VR gun recoil.

Post by **bobv5** » Wed Nov 16, 2011 8:19 pm

I have been thinking about adding recoil to VR weaponry.

It seems to be that the games audio will be the best signal to use. So far I have come up with a circuit that filters all audio that is not from front dead center. (Where the gun would be, at least with no head tracking.) It can be adjusted for software that has a spread on the gun sound, for example, if it was off from the center by a few degrees. I think if the sound comes from front dead center, and you are pressing the trigger, and the extreme high and low frequencies are filtered out,it is probably a gun sound.

Do you guys have any ideas of any other ways I could filter the audio before it to got the the microcontroler?

Post by **WiredEarp** » Wed Nov 16, 2011 9:45 pm

Really, the 'only activating when the trigger is pulled' bit is the best idea for reducing unwanted effects in a universal way. The idea of only using some channels is also good.

Short of memory hacking this is the best you'll get I think. I think my idea of trapping sound effect playback would provide better reliability (if it worked at all) but would require customization for every game. your idea is definitely worthwhile...

Post by **mAchiNE** » Thu Nov 17, 2011 10:33 pm

Do mean filters sound from the center channel on 5.1? I found that there is no gun sound in the center channel on COD Black ops, at least not enough to trigger my voltage switch, I managed to get it to trigger on the left or right channel. If you are talking about picking up sounds dead center from a stereo source then that sounds like A good Idea to me, should remove most unwanted sound with the exception of explosions, the problem I found is that guns with silencers and some pistols etc did not trigger my setup without adjustment, but your filter sounds a lot more complex than my voltage switch with high pass filter on it so I will be looking forward to your results!

I decided to try out a non audio solution using a selector switch to switch between 3 different modes; semi auto (rifles and pistols) Shotgun (increased force but still semi auto) and Full Auto (for machine guns)

so basically the 2 semi auto modes will be triggered once every trigger pull and the full auto will simply provide constant recoil when the trigger is held down at a set firing rate (somewhere between 15-20 shots per second) I'll probably look into the most common fire rate of commonly used assault rifles, it may not match the in game fire rate perfectly but the effect of feeling the recoil may be just as good and maybe not noticeable that there is a difference. This also eliminates the need to transmit the audio (or switching) information to the controller. I intend to have the selector switch also map 3 buttons so you can for example have a pistol, shotgun and assault rifle in the game and map those weapon selections in game to the selector switch on the controller so the firing mode in game matches the firing mode selected on the controller at any given time.

Obviously the Audio based solution is going to have its advantages over this but the audio based will require more R&D to get working well I think.

Post by **WiredEarp** » Fri Nov 18, 2011 9:05 pm

@Machine: your idea is the most easily achieved. I like the way you had the idea of using different weapon selection buttons to change modes. Some guns have different firing modes however which could complicate things.

Its not really necessary IMHO that the recoil matches up with full auto fire, just that it shakes the gun.

A couple of other ways:

- you could have a 'ramp up' mode where when you pull the trigger, it fires a single shot, waits a 1/2 second, then starts firing fully auto. so when you start rock n rolling there will be a brief pause before the recoil starts (although you still get the first shot recoil) but I dont know if this will be really a problem.

- you could use a 2 stage trigger, and just pull it 1/2 way (first stage) to fire single shot recoil... then when firing full auto just pull the whole way to trigger the full auto recoil motion.

Post by **mAchiNE** » Sat Nov 19, 2011 3:56 am

@WiredEarp
Yes the idea is simple to make that is why I am doing it instead of audio based for now, just so I can get my controller up and running ASAP, I will continue to look into the audio based version at a later stage.

Some guns have different firing modes however which could complicate things.

well those guns will need to be switched firing modes in game which normally requires pressing a button (and also requires pressing on a selector in real life as well), so should not be a problem, although the type of gun you are talking about normally switches between semi auto and burst fire or burst fire and full auto, so I guess I may need to add a fourth option for burst fire mode.

you could have a 'ramp up' mode where when you pull the trigger, it fires a single shot, waits a 1/2 second, then starts firing fully auto. so when you start rock n rolling there will be a brief pause before the recoil starts (although you still get the first shot recoil) but I dont know if this will be really a problem.

Its hard to say without tryint it out, but I think a single recoil followed by a delayed automatic recoil might be off putting

you could use a 2 stage trigger, and just pull it 1/2 way (first stage) to fire single shot recoil... then when firing full auto just pull the whole way to trigger the full auto recoil motion.

This is actually an interesting idea, easy to implement too, might be worth looking into

Post by **bobv5** » Mon Nov 21, 2011 9:46 pm

Yea, I meant a stereo source. As I will be using it with a hmd I have no need for 5.1.

Not really got anything done since I posted that. Got a few days off so should be able to make some progress. If nothing else I will try and post a video of the simulated version of the circuit.

Post by **bobv5** » Fri Dec 02, 2011 7:58 am

Taking more time than I hoped to design the circuit. Got something simulated pretty good now, just tweaking it to work in the real world. Started ordering parts, solenoids arrived today. Bigger than I expected, should have plenty of power.

Post by **PalmerTech** » Fri Dec 02, 2011 1:56 pm

Thanks for the progress update, I can't wait till you get something working. I would love to buy a copy of whatever you put together.

Post by **bobv5** » Fri Dec 02, 2011 3:58 pm

I don't know how long it will take me to get a working prototype, but I would love to sell you one

What sort of package would you be looking for? Just the circuit, or circuit and solenoid?

What sort of power supply are you using? The solenoids in the picture will suck a lot of power if you let them.

Post by **mAchiNE** » Fri Dec 02, 2011 4:47 pm

Yeah, do you plan to make this a wireless controller? if so I think those solenoids will kill your battery pretty quick

Post by **bobv5** » Fri Dec 02, 2011 5:56 pm

I plan for the part of the system strapped to me to be untethered, but the controller itself will be wired to a backpack. I will have to check my maths to be sure, but I beleve they will use about 15 watts each. Not too bad as long as your power system is designed to cope.
As with anything it comes down to cost. Or to be more specific cost vs weight. Idealy this and any other haptic feedback gear will have a dedicated battery as it needs higher voltage than everything else. Other ways are possible, but with diy you would be adding more complexity for very little gain.
A powerful haptics system will always need a powerful battery.

Post by **WiredEarp** » Fri Dec 02, 2011 6:32 pm

A powerful haptics system will always need a powerful battery.

Good rule. Makes sense, to exert energy takes it!

Looking forward to seeing how it all comes together, I think it will add significant immersion.

Post by **cybereality** » Fri Dec 02, 2011 6:49 pm

Very nice! I can't wait to see what becomes of this.

Post by **mAchiNE** » Fri Dec 02, 2011 8:48 pm

another option for powerful haptics with low (electrical) power is to use air pressure and a compressor, then all you need to do is power the compressor, the TN Games 3rd Space vest uses this method (compressor runs on a 12V 1.5A power supply), It should even be possible to use their air compressor if you are planning to use their vest in your setup, the compressor can go in your backpack along with the power supply for HMD, WHDI transmitter etc (the compressor is reasonably small and light). You can then use your audio circuit to control electric pressure valves instead of solenoids, but I guess this makes it more complicated and you still need a cable running to a backpack anyway so your current prototype is probably best bet, but it could be a good backup idea if you find that your setup becomes to heavy because of the battery, or if battery life is just to short.

Also I see you have 2 solenoids, if you are going to put both in one gun controller a idea I have (which I plan do do on the controller I am building at the moment) is to have one pushing the gun back (probably attached to the inside of the stock) and one pushing the gun in an upwards direction to make the recoil more realistic, and have only the backwards pushing solenoid activated on low powered guns (pistols, low powered rifles and assault rifles etc) that have small recoil in real life and both activated on powerful guns (like full auto machine guns and shotguns etc)

In any case it sounds like you have made some very good progress on this, if you can get your audio control circuit working I would also be interested to buy one off you

Post by **bobv5** » Fri Dec 02, 2011 9:33 pm

mAchiNE wrote:another option for powerful haptics with low (electrical) power is to use air pressure and a compressor, then all you need to do is power the compressor, the TN Games 3rd Space vest uses this method (compressor runs on a 12V 1.5A power supply), It should even be possible to use their air compressor if you are planning to use their vest in your setup, the compressor can go in your backpack along with the power supply for HMD, WHDI transmitter etc (the compressor is reasonably small and light). You can then use your audio circuit to control electric pressure valves instead of solenoids, but I guess this makes it more complicated and you still need a cable running to a backpack anyway so your current prototype is probably best bet, but it could be a good backup idea if you find that your setup becomes to heavy because of the battery, or if battery life is just to short.

12V 1.5A is 18 Watts. It's not really that simple but as it runs continuously it is probably still higher overall. A more direct energy conversion is usually more efficient than a less direct conversion. For example, electricity, to motor, to compressor, to air, to movement, is probably going to be less efficient than electricity to movement. Those electric presure valves, they are probably smaller solenoids. For a vest that system might be a good choice, but I think they are not appropriate in this application.

[quote="mAchiNE""]In any case it sounds like you have made some very good progress on this, if you can get your audio control circuit working I would also be interested to buy one off you

[/quote]

I would be happy to sell one to you too

Obviously no obligation, is anyone else intrested in buying this? Would people be wanting just the circuit board, or a complete package?

Post by **WiredEarp** » Sat Dec 03, 2011 12:54 am

I think air pressure would have more initial torque however, so would possibly give more of a realistic sensation (jolt not shove). Certainly, this was the method traditionally used at places like Disneyland etc to simulate gun firing, and it was also used by one of the Virtuality systems that had a shotgun with air powered recoil.

I think air is probably a good way to go for a wired system (however, would also be noisier which can be a concern), but for a wireless system a solenoid would have the advantage being able to be made more easily portable.

Post by **cybereality** » Sat Dec 03, 2011 2:05 pm

I have the TN Games 3rd Space Vest. It is no joke. It really does hit hard.

Post by **bobv5** » Wed Dec 07, 2011 6:12 pm

Sure, you could have a pneumaticly operated system. Like I said before, strong haptics needs a strong power source. It doeesn't need to be electrical. You could run it from a paintball gas canister, you could even strap a tank of propane to yourself and have it ignite in the controller.

I am trying to make a device that can decide when to fire the recoil mechanism. What that mechanism is, is really up to people to decide for themselves.

Cyber, in the driver you hope to make, will it have suport for devices such as TN vest? It looks like a nice piece of hardware, but as you know it is usless without software. I don't currently have the skills to include usb in my device, but if you can emulate a serial port over usb I can give you pretty direct control of the output.

Post by **Synexious** » Wed Dec 07, 2011 7:59 pm

bobv5 wrote:You could even strap a tank of propane to yourself and have it ignite in the controller.

Er, what?

Post by **WiredEarp** » Wed Dec 07, 2011 9:48 pm

Looking forwards to seeing you finish this driver bobv5. I think your idea of using the sound + trigger pull to trigger the effect is the best so far in terms of usability with all games.

Post by **cybereality** » Thu Dec 08, 2011 8:34 pm

@bobv5: Yes. I plan to support force-feedback devices with my driver. This was actually one of the main reasons I wanted to do this. Ideally, I will support consumer devices like the TN Vest, Novint Falcon, etc. and then provide an API for people with more exotic and/or DIY devices. But again, I am in the very early research phase right now. It may be months (or longer) before I actually have something usable.

Post by **bobv5** » Wed Dec 14, 2011 3:33 pm

Aphradonis- I'm not saying it is a good idea for someone to build with junk around the house, but if you have access to machine tools and know how to do the maths, it could give a very powerful, fairly cheap recoil.

I hope to get something working as soon as possible, but at this time of year I don't have a spare penny to go towards my projects. The good news is that I have an SMD rework station in the post, (Thanks Mum

) so will be able to make the boards far easier, and use components that were previously unavailable to me.

Cyber, I'm in a similar situation, I have the hardware and firmware working in a simulator. I hope to have a working hardware prototype early next year. I understand that for your driver haptic feedback is not a priority now. I'm not great at software, what would your driver be expecting from a USB connected haptic device? I know I really need to read up on USB protocols, but far too busy at the moment, so any vague description would be good.

Post by **cybereality** » Wed Dec 14, 2011 9:50 pm

bobv5 wrote:I'm not great at software, what would your driver be expecting from a USB connected haptic device? I know I really need to read up on USB protocols, but far too busy at the moment, so any vague description would be good.

To be honest, I haven't thought about it too much. Still have a lot of ground-work to do first. The general idea would be that the driver would provide an assortment of output values, and there would be a C++ API that would give you access to these values. For an FPS game maybe there would be one floating-point value for the gun recoil. For example, the number 0.5 would indicate a medium level kick-back for a pistol, while rapid values of 0.3 could be for a machine gun. I could also find a way to indicate what type of gun was used, but this would be a little harder and probably would need to be done on a case by case basis for each supported game. Again, I don't want to hype this too much, but I am working on it.

Post by **bobv5** » Fri Dec 23, 2011 4:11 pm

Quick update, recievied the chips for the first prototype today. Will have to see if I can find time to sneak off to the lab.....

Post by **PalmerTech** » Thu Feb 02, 2012 4:07 pm

Any updates on this?

Post by **bobv5** » Thu Feb 02, 2012 9:45 pm

Hoping to get some pcbs made soon. I need to make a small change to the board layout, then I will have to wait untill I can get some time with the equipment. Hopefully sometime this month.

Post by **PalmerTech** » Thu Feb 02, 2012 9:47 pm

Awesome! I have this gun on the way: http://www.amazon.com/STAR-Dragon-AEG-A ... B00547O6O0" onclick="window.open(this.href);return false;

Looking forward to building a nice tracked gun, hopefully with whatever you figure out.

Post by **mAchiNE** » Fri Feb 03, 2012 6:24 am

nice choice Palmer, i can see an analogue stick and some buttons going nicely into the forward grip of that odd looking gun

Post by **bobv5** » Fri Feb 03, 2012 9:28 pm

Palmer, would you just want the pcb or a complete system? The solenoids I am using are designed for 24v. I have no idea how they will perform on whatever voltage that battery pack you are using provides. (18v max I think?)

The output stage is quite robust, (about 30w max for each solenoid, designed to control two independantly). I can make it stronger if needed, but if people want it stronger I need to know soon.

It is too much hassle for me to design a circuirt that can be electronically linked to the trigger on peoples controllers. It will need to have an extra microswitch installed which is opperated by the physical trigger. The reason for this is that one person might use a wiimote, the next might use sony stuff, the next might use something i have never heard of. They are not compatible at the level my circuit works at.

I also have a few other ideas for this, which I have already designed into the hardware. For example, I plan for the reload button to be activated by pulling or pushing a physical magazine. This action will make a small activation of the solenoids. Just enough to feel like a mechanism, rather than a button. (optional of course, but I think it is a good idea)

Post by **PalmerTech** » Fri Feb 03, 2012 9:33 pm

I can step the voltage up no problem, so no worry about that. A complete system is probably best, if only so we both have the same hardware. Better for testing! As far as power, stronger is better in my book.

So, your circuit will need a sound input, a microswitch for the trigger, and 24v for power?

Post by **bobv5** » Fri Feb 03, 2012 10:07 pm

I can provide enough force to break bones, if you have the cash.....

My circuit will need a stereo sound input, a switch for the trigger (plus other switchs for reload etc if you want that stuff), and the voltage that the solenoids need. The solenoids I am using need 24v. I havent actually tested them mounted to a gun yet, but they are about the biggest that could be used without needing a stupidly huge battery.

Post by **PalmerTech** » Fri Feb 03, 2012 10:25 pm

Heh, use your own judgement in regards to force, then. I am planning on a handgun and a rifle, so I would be interested in two units.

Reading up on solenoids, it looks like I might not want to have them drawing from the same power supply as my more sensitive gear, so I will probably get a cheap high draw ni-mh pack to run them.

Post by **bobv5** » Sun Feb 05, 2012 12:06 am

I have designed the circuit to minimize the noise that gets passed to other components on the same supply, but that stuff is basically voodoo. A seperate battery for haptics is not a bad idea, but if I have done my job properly it won't be needed. (I'm learning this as I go, so it is possible I did not do my job properly.)

Post by **mars3554** » Tue Feb 07, 2012 12:15 am

Go to a thrift store like savers or a garage sale and get yourself an electric nail/stapler gun aka brad nail gun. Take it apart and solder together the "wall surface detection switch" wires and remove the magazine. Connect to Wall AC and fire away. Dead simple, good kick, and no worrying about sourcing a solenoid or designing driver circuitry.

Post by **PalmerTech** » Tue Feb 07, 2012 2:31 am

That would not work with a battery pack, though, which is what we are trying to accomplish.

Post by **mars3554** » Wed Feb 08, 2012 6:48 pm

Ah, sorry I didn't read through the entire post. I would opt for something like a autohammer then: http://www.sears.com/craftsman-11818-ne ... 911818000P" onclick="window.open(this.href);return false;

That or a cordless hammer drill on the hammer setting.

The advantage of the two I listed above is they use a motor and cam to generate the linear "shock" so I believe that would be more power efficient.

I think that using a DC solenoid to get a good kick would eat up a lot of battery life.

Firing the electric nailgun I hacked on rapid fire caused the lights in the room to flicker.

Post by **PalmerTech** » Wed Feb 08, 2012 6:54 pm

Yeah, a geared motor system has a lot of advantages. That is one reason I am ordering that airsoft gun, going to mess around with using the firing mechanism (Quality, all metal!) for recoil. The disadvantage is a small lag between pulling the trigger and the first kick, and less flexibility in terms of how the recoil feels for each weapon.

Post by **mars3554** » Thu Feb 09, 2012 1:11 am

Something like this?

http://youtu.be/IXHUFScbcFM" onclick="window.open(this.href);return false;

Post by **PalmerTech** » Thu Feb 09, 2012 1:30 am

Yes, something just like that.

I may very well end up using something like that for my rifle, but a solenoid might end up being the better option for a handgun. Only way to know for sure is to experiment! Great idea on the autohammer, though, that never even crossed my mind. More brains means more ideas means better VR!

Post by **mars3554** » Thu Feb 09, 2012 2:25 am

I purchased it to test it out, the thing is almost too quick (3600rpm or 60Hz) and the stroke is very short.
This video gives a good demonstration:

http://youtu.be/sDu6LIj5LFI" onclick="window.open(this.href);return false;