into the on position for longer than the wait period, and it has the side effect of the magnet always switching on even for a short period of time. Which may be different than real behavior of the switch. I haven't tested this so it might have some additional side effects.

 

As a side note that sometimes you can supply more information than a mouse click via lua as you can pass a value in. Some rotaries that you can specify the increment size (aka in game it defaults to incrementing by 5 but using lua you can increment by any number instead of just 1). Which makes a well interfaced rotary potentially more accurate than using direct bindings in game, although you have to be careful about speed of execution via lua.

[agrasyuk]

The click event tells the sim you have moved it to the start position and the release event is telling the sim you released the pressure. If you never send the release event it will never revert to the on position as the sim thinks your hand is still turning the key...

 

i see what you are saying. Just tried it, indeed DCS does not see the invisible hand releasing the switch that otherwise would have sprung back. it holds it when triggered with key shortcut , or control input, or export. the only time it releases it is when it was engaged with mouse... . that means that i will never receive the command to trip the switch. I can see how even adding logic to handle those special buttons into helios not resolving the issue. somehow i doubt the needed info is exported at the moment. it seems that petition to ED is in order to provide those additional exports, but what is the likelyhood of that...

here goes the perfect plan :doh:

 

 

 

but at least it can reset from the hardware monitor test switch, not much joy, but something... :noexpression:

 

EDIT: actually it can't be so. if someone even without a pit uses just the shortcuts to engage SAS then it means it will never emergency disengage in game. i wonder if there is some sort of timeout for this . going to play with this for a bit.

Edited December 18, 2012 by agrasyuk

[agrasyuk]

went through a bunch of electronic surplus stores and looked through lots of builds here in search for good looking momentary push buttons. for my previous panel attempt i used these (got red, painted them black). I almost ready to reuse them but the bezel bothers me a bit. for my SAS TO trim button i just used a tact with round cap. works , but now i'm looking for something nicer. i really like the clicky switch action of these but getting LED button and painting over seems counterproductive. any suggestions without breaking the bank?

input welcome. thanks!

[Duckling]

I've been thinking on this and for certain situations it could be possible but not full-proof.

 

In DCS interfacing with the cockpit is essentially emulating mouse clicks. Each clickable item is given a click action and optionally a release action. For magnetic switches you have to fire both events or the magnetic portion doesn't work. Essentially if you just fire the click event DCS never thinks you've released the finger pressure on the switch. Think of it like the key in your car. It will stay in the start position as long as you hold pressure, but revert to on after you release pressure. The click event tells the sim you have moved it to the start position and the release event is telling the sim you released the pressure. If you never send the release event it will never revert to the on position as the sim thinks your hand is still turning the key.

 

You can see this in action with the anti-collision switch using Helios. Send a set position command and the game will leave the switch where you tell it to. If you then send the release command it will "do the right thing".

 

In order to get magnetic switches to work in the real pit you need to intercept moving into the magnet position. You'll need to send a "click command" and then after a pre defined wait period send a "release command" when the SAS switch is moved to the on position. Second you make the magnet trigger with in game switch position. If the switch is on make the magnet on. This will work as long as the user holds the switch into the on position for longer than the wait period, and it has the side effect of the magnet always switching on even for a short period of time. Which may be different than real behavior of the switch. I haven't tested this so it might have some additional side effects.

 

As a side note that sometimes you can supply more information than a mouse click via lua as you can pass a value in. Some rotaries that you can specify the increment size (aka in game it defaults to incrementing by 5 but using lua you can increment by any number instead of just 1). Which makes a well interfaced rotary potentially more accurate than using direct bindings in game, although you have to be careful about speed of execution via lua.

 

(Sorry for derailing the thread Anton)

 

Thanks Gadroc. That explain why I have problems to get switch position to trigger the 'on' position even when the maglock is engaged and holds. Not everytime though at thats even stranger but I'll meditate on the quiz over the Xmas and see if I can get it to work.

/Gus

[agrasyuk]

Did some more tests last night. It seems that the Sim releases the "invisible hand" as soon as magnetic switch is held " legaly" by the Sim systems. So yes, SAS will stay up if engaged in cold cockpit due to hardware switch holding it. But if it is engaged "legally" it will drop upon monitor test even though hardware still commands ON. Same with antiskid - it will drop on ebrake even though hardware switch commanding ON. So I will get the drop ON-OFF transition event I'm looking for to trip the breakers as long as I'm engaging only when it is going to be held. I'll take that :)

[Gadroc]

(Sorry for derailing the thread Anton)

 

Thanks Gadroc. That explain why I have problems to get switch position to trigger the 'on' position even when the maglock is engaged and holds. Not everytime though at thats even stranger but I'll meditate on the quiz over the Xmas and see if I can get it to work.

/Gus

 

I have had problem in the past when you send commands "out of the expected sequence" to the sim. For example if you repeatedly send the click command with out sending the release command for a switch. Helios early on had problems with this type of issue and the OSB buttons getting stuck.

 

I imagine this is "non expected" behavior based on normal sim usage and could trigger untested scenarios inside the code that runs those parts of the simulation. My recommendation is to always send a release command for switches that expect them.

[Duckling]

Been thinking a while about the SIM state of the 'hold' position of the MagSwitches on SAS and EAC but haven't done much experimenting on them, backout plan so far is if unable to get the function automated, I could always trigger the emergency release on the stick to at least be in sync with the SIM.

Since there as several conditions that needs to be fulfilled to have the hold condition activated, wonder if that state itself, per function, could be reachable somehow. i.e. if conditions per switch is status 'enabled', power to the coil will be made etc and the that would be regardless of the position of the switch. If condition is disabled, maghold release and the switch position is updated through normal procedure.

(to many words here and bad grammer but I think you know what I mean :-)

 

(have to admit I'm not (yet) using Helios but getting closer by the day, hehe)

 

Best

Gus

[agrasyuk]

Fuel panel construction.

considering using another Arduino UNO for this, while there is not enough digital I/O's i don't think it will be an issue as the few missing switches can still be connected to analog input and handled with analogRead() instead (or if it comes to this sig Amp switch together with test buttons are not really nessesery and can be skipped). i'm still reading about programming UNO to be seen as just another USB joystick.

 

 

not exactly prototypicaly shaped and sized , but resembles the actual thing close enough for me. backplate and spacers yet to be painted, receiver lever assembly is still in works

 

 

testing fit

 

[Gadroc]

Analog pins can be used as digital inputs as well. No need to do an analog read on them.

[agrasyuk]

Thanks Gadrock, I did not know that. My adventure into arduino is just started, lots to learn. I figured I can take input from analog input and "digitize" it with code, but reading it digitally makes for one less thing to worry about.

Could you chime in on how to make arduino talk EOS ? Will it need additional RS shield on it? There is similar question in your workshop forum, but it left unanswered. I remember some libraries and protocol posted on old site, but I didn't give it much thought then.

 

 

Played more with software for the SAS panel - had absolutely stupid moment, i did receive confirmation message in serial monitor but it refused to trip the breakers. pulling my hair out as I curse myself for not providing diagnostics indicators on transistor board. And then it hit me, the power supply probably should be plugged into the wall for this to work .:doh:

Edited December 19, 2012 by agrasyuk

[agrasyuk]

Backlight test. realized that i missed one of the LED strings

[0utLAW]

Nice job!

[Hansolo]

Veri nice indeed :thumbup:

[agrasyuk]

TY guys.

 

on fuel panel started fabbing receiver lever assembly while i wait for more arduino boards, going to tackle the knob as well.

 

meanwhile, AHCP box:

 

 

test fit so not all switches mounted. tried to make the enclosure as shallow as possible, cleared the bottom with less then 1mm to spare. i would love to show it lit but i'm missing handul of LEDs to complete it, so backlight is on hold.

 

 

 

for those wondering why it is in form of standalone box and not just panel the answer will come soon. hope i'm not going to get to much flack for unorthodox solution. for now perhaps a hint - fabricated 2 thingies pictured :smartass:

Edited January 30, 2013 by agrasyuk

[MTFDarkEagle]

Ah cmon, at least get the proper locking switches! :D

 

Looking good man ;)

[agrasyuk]

well , i found the P/N in deadman's thread, but i can't seem to find the actual switches for sale. nor any other locking switches resembling the actual ones for reasonable price. some pop on ebay here and there, but what good is it if i can't get a set of the same for at least ARM and tgp switches. later on i'm thinking to make tubes to dress them up a bit