Periodic pause during Full-Auto

[GoogleCodeExporter]

What steps will reproduce the problem?
1. Full auto
2. High firing rate like lower one
3. keep the trigger pulled

What is the expected output? What do you see instead?
Full auto @ rate of fire selected

when holding it the trigger stop for 1/2 sec than start back sometime i could 
be shooting 30-40 balls one after the other without problem sometime only 10-20 
then there is the delay

What version of the product are you using? On what operating system?
0.7.hex

windows 7

Please provide any additional information below.

Original issue reported on code.google.com by maximesc...@gmail.com on 7 Aug 2012 at 5:08

[GoogleCodeExporter]

Reflashed Chip with same firmware and no problem,

Great software honestly

Original comment by maximesc...@gmail.com on 8 Aug 2012 at 1:38

[GoogleCodeExporter]

I had this issue as well and it bugged me for quite a while on what was 
actually happening.  It turned out to be a loose wire that leads to the battery 
(the red and black wires).  These wires are not very flexible and can lose 
connection easily especially when the trigger assembly is removed.  You may 
need to purchase a 9v terminal with wires and re-soldier the connector to the 
board.  I bought a $0.50 connector (or less, I can't remember) and the wires 
were much more flexible and haven't had the problem since.  I believe in one of 
my videos you actually see the old wire fall off while I was holding it.

For more detail, what was happening for me was:
During full auto, the marker would fire anywhere from 5-10 seconds 
uninterrupted, then pause for about 0.5 seconds and resume firing.  In 3 round 
burst mode, sometimes less than the specified number of rounds would fire, or 
cause 3rb to fire on trigger release rather than trigger pull.  The problem was 
due to an intermittent board reset cause by a faulty ground wire between the 
battery terminal and the board.  Vibrations caused while firing would shake the 
wire, occasionally causing a brief power loss that resulted in a board reset.  
The 0.5 second delay is cause by the board restarting, reading the EEPROM, and 
initializing.

The problem was debugged by adding a large delay upon board startup (for 
testing purposes only).  If the board was truly resetting, this delay would be 
triggered multiple times while the trigger is held during full auto.  If the 
board was not being reset, the delay would only be see once while turning the 
board on.

The result of this test showed multiple long delays proving that the board was 
being reset.  After much testing, it was determined that the wire was faulty, 
and the problem was resolved.

Please let me know if you see this issue regularly.  If you are hesitant to 
solder a new battery terminal as an attempt to solve the problem, I can create 
the above test code again for you to prove the issue.

Keep me posted.

Original comment by MikeDeH...@gmail.com on 14 Sep 2012 at 2:56

[GoogleCodeExporter]

Original comment by MikeDeH...@gmail.com on 14 Sep 2012 at 2:57

• Changed state: Feedback

[GoogleCodeExporter]

seems logical since the delay does always look like its a 0.5 second delay ill 
check it out this weekend i really dont mind soldering a new 9V battery clip as 
this one always scared me

Original comment by maximesc...@gmail.com on 17 Sep 2012 at 5:55

[GoogleCodeExporter]

Also, did you ever saw a random change of mode? After a certain while for no 
reason it would change from 3rd mode to 1st after a on/off 

Original comment by maximesc...@gmail.com on 17 Sep 2012 at 5:56

[GoogleCodeExporter]

Personally, I have not seen the presets change by themselves, but I believe 
what you saw is similar to Issue #2:

http://code.google.com/p/mad-phenom/issues/detail?id=2

I will be looking into this.

Original comment by MikeDeH...@gmail.com on 17 Sep 2012 at 6:04

• Changed title: Periodic pause during Full-Auto

[GoogleCodeExporter]

I think this issue is tied to the faulty battery wire.  If anyone sees this 
issue again after replacing their battery lines, I will re-open and investigate.

Original comment by MikeDeH...@gmail.com on 18 Sep 2012 at 3:17

• Changed state: Done

[GoogleCodeExporter]

When I'm in Fully Auto, I get a pulsating effect. I get the firing issue 
without being connected to the gun. My Etrigger will pause for about a .5 sec. 
I have checked my connections, and everything looks great.

Original comment by william....@gmail.com on 17 Jan 2014 at 2:25

[GoogleCodeExporter]

william.giarc0,

Which version of the firmware are you running?  Also, have you replaced the 
battery wires?  The Tippmann stock wires are rather stiff and break fairly 
easily.  In the past, the on/off behavior in full auto was due to the chip 
resetting after experiencing discontinuity in current from the battery.

Let me know which version you're using and if soldering the battery cables 
helped.

Original comment by MikeDeH...@gmail.com on 17 Jan 2014 at 5:42

• Changed state: Started

[GoogleCodeExporter]

1.0.1

Original comment by william....@gmail.com on 22 Jan 2014 at 12:22

[GoogleCodeExporter]

I will go get a new battery connection just in case. I have two X7 Phenom, one 
works fine and the other one pauses every now and then.

Original comment by william....@gmail.com on 22 Jan 2014 at 12:26

[GoogleCodeExporter]

ok, keep me posted.

Original comment by MikeDeH...@gmail.com on 22 Jan 2014 at 12:41

[GoogleCodeExporter]

Okay I connected a new heavy duty 9 volt connector. I still get a pause every 
now and then. I have two Tippmann X7 Phenom. I bought 5 chips and flashed 3 of 
them. I installed two of them onto the factory E trigger. One works really good 
with no problem. The other one will pause every now and then between shots. I 
re-soider the chip and connection just incase. However it stills pause between 
shots sometime.

Original comment by william....@gmail.com on 22 Jan 2014 at 3:00

[GoogleCodeExporter]

Even with the problem I'm having this codes still rocks. You are the man!

Original comment by william....@gmail.com on 22 Jan 2014 at 3:09

[GoogleCodeExporter]

I'm sorry you're having trouble.  Let me look over the code and see if I spot 
anything.  If you don't mind, I could send you a test build to check if the 
chip is resetting.  It would require you to flash the chip.

Are you experiencing this issue with the default settings?  What is your rate 
of fire set to?  Which preset number are you using for full auto?  Any other 
custom settings like dwell/debounce?

I'll try to get you a test build out tomorrow.

Original comment by MikeDeH...@gmail.com on 22 Jan 2014 at 3:13

[GoogleCodeExporter]

My setting for dwell and debounce is set to 8ms for dwell and 20ms for 
debounce. 
ROF for #1 memory is set to 15 bps
ROF for #2 memory is set to 20 bps
ROF for #3 memory is set to 30 bps
It will do it in all settings. Sometimes it will loop the pauses for a while. I 
will have to let go the trigger to get it out of the looping pauses.

Original comment by william....@gmail.com on 23 Jan 2014 at 1:33

[GoogleCodeExporter]

Here is a debug build that will help us determine if the chip is resetting.  
You can flash it to the chip the same way as the others (just with a different 
file name).

Do not flash this to any other chip than the one you want to test.  This build 
has an intentional 3 second delay when starting up which would be quite 
annoying to run with.

After you flash the chip, turn it on and you should see the red LED display for 
3 seconds before the marker becomes functional.  Then attempt to reproduce the 
issue.

The moment you notice a pause in the Full Auto mode, immediately look at the 
LED and note the color.  

I'm expecting the pauses to become much longer, and for you to see a red LED 
display when they do.

Original comment by mikedeha...@gmail.com on 23 Jan 2014 at 6:35

Attachments:

• mad-phenom-debug.hex

[GoogleCodeExporter]

I will download the code tonight and burn it to the chip in question. I will 
give you feedback afterwards.

Original comment by william....@gmail.com on 24 Jan 2014 at 3:15

[GoogleCodeExporter]

Did you ever find a solution for my problem? Both guns do the same thing. 
Should I re-flash the code on the chip?

Original comment by dalecra...@gmail.com on 30 Oct 2014 at 8:04

[GoogleCodeExporter]

I haven't heard anything back from William regarding the test/debug build from 
comment #17. If you are also experiencing the issue, would you mind trying the 
build in #17? I need to know if there's something causing the chip to reset.

https://code.google.com/p/mad-phenom/issues/detail?id=1#c17

Thanks.

Original comment by MikeDeH...@gmail.com on 30 Oct 2014 at 11:20

• Changed state: Feedback

[GoogleCodeExporter]

Hi, we experienced the same problem, tried the debug firmware and we do get he 
3 sec pause and a red light...

The stock firmware was working fine tho.

We also re-wired the batt conectors and verified the connections..
Same thing.  It somehow reboot quite often for some reasons. Every 2-3 sec on 
average when we hold the trigger.

Original comment by dave.god...@gmail.com on 8 Nov 2014 at 9:56

[GoogleCodeExporter]

Thank you for confirming, I am looking into the issue.

Original comment by MikeDeH...@gmail.com on 11 Nov 2014 at 4:16

• Changed state: Started

[GoogleCodeExporter]

Well it seems like we fixed the problem.

We added a rubber spacer between the hammer and the solenoid motor.
But that could be just a band-aid.

Could be something about the timing of + - maybe need a short pause between.
I dont know. ex: +(10ms) pause(2.5ms) -(10ms) pause(2.5ms), then just spacing 
the pause for slower rate of fire.

My intuition tell me that it's just alternating +- at a given rate, and it's 
why we have issue at any rate of fire.

But what do i know?

Original comment by dave.god...@gmail.com on 15 Nov 2014 at 12:33

[GoogleCodeExporter]

I have yet to duplicate the issue in-house, but I'm still digging.

Were you able to resolve the issue completely?

If so, keep me posted if you experience the issue again.

Original comment by MikeDeH...@gmail.com on 15 Nov 2014 at 12:37

[GoogleCodeExporter]

Tomorrow will be the day in the field, so we will know how it behave.

Is the software alternating +-?  Could be some timing with the clock on the 
attiny, Since it's a reboot it's very confusing.

But that's not my cup ;) so i'll keep you informed tomorrow.

Original comment by dave.god...@gmail.com on 15 Nov 2014 at 1:14

[GoogleCodeExporter]

With the gasket it's working flawlessly.

No more reboot.
Go figure.

I will post a picture of our fix.

Original comment by dave.god...@gmail.com on 15 Nov 2014 at 11:37

[GoogleCodeExporter]

Thanks. I'll post your fix on the wiki in case others have the same issue. I 
appreciate the feedback. I'm going to leave this ticket open for a few more 
weeks while I do more testing.

Original comment by MikeDeH...@gmail.com on 15 Nov 2014 at 11:41

[GoogleCodeExporter]

https://drive.google.com/folderview?id=0B-dFhx5FXmApRlViRE1MS0xId28&usp=sharing

:)

Original comment by dave.god...@gmail.com on 16 Nov 2014 at 6:15

Attachments:

• IMAG0469.jpg

[GoogleCodeExporter]

Hi, i'm the owner of the phenom dave talked about. This weekend i had to change 
the batterie, a soon has i did the periodic pause during the full auto came 
back. Seem like only when the batteries give their full voltage that i have 
this problem. From what i know: it's not a faulty wire(it with work corectly 
with old batteries, it does reboot(we tested it with the debug firmware)and it 
seem to work better with the rubber gasket. could it be that the selenoid 
magnegtic field disrupt the sensor that turn on and off the e-grip, since 
everithing is so close. And could the new batteries with full voltage make the 
magnetic field from the selenoid stronger, and the gasket keep the hammer 
further away from the sensor. If that is the case. would it be possible to 
program the on-off sensor so that it would take a certain amount of time for 
the sensor to be stable before switching it off??

Original comment by n.barbea...@gmail.com on 25 Nov 2014 at 11:54

[GoogleCodeExporter]

Hi, I just programed my x7 phenom tonight and it is having the same issue. It's 
definitely rebooting by watching the green LED blink while holding down the 
trigger. When auto fire pauses the timing of the LED between blinks also delays.

Wires look solid and the gun has only been on a field once, rest of the times 
just back yard shooting.

For reference, I didn't change any of the settings after flashing.

Cheers,
-Mark

Original comment by mwengran...@live.ca on 26 Nov 2014 at 6:53

[GoogleCodeExporter]

I found this forum post about the same issue, I think. Do any of you have the 
magnet in the slot behind the solenoid? Mine doesn't.

http://www.x7og.net/phenom-tech-help/30102-full-auto-pausing-other-issues-3.html

Original comment by mwengran...@live.ca on 26 Nov 2014 at 7:42

[GoogleCodeExporter]

i have no magnet in my e-grip, so i bought a return magnet from tippmannpart 
this morning. the magnet cost me 0.97$ and the shipping cost me around 8$.... 
kinda anoying. Any way i should recive it within 2 week ans ill keep you 
updated if it fix my problem.
Thank you for the hint.

Original comment by n.barbea...@gmail.com on 26 Nov 2014 at 10:55

[GoogleCodeExporter]

I called tippmann tech support today and asked them to send me a magnet for 
free. 0$ for me :D

Original comment by mwengran...@live.ca on 27 Nov 2014 at 6:27

[GoogleCodeExporter]

Thank you so much for that link. I suspect there is something going on with the 
power bus, but I haven't nailed it down yet. I've order another device to help 
me in the debugging process that should arrive this weekend.

Has anyone tried lowering their dwell? If so, did it affect the problem in any 
way?

Original comment by MikeDeH...@gmail.com on 27 Nov 2014 at 6:47

[GoogleCodeExporter]

i just tested my egrip with a lower dwell. It worked perfectly fine up to 4ms. 
At 5ms the periodic pause came back. i didn't shoot any ball with it, and every 
where i read about it, they say tha lowering the dwell to low make the marker 
unreliable.so i'm not verry comfortable lowering the dwell so low. Any way from 
what i read on the link the magnet will probably fix my problem.

Original comment by n.barbea...@gmail.com on 27 Nov 2014 at 11:34

[GoogleCodeExporter]

i finaly recive the magnet... and sadly it didn't work at all....but i got some 
good new, i think i found a final fix, mabe....i continued with the idia that 
the problem came from the selector switch sensor. so what i did is moove the 
sensor. i hade to cut away a peice of plastic so i could moove the sensor more 
freely. after many trial and error i finaly found a position for the sensor 
where everithing work corectly with new or old battery. then i glued it in 
place with hot glue. i had to keep the rubber spacerand the magnet without it 
it didn't seem to work as  well. 

Original comment by n.barbea...@gmail.com on 9 Dec 2014 at 3:33

Attachments:

• IMAG0471.jpg

[GoogleCodeExporter]

Ya I just got my magnet today, didn't help. What should the dwell be set to? 
What becomes unstable at 4ms?

Original comment by mwengran...@live.ca on 11 Dec 2014 at 5:51

[GoogleCodeExporter]

changing the dwell on mine to 4 also seems to resolve it. I wonder why this 
happens. Maybe i'll give tippmann a call and see what they have to say about it.

Original comment by mwengran...@live.ca on 11 Dec 2014 at 6:14

[GoogleCodeExporter]

I apologize for not getting back sooner. I intend on testing the dwell settings 
using a digital probe compared to the original chip. There's a possibility that 
4ms is not truly 4ms in the software. My timing might be a little off causing 
more dwell than intended. I'll try to get to testing this weekend.

Thank you for testing the dwell as a work around. If you notice anything 
strange with a lower dwell, let me know.

Original comment by MikeDeH...@gmail.com on 11 Dec 2014 at 6:31

[GoogleCodeExporter]

After some analysis, it looks like the chip is losing power externally. This is 
good new because it means the software is working as expected, but it's bad 
news because it means something else is wrong with the board.

I've attached a screenshot of the digital probe output where you can see the 
power to the chip is dropping off fairly regularly. I was able to duplicate the 
problem to the point where power was dropping for ~86ms every 0.8 seconds.

I have more experiments to run, but I'm wondering if this has something to do 
with electromagnetism causing the power hall effect sensor to reset. This could 
be caused by the solenoid being on long enough to trigger the power hall effect 
sensors debounce threshold, or the trigger magnet getting too close to the 
power sensor.

I will test these theories shortly.

Original comment by MikeDeH...@gmail.com on 14 Dec 2014 at 6:30

Attachments:

• chipReset.png

[GoogleCodeExporter]

Confirmed.

This issue occurs when the magnetism to the power hall effect sensor is 
present, but weak.

This theory has been confirmed by holding a magnet near the trigger sensors 
while testing the power hall effect sensor with a magnet at various distances. 
The expected result would be that the board powered on fully once the power 
magnet reached a certain distance and powered off when the magnet was far away. 
While this is true, a third state exists.

When a magnet is at a certain distance from the power hall effect sensor 
(varies by strength of magnet) the board will reset periodically (every ~0.8 
sec) when the solenoid is active. If the solenoid is active, no such reset 
occurs.

I've attempted the fix in the above comment. The problem improved, but did not 
resolve completely.

https://code.google.com/p/mad-phenom/issues/detail?id=1#c36

I did attempt a "workaround" that worked completely, though it's far from 
ideal. I bypassed the hall effect power sensor, essentially bypassing the 
selector switch and forcing the board to stay powered on. This workaround will 
disable the mechanical semi-auto and will require you to pull the battery to 
use mechanical semi-auto and turn off the board.

I've attached a picture of the sensor bypass (it's the terrible blob of solder 
connecting the last two leads).

Attempt this at your own risk. I am not responsible for broken boards, bend 
leads, or otherwise broken property. I provide no warranty.

Unfortunately, since this issue is outside of software control, there's nothing 
I can do to fix it programmatically. I'm going to keep playing around with a 
more permanent hardware solution, but until I found something better, I'm going 
to close the issue.

Original comment by MikeDeH...@gmail.com on 14 Dec 2014 at 10:39

• Changed state: Done

Attachments:

• sensorBypass.png

[GoogleCodeExporter]

I've ordered a new hall effect sensor in case the one on my board is just not 
working. I'll post an update after they arrive.

Here's the part I ordered:

http://www.mouser.com/ProductDetail/Honeywell/SS451A/?qs=%2fha2pyFadujv2DegI%2f6
G%2fg5sFVAZv3HpBfdL93OF9b4%3d

Original comment by MikeDeH...@gmail.com on 14 Dec 2014 at 10:43

[GoogleCodeExporter]

My new hall effect sensors arrived today and I believe they have fixed the 
issue. I took some measurements and discovered that the sensor did not line up 
with the selector switch magnet. When I soldered my new sensor, I raised it up 
a little to align with the magnet a bit better and I have not been able to 
duplicate the issue since.

Check out my attached pictures for a look at what I did. This change is not for 
the faint of heart as it requires soldering and modifying the original board's 
plastic inserts. I would only suggest doing this if this issue is affecting 
your game play.

ATTEMPT AT YOUR OWN RISK.

Original comment by MikeDeH...@gmail.com on 18 Dec 2014 at 3:53

Attachments:

• IMG_20141217_214541089.jpg
• IMG_20141217_214614912.jpg