Another thought: You could probably design a single front panel that supports JST, dupont, and PTH.
I am happy to accept pull requests to a new “kicad” subdirectory to the joulescope_front_panels repo. If want to do the design but not mess with git, send the design to me and I can commit for you (with credit of course!). I can also create links in the projects to your own repos/design if you would prefer to keep full control.
Yeah, I’m unlikely to go for option 2 or 3! Mainly the PD (cautionary note taken - and thanks for being up front about what’s possible if people want to push the boundaries) and basic HS lines.
I like @Philip’s point about just giving the 5.08mm pitch holes of suitable diameter and a bit of space and then people can add the terminal blocks they want to. If the hole is larger, you can always “fill it” with solder for smaller terminal block legs.
I second the phoenix style connector.
I was thinking a couple of days ago that I’d like to make a front panel with a linear regulator and a knob for setting the voltage (and maybe a disconnect switch so that you can set the voltage prior to attaching the DUT). You can use the readout through the software to set the voltage, then flip the switch to attach the DUT.
Why would I want this rather than just using one of my fancy bench supplies? Well, I haven’t started the thread on it yet but I’m seeing some weird behavior that I think is related either to the line inductance or possibly to the frequency response of my bench supplies. It’s possible that it’s neither, I haven’t worked it out yet, but it would be nice to eliminate that variable with a solid local regulator. It might be doubly nice to make it a 2 quadrant, because that may also be part of my problem. As a quick note, the problem I’m seeing is an inappropriate range change to the -2A range with a buck converter switching event, causing a ~-2uC glitch in my measurements that swamps my signal (my device is a BLE device with an average draw of ~10uA, confirmed through other reliable means, and I’m getting 40 of these -2uC glitches per second, so rather than reading 10uA I read -70uA, which is impossible anyway since my device doesn’t generate power from the aether).
@mliberty here comes the follow-up product! I can see a stacked-PCB mini power supply that you bolt-on sitting nicely in your product line-up. You don’t even need a display as that’s all on the PC GUI. Could Joulescope do hard on-off current limiting?
I do plan to implement a programmable soft-fuse feature to the existing JS110 Joulescope. The hardware is just waiting for FPGA and software!
Hi @NickLL, if you turn off Joulescope autoranging and set it to 180 uA, does the measurement seem right? You mentioned -2A, so is this related to Possible autoranging with negative microamps issue (and note about imbalance) ? Does the problem still exist with 0.5.0 and the firmware update?
Wait, does this mean existing HW just needs FW & bitstream or does it need additional FPGA HW?
Bitstream! So, we are waiting on Verilog, C and Python. I haven’t figured out how to add an FPGA on shipped products 
-2A is probably -1.2A or whatever it was (I’m going from memory right now). This isn’t related to the other issue and it does persist in 0.5.0. I haven’t had a chance to test the negative microamps issue again with 0.5.0.
And, yes, when I switched to the 180uA range the negative spike disappears and the numbers are more reasonable, however the resolution at 180uA seems to be limiting. My device spends almost all of its time drawing ~700nA.
I do plan to dig in a bit more on this issue and will make a new thread when I have more useful quantification of the problem.
Come on - FPGA on the front panel standard!
@simonmerrett - I just updated the joulescope_front_panel GitHub repo with an improved fp_baseline.pdf and PDF schematics and layout for fp_usb. Let me know if any info is still missing!
@NickLL @Philip how’s this for a starter for 10? Terminal block of your choice on 5.08mm / 0.2" pitch and two pitches of JST connector for lithium battery powered setups. What would you add/remove/change?
Or there’s this, with 0.1" pin headers and a clear dividing line between input and output connectors:
Looks like a nice start. My suggestion is to align everything a bit better.
For example, you have pin 1 as the ‘-’ on the IN 0.2" header and pin 1 as the ‘+’ on the OUT 0.2" header. I understand that these match the connector well and if you were to swap it then one of them would be facing inward if the connector was installed correctly. My suggestion is to simply rotate them both 90 degrees. If your concern is that the wires coming out the top or bottom would be ugly, I’d suggest that they’d also be ugly coming out the left and right and that the best solution to that is to use vertical headers (like https://www.digikey.com/product-detail/en/phoenix-contact/1755736/277-1150-ND/260518).
I would also suggest aligning all of the other connectors and making them all horizontally oriented (the 0.100" headers too) and horizontally aligned with each other, and all more toward the middle than the LED. It might look a little more cramped but that’s where the BNCs are placed and that looks fine. If you move all of the silkscreen labels to the side then you should have room to stack all 4 connectors on the left of the line, I think.
Anyway, those are my suggestions.
Thanks @NickLL, I get your point about alignment but I suppose I was mainly thinking about being able to grasp the connectors properly for removal, so tried to leave as much access around the sides as possible.
Here’s a vertically aligned version.
I’m not sure I agree about making the terminal block openings face up, rather than outwards but I’m glad to have your opinion. Hopefully others will voice theirs. It seems that if you populate both JSTs (optional if you don’t envisage using one of them) that a properly aligned vertical or horizonal layout would hinder fingers on one side or another, at least for the shallower of the two shrouds.Edit - this makes me think we ought to have the shallower 2.0mm JST on top…
If you horizontally orient and horizontally align them all in one column, wouldn’t you still have better finger access to all of them by grasping the sides?
Also, I think you misunderstand what I mean by “vertical”. I mean a vertical mount connector rather than a right-angle connector. The link in my last reply should clarify. A “vertical” phoenix style connector would have wires coming out of the screen relative to your image, in the same way that every other joulescope front panel connector works, and the way the JST and 0.100" header connectors on your panel would also work.
I’m not sure that’s better than having access to three or four sides. My limited experience of JSTs is that they can be quite stiff.
Sorry, I didn’t misunderstand with that layout, I just wanted to see what it would look like. Here’s something closer to what I think you meant:
@mliberty you’ve been thinking about this HW for waaay longer than any of us. What are your thoughts on wire/connector layouts and orientations.
I don’t have any strong preferences when it comes to front panel cable “flow” since that depends upon the application. For the front panels to date, I have tried to keep cables coming out horizontally either out the sides (USB front panel) or out the front (all other front panels). Many terminal blocks are at 45° diagonal or up/down, so keep that in mind. Also locate the terminal block so that physically larger terminal blocks components do not interfere with the light pipe. For example, the 277-6722-ND previously requested by @Philip appears bigger than the footprint (did not check actual dimensions though).
Another consideration: for lowest voltage drop, especially for the higher current-capacity connectors, you want to keep distances as short as possible.
Yep, that’s a beast! Not sure it would be able to share with all the others, although perhaps you could use the same PCB and have “either or” with the big one or the range of smaller ones. To accommodate that connector, the 5.08mm holes would probably only support a sideways cable orientation.
@Philip are you a Kicad user and would you be willing to knock up a footprint for that connector you referenced?
I use Altium, and I am currently vacationing, so can’t do any CAE work. The connector I referenced isn’t one I specifically love, it was just a starting point that showed that Phoenix has a multiple parts with the same footprint, and different ways of holding wires, including a lever type.
