Registration Notice

Due to increased spam, forum registration must be manually approved by a moderator! Please see this post for instructions.

*

Offline Lotuswins

  • 155
    • View Profile
  • Roseburg, oregon
Re: Testing TCB Firmware with Stock Arduino Mega
« Reply #195 on: April 29, 2020, 12:40:28 PM »
Hi Luke,

Thanks for answering.  My shield does not use the 5v output from the Arduino reg, but the digital pins do output a 5v signal, so too many outputs for the little reg they installed?  However unlikely that is, I did rob an old reg chip off a parts arduino and both new ones are now working fine (it is an 800ma output also), so as you surmised, it is probably the chinese made issue.

Oh well, I've ordered from another source so we will see if they will be more reliable.  I have two Robotdyn ones functioning okay, from older manufacture, and these two new ones failed in one week.....only discernible difference (to my humble observation skills) are the 5v reg stampings.

Jerry


*

Offline Lotuswins

  • 155
    • View Profile
  • Roseburg, oregon
Re: Testing TCB Firmware with Stock Arduino Mega
« Reply #196 on: July 12, 2020, 02:41:46 PM »
Hi Luke,

I'm just about finished with my shield, but am having difficulty with the airsoft.  I measured 1.7 amp max on my DVM for the airsoft unit, and I don't think the BC337 is up to the task since it won't even start the motor running.  BC337 is rated at only 0.8 amps, and the surge must be too much on motor start.

Would you have a recommended TO92 replacement in mind?

*
FatFingersOPTCBRev2.pdf
(45.81 kB ~ Downloads: 65)

*

Offline Lotuswins

  • 155
    • View Profile
  • Roseburg, oregon
Re: Testing TCB Firmware with Stock Arduino Mega
« Reply #197 on: July 13, 2020, 01:48:38 PM »
Hi Luke,

I just ordered these up to test...should work?  I hope not to have to modify the boards/submit a new board request since it took 3 months last time from Seed  :'(

jerry
*
ZTX650.pdf
(67.43 kB ~ Downloads: 71)

*

Offline LukeZ

  • 1014
    • View Profile
  • France
Re: Testing TCB Firmware with Stock Arduino Mega
« Reply #198 on: July 13, 2020, 04:47:48 PM »
Hi Jerry, sorry for the slow reply. The ZTX651 looks like it should work fine, though I note the order of the pins appears to be reversed from the BC337, so you will just want to double check and probably flip it around when you install it.

Also, to get the full 2 amps out of that transistor you will want to reduce the base resistor to around 200 ohms, this is R4 on your schematic.

Let us know how it works!
NO SUPPORT THROUGH PM - Read why
Need a forum account? Read here
Open Panzer FAQs

*

Offline Lotuswins

  • 155
    • View Profile
  • Roseburg, oregon
Re: Testing TCB Firmware with Stock Arduino Mega
« Reply #199 on: July 13, 2020, 09:22:47 PM »
Thank you Luke, especially for the advice on the bias resistor. 

I hope to finalize everything soon if this works, all else seems okay.   I plan to share the eagle/gerber files along with BOM and details on assembly.  It should be a much easier build than the original, but same footprint and connectors though without some of the frills which I have never used in any of my 12 tanks that have OPTCBs.....

Anyway, hopefully it will help your awesome design continue on.  I cannot thank you enough.   :)

Jerry

*

Offline Lotuswins

  • 155
    • View Profile
  • Roseburg, oregon
Re: Testing TCB Firmware with Stock Arduino Mega
« Reply #200 on: July 25, 2020, 03:10:18 PM »
Hi Luke,

Well I think I'm done with my shield.  I called it 'FatFingers' since us old folks can't seem to manipulate the SMD components too easily. 

I've included the Gerber zip file, and in the Here is the FatFingers V2 boards zip is the PDF of the schematic, the BOM in excel, an assembly writeup in Word, and the Eagle files.

I hope this helps someone who wants a TCB but can't manage the SMD and flow soldering.  Let me know if you have any comments?  I've assembled 4 of these so far, with 6 more to go.  Takes me about 3-4 hours for one, but doing multiple takes less time per board.

Jerry
*
Here is the FatFingerV2 boards.zip
(859.37 kB ~ Downloads: 64)
*
FatFingersOPTCBRev2_2020-07-25.zip
(387.71 kB ~ Downloads: 62)

*

Offline LukeZ

  • 1014
    • View Profile
  • France
Re: Testing TCB Firmware with Stock Arduino Mega
« Reply #201 on: July 26, 2020, 01:43:40 PM »
Jerry this is really great, I read through your write-up and examined the board in Eagle. I can tell you put a lot of time and thought into this, I was surprised to see you even kept the same mounting hole dimensions and connector layout! Thanks for being so thorough in the documentation and creating the bill of material, etc...

I have created a new section on the Downloads page called "Member Designs" and yours is the first design to be posted, and I've included links to the various board files and resources. I also put a link to your board on OSH Park under the OP account.

I noticed you mentioned trouble getting boards of adequate thickness from OSH Park, but although they do offer a thinner option, I believe the standard selection is 1.6mm thick which is what you recommend and which I agree is better. I did a test order at OSH Park and it looks like it comes to about $40 US for 3 boards, which seems reasonable. However SEEED is also a good place to order them so that is fine as well.

This may have already been mentioned earlier in the thread, if so forgive me, but did you ever find any other Arduino Mega boards besides the RobotDyn that uses the same footprint? Jürgen mentioned the generic Chinese "Mega2560 Pro Mini R3" boards which look very similar but I can't tell for sure if they are actually the same footprint or would be compatible with your shield. Anyway, if there are others you know of that would work let me know and I'll list them in the comments.

Over the years people have built quite a few different kinds of shields and partial shields, or assembled mostly functional TCB substitutes from various parts, but this one I think is the most complete project so far. It maintains all the functionality of the original board, can be built without advanced soldering techniques, and is well documented. I really do hope you get to experience the enjoyment of seeing someone else build and run your design!
NO SUPPORT THROUGH PM - Read why
Need a forum account? Read here
Open Panzer FAQs

*

Offline Lotuswins

  • 155
    • View Profile
  • Roseburg, oregon
Re: Testing TCB Firmware with Stock Arduino Mega
« Reply #202 on: July 27, 2020, 11:29:42 AM »
Hi Luke,

Thank you so much for the kind words, especially since mine is obviously an amateur effort.  One doesn't appreciate what effort goes into a boards design until one attempts his own!!  Your layout is really a piece of art!!  I can't imagine how long it must have taken to get all those bits and pieces to align and fit on a 2 layer board, it boggles the mind.

The Arduino I've located to replace the Robotdyn that works is from ebay: https://www.ebay.com/itm/274262551130.  I have two operating, so far so good.  I hope Robotdyn has fixed the 5v regulator issue, as I fear the ebay source will eventually dry up/change.... I will try again at Robotdyn, in the near future.  I could not find another 'commercial' outlet for the board, that would be a better source?

I think the OSH boards are better quality than the Seed version.  I was replacing that pesky transistor Q7, and the pads wanted to distort/lift on me with the Seed PCB.  The OSH pads seem to be better anchored to the substrate.  The reason I got the thinner board from OSH is that they don't offer the 1.6mm board and 2 oz. copper.  I have a good friend who I went to school with that went into the electronics field, and he recommended I use 2oz copper.  I'm not sure how 1 oz would work?  I really didn't do any current calcs on the board, simply went with as large a trace that would fit in the confines for other than signals (smoker, airsoft, etc.).   Also, the purple is a nice relief from the standard green color  8)

I feel honored that I'm the first on the Members Design!!! thank you.....Jerry


*

Offline LukeZ

  • 1014
    • View Profile
  • France
Re: Testing TCB Firmware with Stock Arduino Mega
« Reply #203 on: July 29, 2020, 09:35:42 AM »
Thanks for the link to the eBay boards, and on closer examination it looks to me that the pro mini boards Jürgen mentioned will also work, there are minor differences but overall it's the same footprint. I made a note about these in the description on the Downloads page.

That is a good point about copper thickness, you're right OSH only offers thicker copper on their thinner board for some reason. I did a rough calculation for the smoker portion of your board and even with the 32 mil trace width you use, and even at 2 oz copper, the maximum calculated limit of those traces is only 1.7 amps. At 1 oz copper it drops to about 1 amp. I don't recall exactly how much current the generic Taigen/Heng Long smokers pull, the heating element itself was about 1/2 amp but then there is the current of the motor as well. (Actually it seems you measured one at 2.1 amps once)

It highlights yet another of the many aspects of board design, aside from merely squeezing everything into a finite area. This is why we often use planes rather than only traces for high current signals, or polygons if we don't have the luxury of covering the majority of a board layer with copper. Another method is to run double traces, one on top of the board and one on bottom, this effectively doubles your copper thickness without increasing the trace width, though of course you need to have a clear run on both the top and bottom for that to work. Component placement also helps, we can try to locate high current outputs/transistors as close as possible to the incoming power source, in order to keep trace length short. We used all these and more on the original TCB, which was produced using only 1 oz copper. However I will say, that although more difficult to assemble, an SMD board does make life easier in other ways, for one we need fewer traces on the bottom of the board, and fewer interruptions generally on the bottom due to fewer through-holes, so this makes it easier to to have large power planes on the bottom layer.

In your case I rather doubt it will be an issue, but you can keep an eye on your boards for signs of overheating. There is still a little room to increase power handling on your design but I'd just use it and see what empirical evidence tells you before feeling like you should launch into an overhaul.
NO SUPPORT THROUGH PM - Read why
Need a forum account? Read here
Open Panzer FAQs

*

Offline Lotuswins

  • 155
    • View Profile
  • Roseburg, oregon
Re: Testing TCB Firmware with Stock Arduino Mega
« Reply #204 on: July 31, 2020, 01:09:05 PM »
Hi Luke,

2.1 amps....back in 18......must have been an abused smoker  8)

I seem to have melted a few down too..... :-\

So I pulled out two new, one used Taigen V2 smokers out of the spares bin and messed around for a little while. 

I hooked up my FatFingersV2 board, and put the Fluke in series measuring amps in the AVG (average) mode.  At 80 percent setting at idle, I got 0.56 to 0.60 amps on all three smokers.  It goes up a bit when increasing the throttle to 100% as set up on the board, but not much.

So I then plugged the smokers directly into the battery at 7.89volts unloaded 6 cell NiMh.  This yielded 0.73 to 0.80 amps.   

A curious thing is happening though when you let the smoker run for a while.... The current reading, starting at 0.6 amps, would slowly rise.   At roughly the 2 minute mark, the amps had increased a little over 10%.  The front of the smoker got hot, could hardly touch it....I wonder if my previous readings were maximums with a very warm smoker?  Perhaps.....

So the resistor in the stock Taigen V2 units is a NiCr wire, wrapped around a fiberglass or silicone fabric cylinder, right?  Small, but effective..  I've seen them, after much use, turn all black, burnt to a crisp.  I wonder if the heating action eventually turns the insulator to carbon, reducing the resistance causing even more current to flow?  Right now, I've given up on the smokers, just replace them as they fail since they are cheap.   HL now has a newer version though, that has an improved resistor configuration (V6.0 I think) which has the wire wrapped piece inside an insulating tube, which wicks up the fluid.  The resistor is smaller than the Taigen version, and instead of a wad of cotton to wick the fluid up, they use a little sponge.  I have one in my JS2, and so far it hasn't failed me.  I haven't done a current read on it yet, but the resistance is in the same ballpark as the Taigen IIRC.

Thanks for doing the current calcs, at least it looks like the board will survive a stock smoker.  No temperature rise on the traces was detectable after the 2 minute run above.


*

Offline LukeZ

  • 1014
    • View Profile
  • France
Re: Testing TCB Firmware with Stock Arduino Mega
« Reply #205 on: July 31, 2020, 03:53:31 PM »
Thanks for measuring that Jerry, those numbers are much more reasonable thankfully! That sounds closer to what I would have expected, and well within the tolerances of your board even with 1 oz copper which is good to know. Who knows what you must have done to that poor old smoker to have it draw over 2 amps! But these things are cheap and who knows what can happen. Maybe the little air-pump got gummed up with old fluid.

I am not surprised the current rises slightly over time as the smoker gets hot, there is definitely a relationship between the resistance of a conductor and its temperature, which principle as you know is used in reverse with thermistors to measure temperature! I've only taken apart a handful of older version Heng Long smokers and I thought they just used an actual resistor, but I would not be surprised if some of the newer ones do use nichrome wire as you have found.

It is hard for me to keep up with all the continuous developments but it seems Heng Long has really been upping their game. I notice in pictures their new smokers drive the heating element and fan separately, which is a much better arrangement (we added that feature to the TCB last year). I've also seen pictures of their new differential transmission similar to the DKLM gearbox, but haven't seen it in action. I have one of the DKLM geardrives and I really like it, but I don't know if Heng Long plan to release versions of theirs compatible with all their old models or if it will only fit in certain newer ones.
NO SUPPORT THROUGH PM - Read why
Need a forum account? Read here
Open Panzer FAQs