Is there enough room for those to be right angle pins? It looks like the module pins are in the way.
And where’d gpio 16 go?
There is a lot of room.
We lost an output to the gpio.0 pin. But since I don’t think many if any used all 4 we now have three more useful selectable outputs.
1 Like
I’ve made thousands of banana-to-ring adapters over the years. They’re literally the most “borrowed” item in my test kits.
I’m about to make up another batch.
I’d blow people’s minds if I showed what my breakout boxes and breakout cables look like.
same experience here.
I like them too, but the $Limits is pretty good.
I think there needs to be a specific section of the documentation describing how to test and diagnose endstop issues.
I think the only bummer is losing those filament-as-fiber-optic indicators on the gantry.
Can you leave an SMT pad near the diag pin on the socket?
I could, but it is a pretty tight spot, right in the mix of all the driver stuff. It anything It would be a via to the bottom but that seems less useful.
That would work fine.
To bring up the elephant on the board: I think this should have an embedded ESP-32 with the PCB antenna, with the antenna hanging out from the outline of the jackpot PCB.
I think as shown with the external antenna, it will be an unmitigated disaster. Too fiddly and prone to user errors and problems.
The current jackpot puts the antenna in the center of the board, with the jackpot pcb just under the socketed module. That isn’t ideal either, but is because the usb connector on the ESP-32 needs to face the outside of the board.
Going to an integrated ESP-32, why not try to use it as it was intended with the antenna protruding just beyond the rest of the board?
Also, just to say it- you’re begging for regulatory attention if you package antennas and esp-32 separately.
I’m struggling to see how that’s likely to be the case? Those connectors are small but they snap on in quite a positive way and then they have good retention. The main issues are usually trying to unsnap them…
This would be bettter but it’s still remarkable how much more garbage the chip antenna is vs even a basic external whip. At my place that’s the difference between -88dBm and high packet loss to -60dBm and 100% reliability for one module. I don’t know that it’s ‘that’ important when you can run an AP nearby, but then the whole station-only approach seems mildly distasteful to me anyway so I’m probably in quite a different boat.
I don’t think that’s the case for selling individual items like that. This isn’t a standalone product, it’s a component. I’ve never heard of anyone getting pulled up on things like that with bare boards.
I have no issues with that at all!
I am actually worried about vac hose static and that antenna, actually. On the LR4 that puts them in the same proximity.
I trust both your opinions very much. What are the rest of you thinking?
I could offer both…but we lose out on bulk cost savings that way, in a fairly significant manner.
Built in antenna will be less expensive and less work, with possibly a better signal…but the built in will be significantly better than the V1 jackpot because it will not have a populated PCB under it and can hang out of the case if so desired. much better position.
We have every part of the user spectrum in our community. We have folks who could design their own controller or write firmware from scratch, and we have folks who struggle to even plug in VMOT power.
We have folks with young eyes and good dexterity, and we have folks with older eyes or fat fingers.
We shove all this stuff into a self-printed enclosure and struggle to keep rats nests of wiring clear of any antenna. Then we shove all that into an X axis beam and surround that with an enormous ESD source (vaccum hose) and electrical noise source (spindle).
Every single one is a unique custom build. Every antenna gets stuck by the builder in some different random spot.
As for the antenna, the little external bowtie these would come with is an improvement overall vs the stock PCB antenna, but isn’t nearly as good as you say- getting even a little stubby monopole to mount on the outside of the enclosure.
I’m not super convinced the bowtie that much better than just moving the ESP-32 to the edge of the PCB and giving the stock antenna good line of sight to the outside world.
I’m curious what the less EE crowd thinks about this. We (you, me) aren’t the group I’m worried about having trouble with the external antenna. If this gets done it needs really clear instructions and maybe a good standard enclosure to install it in .
1 Like
I think the entire point of this is that everything is different so the external antenna is exactly why it makes sense. It gives an axis for change and customisation. If you want to put the entire thing in an aluminium project box, go ham, just run a pigtail to a bulkhead RP-SMA connector. If you want to put the entire thing in the beam and then use ACM for the strut plates, go ham, just have a hole for the whip antenna to poke through or run a cable out to another one, etc.
I get that there needs to be concessions for all types of people but I struggle to see this as really any harder than wiring to those terminals that are already on there, fishing wires through the slots in the LR4 carriages or working with small screws. It could well be that I’ve done it enough that I can’t see the forest for the trees, but trying to figure out a litmus test like ‘could my fiance do this with written instructions and without damaging anything’ and I have no doubt that she could, or indeed anyone else I know who has made one of these things… And in the worst case, finding assistance from someone with younger eyes, using magnification, using some needle nose pliers etc. all improve the situation…
Regardless, I don’t think this has any clear answer and is just a matter of opinion. As long as they’re both present in the shop I’d say let one with sockets out in the world and see what happens, or maybe do a single batch of that one and see what the feedback is.
At the end of the day, I’m not personally worried because the first thing I’d do is hot-air the module off and replace it, no skin off my nose 
1 Like
I’m assuming usage of a plastic whip antenna and coax pigtail so I wouldn’t be particularly worried… That may be optimistic. I’ve never had issues with exposed whip antennas leading to module deaths and there are plenty of other consumer things with ESP32 modules and exposed antennas like that. There’s very little about this design that I would have called particularly static protected so it wouldn’t surprise me if the antenna were one of the more robust points of connection in that respect. I’d also say if there’s significant risk of static based issues like that, the answer is probably to explain the mitigations and encourage people to use them, not just try to harden the device against epic levels of static interaction…
I definitely get that this will be a heck of a lot better than the V1 Jackpot, that’s definitely a bonus but from my experience even with the PCB antenna being in a good spot they’re kinda just ‘ok’ at best. I typically hang the modules off the edge of the board as well and then enclose them such that the antenna is well clear of any cabling, mounting features etc. and I’ve still had some situations where they just don’t perform acceptably.
If you’re sticking with the whole ‘operate in station mode and run it from a nearby tablet’ approach then I doubt there’s much of an issue there. Proximity is very close and if you’re willing to stick with the ‘standard’ install being outside the beam then there’s really not ‘that’ much to complain about.
For me personally, it’s the things where I see people stuffing the entire controller into the beam and then some of the threads that go by where there are people getting weird behaviour or UI timeouts or whatever and I always wonder if that’s a byproduct of crap signal strength and excessive packet loss.
Like I just said above, I think external antennas are the better approach because they’re more flexible, but they’re definitely an added complexity. If you’re ok with the performance as it stands with the Jackpot then I don’t see a huge reason to change other than if that flexibility would be useful. Someone who is particularly bothered by this, like I would be, can just replace the module outright, either themselves or with assistance from someone else. That’s 10x harder than just plugging in a u.fl connector in the first place, but seems like a reasonable compromise for someone who wants to go the extra mile and hide their entire board in a beam with metallic struts or whatever.
1 Like
as someone who runs into connection issues regularly I would love to have an external antenna.
1 Like
For me, with an integrated ESP32, I tend to agree with @jono035 and think I would prefer the external antenna.
The signal I get from my current Jackpot is only marginally as good as I need it to be to connect to my wifi.
I haven’t done it yet, but was about to order an ESP32 with the external antenna connector to see if its better enough to keep my connection stable.
Even though it’s in the instructions, we have seen a lot of people have connection problems and bad experiences due to wiring running over the antenna, etc., even in AP mode. It feels like having the external antenna would mitigate some of those issues.
When I built my MP3DP v5, the Manta board with CB1 came with a crappy external antenna like this
I upgraded mine to one of these
and it made a massive difference in the connection strength that I had to my home network.
With most of our boards being enclosed, and mounted inside an enclosure, and under the dust hose… I’m really interested to see how much better my connection will be.
4 Likes
I personally would prefer the integrated antenna but I also have strong wifi in my shop. It’s just another part. I’d rather not have something sticking up but I guess I could use one of those flat antennas and keep it in the case. I’m not worried about breaking the connector.
I see that Bart uses the external antenna versions on most of his controllers.
I’m fine with whatever decision is made.
Dang…maybe I make two batches and see where people vote with their wallets, the external will be a few dollars more (in this case that is about 5%) with an included
.
I will look through LCSC to see what the bulk monopoles look like.
1 Like
Depending on which ESP32 you want to integrate, there are some with socket for external antenna already on them. You could just extend that to the side of the board to be exposed on the enclosure for anyone who would like to make use of it?
Also just leaving socket somewhere on the board will be easy enough for people in need of it to use… I think
I am not seeing one for the Wroom that we use. The wrover (that is not good for fluid) version has a solder pad that need to be bridged to use the external one, the rest are versions that are not completely fluidnc compatible yet.
If I missed one, shoot me the LCSC link. Here is my search, Search by "esp32-wroom-32"
Oh! I thought you want to integrate whole board like for instance esp32-sN and I have seen some on those having ext antenna socket… sorry - my mistake.
If tests show vac hose static is a concern, then, would be good to ensure doc strongly recommends grounding one end. Think Maslow docs/forum recommend this too.
Guessing existing/updated enclosure has room and ties to keep wires clear of whatever antenna we end up with.
Personally prefer whatever has most reliable strong signal strength. Time burned troubleshooting network connection issues is never fun, especially for devices without screens or other ways to interact.
@Michael_Melancon, cheers for the antenna recommendation. Don’t have WiFi repeater in my garage, but have many devices and multiple nearby networks, so, decent external will be neat for my situation.
Currently, those are not officially supported by fluidnc, when they are we will make a new board.
1 Like