Adding switches to devo 6s

There have been only two changes to this in the nightly builds: the original merge and the off-by-one bug fix.

I just realized that this extra pins can also be used for controlling a third extra RF module. I've got a CC2500 I've been wanting to swap into mine. So another project I can't do while my knee heals .

Is there a consensus on the labelling changes?

mwm wrote: I just realized that this extra pins can also be used for controlling a third extra RF module.

Hmm, got me thinking about having the physical switches but also parallel wiring for the RF module, as an either/or thing... which then posed the question of live hardware.ini editing via the TX itself, eg. hit a combo of buttons to bring up a checklist edit of the ini (after first disabling RF output), check or uncheck your desired settings then hit "reboot" to enable your new settings. So a sort of on-the-fly ability to swap between hardware configs that you omay have physically setup in parallel, ie. choosing between having the switches or having the extra RF module?

mwm wrote: Is there a consensus on the labelling changes?

Was the current code based on how the 8s is configured? If so, then I think yes keep it how it is and we document which pad does what, and I change my hardware layout (easy enough to do).
If it's not due to similarities with the 8s, was there any other rationale behind how it ended up that may need to be discussed before a final commit?

Hope the knee's on the mend

Well, the button matrix is now identical to the button matrix for the 8.

The two switches also use the 8's input, but the 7E's names.

But since they don't share code (at least not yet), we can easily swap them around if there's a reason to do so.

BTW, I just checked, and we can't use the two switch inputs as analog inputs. The pins don't allow that. I was hoping to coopt one into a slide switch on the side of the tx for jib control. Oh well.

Can someone confirm for me that the labelling on the most recent photo in this thread is correct? In particular, the switch labelling. I'm about to open up my 6S and solder in some headers, and want to try using the switch pins as the third module enable/disable pin. That should work with no further software mods if you're interested in getting four rf modules in a 6 without a multimodule. The 3-in-1 being discussed elsewhere would be particularly interesting for this.

Yep Deal57's last photo is correct, when looking at the PCB with the stencil text facing you (P19, HOV-T etc) "B" is on the left and "A" is on the right. Also the "right" trim is on the left, and the "left" trim is on the right. Takes some mental gymnastics but it makes sense when you mate it up to the back case.

Both trims now correctly read left-to-right as "positive-common-negative".

So yeah if the matrix is the same as the 8S I say we keep it how it has now been confirmed.

Finally received some actual momentary toggles for the trims; these ones . I'd purchased what were listed as identical ones previously that turned up as latching (like a 3-way), and when I raised the issue they promptly sent some double-pole momentaries, which are way too wide to fit in this instance. No guarantee that if you purchase the exact same one as I've listed that you'll receive the same, as no-one seems to check beforehand.
I'll post one day a rundown of my physical setup, but one thing to note is that these generic red-bodied switches can actually have quite a significant chunk of the red material whittled away with a sharp blade; I had to do this on one corner of the trims to squeeze everything in.

My plan is to use one of the switches to add a third rf module, and then add two honest to goddess trim switches for the trims. I want to mount them between the halves of the case so they land underneath my index fingers. The other switch I'll leave alone for now.

If I had an analog pin, I've got a slider to mount in place of one of the trim switches.

I'm using rocker switches for use as TRIM_L and TRIM_R ( www.ebay.com/itm/201331239029?_trksid=p2...e=STRK%3AMEBIDX%3AIT ). They'll fit nicely in the back shoulder, but I'm not sure how to tackle cutting a rectangular hole for them. Any suggestions?

i would use a round dril and then use a file to make the opening rectangular.
but probably there are better ways nowadays.

Those are my thoughts pretty much exactly. I'm hoping someone can suggest one of the "better ways".

I would try a rotary tool... use a drill-bit style cutter and put a die framing the area to cut. I would think it will still require a steady hand but there's less chance of drilling outside the line.

I used a rotary tool with a disc-style blade for this soft plastic once and made a right mess of it. Maybe a little practice first would have helped.

Well I haven't gotten around to (!) swapping out my trim toggles from latching to momentary, nor even swapping them to the correct shoulders, nor actually using any of the switches for a model yet... But just a note that I tried using a standard trim the other day and nothing was responding, turned out one of my latching trim toggles was on, and when in a held-on mode it stops other trims reacting! So just as a word of caution, you could theoretically have these new trims set up as additional three-way switches BUT when not centred they will negate your "normal" trims, or at the very least make them not respond as/when desired

Was going to start another thread but it relates so I'll ask here; what is the exact purpose of the IN4148 diodes in the 7E switch mod? I think I recall something about it somehwere along the line but none of my searches are turning up an answer.
Reason is regarding my previous post; for the button matrix, does adding the diodes isolate the lines or something, so they don't interfere with each other as I am seeing with the latching switches?

This kind of basic info should probably be included somewhere in the mods section of the wiki, a quick "why do we need to do x for y to work" explanation in one central location. I can add it if I can gather up the data.

PS. I have started using SW_A for launch mode for my mini DLG (discuss launch glider); as it's a latching 2-way, I've turned it into a momentary by holding it back with a rubber band! Once I get around to swapping out the extra trims for the momentaries I might move it over to those instead.

I think www.deviationtx.com/forum/how-to/1436-ad...tches-to-the-devo-7e would have been a better choice for a topic, but topic drift doesn't bother me much.

Anyway, you do know that deviation has support for momentary buttons in the mixer, right? You can use the various navigation buttons or the trim buttons for that.

RoGuE_StreaK wrote: Was going to start another thread but it relates so I'll ask here; what is the exact purpose of the IN4148 diodes in the 7E switch mod? I think I recall something about it somehwere along the line but none of my searches are turning up an answer.
Reason is regarding my previous post; for the button matrix, does adding the diodes isolate the lines or something, so they don't interfere with each other as I am seeing with the latching switches?

I used them for the Ultimate7E switch mod to prevent button matrix ghosting. Basically since the switches stay in position, we don't want false positives/negatives due to the circuit paths. Google for "button matrix ghosting" or "keyboard ghosting" for more information.

Mike, I put the topic here because it specifically applies to this mod; so far we have not mentioned or discussed the use of diodes for this mod, only wiring straight to switches. But what I found when using a latching switch on the trim line was that it certain other trims responding, so I'm querying whether we should list diodes as a requirement to stop this happening.

RE: momentary buttons, the requirement for a DLG is to have a shoulder-mounted momentary that you can pull-on with your index finger while holding the TX one-handed while bodily spinning while holding and throwing the DLG with the other hand / fingertips The switch is released a split-second after the plane is released. A modded 6S or 7E is a perfect accompaniment to such an activity due to it's compact easy-to-hold-one-handed nature.

Thanks slipstream, I believe that sheds some light on the nature of the button matrix and why the diodes should be introduced for any switches connected here. In essence, if there is no diode then backflow can occur and block the recognition of other buttons on the same matrix lines, but putting a diode in place blocks backflow so the buttons can be seen simultaneously?

The requirement for diodes should however only be applicable to the Trim switches, correct? ie. SW_A/SW_B have their own dedicated circuits separate from the button matrix so don't require additional hardware.

Sorry RS, I saw the "Devo 7E" comment and naturally pointed you there.

You can do use the trim switch extension on the 6S to get momentary shoulder buttons. I installed two-way rockers to get something that acted like real trim switches on mine. Using one of those should do the trick.

Yes, the 6S extra switches uses dedicated switches ala the 8/10, so doesn't need diodes. Given that using those gets you to the same set of switches as the 8 and 10, and treating button matrix inputs as real switches would complicate things quite a bit, I didn't see much point in that. Doing that on both sets would mean 4 more 3-way switches, though.

Hi Mike, I do have replacement momentary 3-ways [(on)-off-(on)], but I'd chucked in the latching ones for the original testing purposes and haven't had the time to swap them out, which will also involve resoldering most of the connectors etc. to make it a bit more professional. But because I used the latching ones I inadvertently came across the ghosting issue, so thought I'd raise it and the solution in case someone does decide to use the trims as extra 3-ways. Don't know what you'd do with that many 3-ways, but you never know.

Actually wouldn't it either give 3x 3-ways or 2x 3-ways and 2x 2-ways?
I never did get around to testing a physical 3-way on SW_A, so I don't think its current behaviour in the code has been confirmed by anyone to be accurate. By my logic it was wrong, but my logic has been known to be... illogical

Also just realised that I fell into the trap of calling silpstream "slipstream"

I was just looking at the switches from the trim buttons, which is 2 3-ways for each pair of buttons. Or lower. You can then get either another 3-way or a pair of 2-ways from the P19 positions. So the max EXTRA would be 5 3-ways or 4 3-ways and 2 2-ways. But again, there's no support for switches on the trim buttons.

I assumed we wouldn't need diodes, because I understand the 7E only needs them for the 2 three-way conversion. But I could misunderstand, or this could just be different.

You may be right about the 3-way on SWA. I tested it in the emulator, but that doesn't insure that the hardware is handled properly. The way it combines the two inputs from two 2-ways into a 3-way should work, but there could be a bug/typo/etc. in it. Unfortunately, I can't really test it now because I've got an nrf24L01 hanging off of one of those pins.

@RoGuE_StreaK, no worries on the name... lol happens to everyone including myself when auto-correct is on.

Anyhow, your understanding of the back flow is correct. I think diodes should work to prevent problems with other trims, but if the latched switch is not centred then the trim switch that is is connected to will give you problems still I think, but I've never tested it. How does a latched switch connected in parallel to the trims not cause the 6s to assume that the trims are being pressed and held instead? Won't this cause your trims to go to either extreme for that axis?