Lane Posted January 20, 2019 Author Report Share Posted January 20, 2019 On 1/11/2019 at 6:19 PM, Armed Eye Doc said: It looks like This one's for you sir. I just got a powder trickler in the mail yesterday. I think it needs a larger size stepper motor to make it look nice; but I'm not paying someone $200 to attach a rubber band between a motor and a trickler. If you've ever seen or heard a 3d printer run; you'll know stepper motors can be made to sing. That could help ensure single kernel output and perhaps help reset any scale errors with the vibration. Quote Link to comment Share on other sites More sharing options...
Armed Eye Doc Posted January 20, 2019 Report Share Posted January 20, 2019 Excellent work ? Quote Link to comment Share on other sites More sharing options...
98Z5V Posted January 20, 2019 Report Share Posted January 20, 2019 1 hour ago, Lane said: That's how this timer works as well. I thought I remember you saying earlier you can press GO and let it run all day long though. Ah, valid observation - and correct. If we ran the timer on a course of fire, and just set the timer on the table afterwards, it would keep going and recording shots, as long as it has something to record. If people continued to shoot, all day long, it would have one LONG shot string recorded. It times out after that last shot recorded, and shuts itself off, if there are no more sounds to log. If the shooting never stops, it lists a big shot string. Quote Link to comment Share on other sites More sharing options...
Lane Posted January 20, 2019 Author Report Share Posted January 20, 2019 4 minutes ago, 98Z5V said: If we ran the timer on a course of fire I'm definitely good with a big long shot string; should be able to diagnose full auto fire too. I never thought about an auto-shutdown time. That would do a lot in terms of battery savings. How long do you care about leaving it on and running? That was and still is a huge issue in terms of RAM calculations which I haven't finished yet. Quote Link to comment Share on other sites More sharing options...
98Z5V Posted January 20, 2019 Report Share Posted January 20, 2019 8 minutes ago, Lane said: How long do you care about leaving it on and running? That was and still is a huge issue in terms of RAM calculations which I haven't finished yet. I usually catch it when I do that, and laugh about the 390-second shot string we just logged... Then I turn it off, because the other firing is still going on. I have come back to it much later, and it's sitting on the table, powered down from inactivity. It clears all shot strings when it's turned off (or goes off), so I never know how long it ran before it ran out of things to record and shut down due to inactivity. Quote Link to comment Share on other sites More sharing options...
Lane Posted January 20, 2019 Author Report Share Posted January 20, 2019 22 minutes ago, 98Z5V said: I usually catch it when I do that, and laugh about the 390-second shot string we just logged So even a 10 minute window is acceptable? Does anyone's pager style shot timer do better than 1ms resolution? Quote Link to comment Share on other sites More sharing options...
Robocop1051 Posted January 20, 2019 Report Share Posted January 20, 2019 2 hours ago, Lane said: I was pretty sure you needed to have that worked out already... "Copper" grounding rod is really steel core; but C101 is about $24 a foot in 5/8"; if each one is 3" long that's about $6 per projectile. I'll turn some test projectiles on the Unimat when I get all this other stuff all finished up, (just for fun of course). ... 911 grains to grams is about 60; thats about two troy ounces, a forged gold projectile would be about $2,570 a round; but silver on the other hand is closer to $31 a pop. You know; just in case... We are actually working on our own composite material that we have been testing in .30 cal rifles. We are also entertaining the concept of brass vs copper for value. Quote Link to comment Share on other sites More sharing options...
Lane Posted January 20, 2019 Author Report Share Posted January 20, 2019 2 minutes ago, Robocop1051 said: composite material Carbon based; or plastic type? I'm just fine with less lead in the projectile. Quote Link to comment Share on other sites More sharing options...
98Z5V Posted January 20, 2019 Report Share Posted January 20, 2019 (edited) 21 minutes ago, Lane said: So even a 10 minute window is acceptable? Does anyone's pager style shot timer do better than 1ms resolution? That would be plenty. I'm not sure what the resolution is on mine, but I'll try to find out. EDIT - Couldn't find any data on resolution, except for one review for a Competition Electronics Pocket Pro II timer, which "measures near 1 ms..." Great info on specs in this article/review on it: https://www.usacarry.com/pocket-pro-2-shot-timer-review/ Edited January 20, 2019 by 98Z5V Quote Link to comment Share on other sites More sharing options...
Lane Posted January 20, 2019 Author Report Share Posted January 20, 2019 14 minutes ago, 98Z5V said: "measures near 1 ms..." Sounds good; I'll try to land well better than that. Quote Link to comment Share on other sites More sharing options...
Lane Posted January 22, 2019 Author Report Share Posted January 22, 2019 I'm not sure if I'll have this finished by the time the show opens or not. I do get a few bonus hours since it's in a different time zone. I ran into a fair number of glitches trying to get everything running flawlessly. A few strange ones where it would crash after five, or thirty minutes of run time. That stuff is all resolved once again; but adding the rotary encoder hardware made it crash too. What do you know, the internal pullup resistors are apparently too weak. Now; even with encoder/button interrupts enabled, the shot timing code is still stable. I think everything else is cosmetic from here. A bit of parsing data in the shot timer programs, and menu code is pretty much all that's left. That and turning off all my debug code for speed and elegance. I had forgotten there wasn't mail on Monday; still missing a few minor pieces of hardware, the loud buzzers, and LiPo charging boards. Quote Link to comment Share on other sites More sharing options...
Lane Posted January 23, 2019 Author Report Share Posted January 23, 2019 I tried; but got hung up on some code here and there. The rest of the parts either showed up in the mail today or were harvested from other electronics. Two blue boards and one green board are lithium battery chargers with and without protection, and a 3.3v DC/DC converter. The two red boards were lithium battery alarms. I broke the first alarm sounder trying to remove it, but I cut the second one off the back of the board, thereby ruining the circuit board but saving the alarm. This type of alarm is much louder and still frequency tunable; so that's settled. I had actually ordered 10 of those parts last week; but the order got canceled on Sunday because of low stock. Pretty much the only urgent matter left in some code is the menu system; which is the main user interface on the display. The software needs to differentiate which routine is running at the time; and process the rotary encoder and button presses accordingly. Then to build some portable battery powered units for some more real world testing... Quote Link to comment Share on other sites More sharing options...
Lane Posted January 23, 2019 Author Report Share Posted January 23, 2019 On 1/20/2019 at 10:39 AM, 98Z5V said: It clears all shot strings when it's turned off (or goes off) I'm not totally sure what you mean by this... My shot timer saves every shot string up to the point the flash memory is full. Let it time out, or turn it off; take it home. You can also log into the web interface over WiFi and download the data at any time; even while you are shooting if necessary (that's what I did the other day from my phone when it was cold outside). Not much of an update here; but the menus finally animate the way they were designed to. Every thing I have left is minutia; adding or deleting a single line or two of code here and there. Soldering connectors on the parts, and headers on the boards. Quote Link to comment Share on other sites More sharing options...
Lane Posted January 24, 2019 Author Report Share Posted January 24, 2019 A bit out of focus; but I'm preparing for the two "new" hardware builds. I think my original prototype will be assembled as a third build actually, but all the dimensions are different because of the detached screen module. Just a few more connectors to prepare, and then I can solder it all together. The first enclosure has all the parts inside, but the top rotary encoder board needs to be remade and replaced. It's just too tight against the microcontroller board's usb port which needs to have access for the user (I just don't want that to break off because of my bad design). Not a problem, I have plenty of those parts and perf board laying around. Quote Link to comment Share on other sites More sharing options...
98Z5V Posted January 24, 2019 Report Share Posted January 24, 2019 On 1/22/2019 at 11:20 PM, Lane said: I'm not totally sure what you mean by this... It doesn't write anything to memory, that you can recall or review later; only the lst recorded shot string is what's viewable. It only keeps it's last recorded shot string in memory, and that will clear from memory as soon as you start (GO button) a new shot series - or it clears everything when you shut it off, or it shuts itself off from inactivity. Nothing is written to memory, in any permanent fashion or manner. Only thing stored (be it temporarily) is the last string it recorded. Quote Link to comment Share on other sites More sharing options...
Lane Posted January 24, 2019 Author Report Share Posted January 24, 2019 13 hours ago, 98Z5V said: It only keeps it's last recorded shot string in memory, and that will clear from memory as soon as you start (GO button) a new shot series - or it clears everything when you shut it off, or it shuts itself off from inactivity. Thanks for the clarification. Up until recently having actual storage on a microcontroller was unheard of. They could have added an external flash chip for storage, but it sounds like a design built from 1980s technology when that wasn't possible. That is what's nice about this particular board; it has flash storage (built in, and more on an external chip), as well as built in WiFi for less than $3.00. With the attached display it's only $6.50 in single quantities. At these prices one could even employ a second dumb microcontroller to do all the heavy lifting, and still come in at less than $10 for the whole setup. Quote Link to comment Share on other sites More sharing options...
unforgiven Posted January 24, 2019 Report Share Posted January 24, 2019 That is very cool. My limited experience in the mill I became an electronic repair man apprentice. We were studying transistors, and gates, or gates ??? That enabled me to work for Xerox troubleshooting/ repairing mid volume copiers. Night and day difference. Quote Link to comment Share on other sites More sharing options...
Lane Posted January 25, 2019 Author Report Share Posted January 25, 2019 5 hours ago, unforgiven said: We were studying transistors, and gates, or gates ??? The funny thing here to me is; what was old is new again. These days if you want extreme performance you can use chips like Field Programmable Gate Arrays (FPGA), or Complex Programmable Logic Devices (CPLD). In both cases a huge array of individual logic gates are configured to achieve the desired functional output. One can load an entire CPU in the array and still have room left to drive a display with the hardware, and whatever else is necessary. I took apart a car radio a while back that was little more than two chips (one was an FPGA), and an LCD touchscreen. Quote Link to comment Share on other sites More sharing options...
Lane Posted January 25, 2019 Author Report Share Posted January 25, 2019 Testing out the first hardware build and realized I forgot the pullup resistors for the rotary encoder. It only moves twice before it stops. I checked the wiring and it doesn't appear there is a short or anything. I'm going to solder the resistors to the back of that carrier board and hope for the best. The only other thing I have yet to figure out is how to attach the tiny protection diode for the input to protect the ADC. I will also need to make a few splits off the power and ground cables to get everything connected; but the lithium charger is already working along with battery powering the device. Quote Link to comment Share on other sites More sharing options...
Lane Posted January 26, 2019 Author Report Share Posted January 26, 2019 No external pullup resistor needed. After my last post I got to thinking about the problem in more depth. Seems like I might have had solderless breadboard issues with my earlier build. This problem was only on one of the two pins. I played around with the internal pullup resistors and finally resoldered my whole board. It was a loose pin (or five). Everything is solid and I can begin to assemble the final unit now. Only two external components for the audio input circuit, and two more for the buzzer as needed. Quote Link to comment Share on other sites More sharing options...
Lane Posted January 27, 2019 Author Report Share Posted January 27, 2019 After much fumbling around with the parts and wiring the unit is self contained and running on its own power. I only got one of the two batteries inside it so far; but that should get better as the design is refined. At this point I can actually carry the unit around and test it outdoors freely. It seems like the plastic enclosure is transmitting the vibrations well enough; but I haven't tried a new live fire test. It might be necessary to punch a hole to make it work better; but I won't know until I try with some quiet ammo. I still have a few minor additions to test out before I begin the second hardware build; hopefully building on what I learned this first time around. The wiring in particular needs to be done carefully and with respect to length and routing. Quality wire also makes a world of difference when trying to cram things into a tight space. That alone will make up the extra space for the second battery. I really should have laid out the design first and then cut each wire to length; but at least this current layout doesn't seem to have introduced any appreciable noise (I was most concerned about the power supply). One of the front panel buttons needs to be wired between the Wake and Reset pins to allow the user to wake from the low power sleep mode. At the moment it doesn't have a hardware power switch either so it's always on until the battery dies. It should be as simple as calling to the sleep command, and attaching the two wires for the wake circuit. I'll start testing the wake function on the breadboard build, and laying out my next hardware build as I work the kinks out of this one. I still have to harvest one more of those loud buzzers from a lipo alarm. I found a pin vise though, so that should make quick work of it. One last thing is to run through my code and make sure the front panel LED is OFF most of the time. I find it annoying to see it on inside the case. Quote Link to comment Share on other sites More sharing options...
98Z5V Posted January 27, 2019 Report Share Posted January 27, 2019 This is amazing, watching all this go down... Fucking BADASS!!! Quote Link to comment Share on other sites More sharing options...
Lane Posted January 28, 2019 Author Report Share Posted January 28, 2019 15 hours ago, 98Z5V said: This is amazing, watching all this go down... I'm having a good time with this project. As you may have noticed I didn't study commercial products too deeply before beginning my own build. I figured that it would be better to explore the limits of the hardware; than to get distracted by how someone else's timer functions early on. Spent a little while preparing brass for reloading while thinking about what's next on this shot timer project. I put the battery powered unit on a demo loop to see how long the single battery lasts without turning it off; I got 2.5 hours running everything. I harvested the second buzzer and tested them both, with and without transistors. Much louder with the transistor, so I'm going to have to figure out how to wire one in without making a mess. Return from the deep sleep function only requires pulling the reset pin low; and was a single line of code to implement. The reset button can be actuated by pressing on the case above the screen. Turns out the battery dying puts the unit in to deep sleep all by itself. Took me a while to figure that out when the unit wouldn't turn back on even after being fully charged. Started assembling the second unit as carefully as possible, but the plastic box shattered in a few places when I was trimming it. That; and the different rotary encoder position revealed that the FLASH button on the back of the board was usable through the case by pressing on the screen in the bottom right corner. Unfortunately that switch, and the encoder button both fail to trigger the ISR. I have no idea why that is; but it may be another bug in the older libraries I'm using. I can't find anything informative about the problem yet other than a bug report closed because of no input. I've tested numerous boards and checked the circuits with a meter. Debug code even revealed that the circuits are fully functional, and changing the pin status if read manually. I likely have enough clock cycles left to read the pins in the main loop if I absolutely need to go that route (with no functional interrupts); but the rotary encoder fires the interrupts just fine. There is already an animation loop, so this could offset that in terms of timing if necessary. I set up a second battery test with the board running at 160 MHz; and that bright blue LED is finally off. I'll let that go for a while while I work on other software issues. I'm sure the ISR thing will work itself out pretty easily. My current code is well below 1ms in resolution; but I don't know how low exactly. It literally reads a 0 in that field on screen ,so I will have to check it a different way. I realized I need a facility to rename the shot time data file to keep things running smoothly. I knew that the shot data file growing too large would cause problems. This would allow for an onboard archive to ensure maximum performance and use all of the available data storage. This really doesn't happen often; but I've been pushing those limits on purpose. If you download and clear shot data every few months (years?), you'll never have to move/rename a shot file. Looks like firmware can be uploaded via Android devices with an appropriate USB OTG cable (or a computer). I haven't tried it yet, but might as well see how that goes. Reports are that it is a very slow process, but works reliably. Quote Link to comment Share on other sites More sharing options...
Lane Posted January 28, 2019 Author Report Share Posted January 28, 2019 58 minutes ago, Lane said: 160 MHz This thing has long surpassed the run time on 80 MHz test builds... I had read somewhere a while back that running 160MHz uses less power; but this blew my mind. Double the speed for less battery consumption? Less powder, for a heavier bullet? Quote Link to comment Share on other sites More sharing options...
98Z5V Posted January 28, 2019 Report Share Posted January 28, 2019 Weirder things have happened, brother - run it and see what's up! Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.