Oshpark Projects

1808 posts / 0 new
Last post
MRsDNF
MRsDNF's picture
Offline
Last seen: 1 month 2 weeks ago
Joined: 12/22/2011 - 21:18
Posts: 13473
Location: A light beam away from the missus in the land of Aus.

Thanks wight. I now understand the MCU sits in the middle hence the 20 pins. I was looking at the four on the outside edge.

 

djozz quotes, "it came with chinese lettering that is chinese to me".

                      "My man mousehole needs one too"

old4570 said "I'm not an expert , so don't suffer from any such technical restrictions".

Old-Lumens. Highly admired and cherished member of Budget Light Forum. 11.5.2011 - 20.12.16. RIP.

 

HarleyQuin
HarleyQuin's picture
Offline
Last seen: 6 months 3 weeks ago
Joined: 03/29/2013 - 04:47
Posts: 588
wight wrote:
What is our collective experience on OSH Park’s removing copper from the edges of the board? We are all aware that this is normally done. They require a 15mil setback for traces, but since our copper pours are not really traces it’s standard for us to specify copper right up to the edge of the board. Somewhere along the line a little copper is removed in order to prolong the life of the router bits used to cut the boards. How much are we talking about? I’ve just checked several boards using a feeler gauge and to me it looks like less than 0.2mm in all cases.

I measure a diameter of the GND-copper that is always 0.6mm smaller than the board. Got 3 very recent boards here (all ordered last November):
22mm board, 21.4mm copper
17.4mm board (I’m long since overshooting and filing down…), 16.8mm copper.
15.4mm board, 14.8mm copper.

It’s already difficult to solder the boards to some pills. With this added 0.3mm all around it’s nigh impossible.
Thus the workaround.

The 0.8mm PCB is a great option, by the way.
HQ

Oshpark Boards:
HQ ProgKey: Universal Driver Programming Key . Boost: HQ 15mm/17mm programmable boost driver with ATtiny13A
46mm Triple-Channel: BLF SRK FET v3 . 17mm Linear: HQ10D / HQ4D / HQ4S . Contact Boards: 22/24/26mm

wight
Offline
Last seen: 1 year 3 months ago
Joined: 11/27/2013 - 16:40
Posts: 4969
Location: Virginia, USA

Thanks HarleyQuin. My measurement of ~0.2mm was for only one edge. I did not measure the overall dimensions of any of the drivers. Still, I think the smallest feeler gauge in my set was 0.25mm and the missing material was less than that much for me. In any case our measurements are pretty similar: ~0.4mm overall and ~0.6mm overall.

Still fine, still on a break. One day I’ll catch up with you folks! previous wight catchup Wink
list of my drivers & variants (A17DD, FET+1 stuff, WIP stuff, etc)

Pablo E.
Pablo E.'s picture
Offline
Last seen: 4 years 3 months ago
Joined: 08/08/2015 - 13:54
Posts: 234
Location: Spain (GMT+1)
wight wrote:
The Zener mod requires two additional components: a low-value resistor (such as 200 ohm) and an appropriate zener (such as MMSZ5229B for example). The resistor should be placed between MCU Vcc and Vbat. The Zener should also be attached to MCU Vcc – it will pull the voltage down to the appropriate level.

Thanks, now is perfectly clear.

wight wrote:
Unless I am mistaken, this means that for a Zener build a person would replace D1 with the 200 ohm resistor, then place a Zener between the lefthand pad of the resistor and the lefthand pad of C1

That is what i think, but i had found this thread with C1 between resitor and gnd, and i knew that this cap position is something from the past on 1S configurations, but i didn’t know if it still was doing like that for zener mod.

wight wrote:
I think that this will be a pain as you have things setup now: trying to set a 0603 or 0805 component on those little diode pads will result in the component wanting to walk/spin… probably getting stuck to C1 sometimes. Stacking the Zener on top will probably also be a bit of a pain.

You are completly right, i was so concentrated on finding where to place zener, that i didn’t realize that resistor would have to be soldered on those tiny pads.

wight wrote:
Everything RMM brings up here is important and worthwhile to think about.

That is someting you can take for granted i do, i just was surprised.
I have narrowed exposed GND ring to 0.5mm, i hope it is enough.

I am still working on zener mod capability, but here are the changes I’ve done for the moment:
-Wider traces
-Some components relocated (far from vias)
-Narrower exposed GND ring

Thanks all for the sugestion, if there is someting more you think could be improved, let me know.

https://oshpark.com/projects/BDWjf6DU

nitroz
nitroz's picture
Offline
Last seen: 3 months 1 week ago
Joined: 07/11/2012 - 16:57
Posts: 206
Location: Monroe

Has anyone laid out an 8mm XM-L, MCPCB?

If not, who can?

Thanks!

Halo...
Halo...'s picture
Offline
Last seen: 4 years 10 months ago
Joined: 12/15/2011 - 02:39
Posts: 3304
Location: Halo island

nitroz wrote:
Has anyone laid out an 8mm XM-L, MCPCB?

The problem is that custom MCPCBs (metal core PCBs) are quite expensive. Do you have a source with a decent price?
nitroz
nitroz's picture
Offline
Last seen: 3 months 1 week ago
Joined: 07/11/2012 - 16:57
Posts: 206
Location: Monroe
Halo… wrote:
nitroz wrote:
Has anyone laid out an 8mm XM-L, MCPCB?
The problem is that custom MCPCBs (metal core PCBs) are quite expensive. Do you have a source with a decent price?

I wish.

led4power
led4power's picture
Offline
Last seen: 8 hours 1 min ago
Joined: 12/29/2012 - 09:48
Posts: 1325
Location: Croatia,EU
WarHawk-AVG
WarHawk-AVG's picture
Offline
Last seen: 1 year 2 weeks ago
Joined: 01/04/2014 - 06:47
Posts: 5071
Location: H-Town
led4power wrote:

http://smart-prototyping.com/Prototyping-Services/Electronic-Prototyping...PCB-Prototyping.html

Starting price 30$ for 10pcs 5×5cm.

OH!!!!!

Wonder if gerbers sent to them, the pad under the heat slug could be left open and auto filled with HASL from the factory Wink

nitroz
nitroz's picture
Offline
Last seen: 3 months 1 week ago
Joined: 07/11/2012 - 16:57
Posts: 206
Location: Monroe
led4power wrote:

http://smart-prototyping.com/Prototyping-Services/Electronic-Prototyping...PCB-Prototyping.html

Starting price 30$ for 10pcs 5×5cm.

Awesome! Thanks.

Halo...
Halo...'s picture
Offline
Last seen: 4 years 10 months ago
Joined: 12/15/2011 - 02:39
Posts: 3304
Location: Halo island

WarHawk wrote:
Wonder if gerbers sent to them, the pad under the heat slug could be left open and auto filled with HASL from the factory Wink
I don’t think it’s likely that they will be setup to do that. We also don’t know at what point the fiberglass (dielectric) layer is bonded to the metal core.

No harm in asking though. Worst they can do is say no.

wight
Offline
Last seen: 1 year 3 months ago
Joined: 11/27/2013 - 16:40
Posts: 4969
Location: Virginia, USA

Pablo E. wrote:
-Narrower exposed GND ring
If you make this too narrow and don’t make your PCB larger than 17mm diameter you’ll find that the exposed area of GND ring will sometimes disappear during production. If you look carefully at some other drivers you’ll note that they are setup with a thin exposed GND ring plus larger exposed areas where possible.

The A17PZL is a good example actually. Take a look at the OP in that thread. Note that there are narrow GND rings on the top and the bottom. The primary ring on the bottom is skinnier than the ring on the top. The exposed part of one on the bottom completely disappears when OSH Park produces the board! Fortunately the bottom of that board has exposed scallops which are still there. You can see of of that in the 4 pics in the OP.

Still fine, still on a break. One day I’ll catch up with you folks! previous wight catchup Wink
list of my drivers & variants (A17DD, FET+1 stuff, WIP stuff, etc)

RMM
RMM's picture
Offline
Last seen: 1 year 3 months ago
Joined: 07/23/2013 - 13:47
Posts: 4006
Location: USA

What you'll find is that if you are pushing the limits too much on copper thickness, trace thickness, drill hits, solder mask, etc., that you will often get boards that aren't 'as expected'.  Oshpark uses several different fabs, so sometimes you can get a board and it will be different than other boards.  The Chinese fabs seem to be even more fond of making arbitrary changes, so be careful.

Mountain Electronics : batteries, Noctigon, and much more! What's new? 

nitroz
nitroz's picture
Offline
Last seen: 3 months 1 week ago
Joined: 07/11/2012 - 16:57
Posts: 206
Location: Monroe

Can anyone here convert an autocad to gerber for the place listed above?

PM me if you can. Thanks!

mattlward
mattlward's picture
Online
Last seen: 2 min 45 sec ago
Joined: 06/19/2015 - 09:20
Posts: 2861
Location: Illinois, USA

This has been ordered, but not received or built yet…
Courui D01 Contact Board V3

Matt

EDC rotation:
FW1A, LH351D 4000k (second favorite)
FW3A, LH351D 3500k
FW3A, SST20 FD2 4000k
FW3A, Nichia 4000k sw40 r9080 (favorite light!)
FW3A, Cree XP-L Hi 5A3
Emisar D4V2, SST20 4000k
S2+, XM-L2 T6 4C

Halo...
Halo...'s picture
Offline
Last seen: 4 years 10 months ago
Joined: 12/15/2011 - 02:39
Posts: 3304
Location: Halo island

Autocad is not for pcb design.
So there is not going to be any automated way to convert autocad to gerber.

nitroz
nitroz's picture
Offline
Last seen: 3 months 1 week ago
Joined: 07/11/2012 - 16:57
Posts: 206
Location: Monroe
Halo… wrote:
Autocad is not for pcb design. So there is not going to be any automated way to convert autocad to gerber.

Let me rephrase it then. Can someone recreate the autocad into a gerber?

wight
Offline
Last seen: 1 year 3 months ago
Joined: 11/27/2013 - 16:40
Posts: 4969
Location: Virginia, USA

nitroz wrote:
Let me rephrase it then. Can someone recreate the autocad into a gerber?
Did you post your DXF/whatever somewhere? A straightforward conversion would be to export bitmaps and then use Eagle or KiCad to import the bitmaps into layers. From there you can export a set of gerbers.

This process can generate rough edges. In the past for my own purposes I’ve exported high resolution bitmaps at 4x my desired size. I would then import them into KiCad and save that. At that point I’d use a web service to reduce them to 25%. (the web service is intended to help fix stuff like putting logos on silkscreen art). Once they were reduced they were the correct size of course and I’d reopen the files in KiCad and export gerbers. It’s a lengthy process because you must do it for every layer. Good luck!

Still fine, still on a break. One day I’ll catch up with you folks! previous wight catchup Wink
list of my drivers & variants (A17DD, FET+1 stuff, WIP stuff, etc)

Pablo E.
Pablo E.'s picture
Offline
Last seen: 4 years 3 months ago
Joined: 08/08/2015 - 13:54
Posts: 234
Location: Spain (GMT+1)

I got a batch of boards with empty ring on the edge, but i thought it was a fault because the LED+ was not exposed and in the second batch were copper to the edge.

Seeing the second board i didn’t know why OSHpark recommends at least 15 mil (0.4mm) clearances from traces to the edge of the board, but now i know there are variation between different fabs, it makes sense .

wight wrote:
thin exposed GND ring plus larger exposed areas where possible.

Great tip, i’ll put into practice.

RMM
RMM's picture
Offline
Last seen: 1 year 3 months ago
Joined: 07/23/2013 - 13:47
Posts: 4006
Location: USA

Yeah, that kind of change is a bummer.  I have about 500 17mm boards with no exposed copper behind the FET.  Same exact file I had already ordered dozens of times, but the time I went to order the most look what happened!

If you look at the silkscreen and milling/drill hits you can often tell which Oshpark fab they were from.  I can distinctly tell the differences between them, but you never know which one you're going to get.

Mountain Electronics : batteries, Noctigon, and much more! What's new? 

Pablo E.
Pablo E.'s picture
Offline
Last seen: 4 years 3 months ago
Joined: 08/08/2015 - 13:54
Posts: 234
Location: Spain (GMT+1)

It has been a lot of time since i posted v004 (I have been a bit busy due to exams), but today i’ve finished a new version.

Theese are the changes from v004:
-0603 diode pads (for zener mod)
-More exposed copper on top GND ring
-Added copper pour under MCU for temp sensoring
-Wider bottom GND ring near FET vias

https://oshpark.com/shared_projects/tjm3DzQH

As always, any suggestion will be welcome.

finges
Offline
Last seen: 2 days 15 hours ago
Joined: 11/19/2014 - 14:50
Posts: 504
Location: Germany

Nice, one question is it “only” a design change to wights A17DD-L or is there any extra functionality?

Is it possible to drill through that LED+ pad and push a cable through? Could you already make a hole there? This way you can use the same cable to bypass the spring and to get to the LED.

Pablo E.
Pablo E.'s picture
Offline
Last seen: 4 years 3 months ago
Joined: 08/08/2015 - 13:54
Posts: 234
Location: Spain (GMT+1)

finges wrote:
Nice, one question is it "only" a design change to wights "A17DD-L":http://budgetlightforum.com/node/36710 or is there any extra functionality? Is it possible to drill through that LED+ pad and push a cable through? Could you already make a hole there? This way you can use the same cable to bypass the spring and to get to the LED.

 

Hi finges,

there is an extra funcionaltiy, pads for a pull down resistor (PDR) to avoid "turbo to moon" blink.

 

I doubt a hole could be drilled, but i can make it in the design.

 

Stay tuned.

Pablo E.
Pablo E.'s picture
Offline
Last seen: 4 years 3 months ago
Joined: 08/08/2015 - 13:54
Posts: 234
Location: Spain (GMT+1)

This version has been replaced by v006: http://budgetlightforum.com/comment/914429#comment-914429

Done!

Hole (actually is a via) is 2mm/0.0787inch in diameter.

https://oshpark.com/shared_projects/rVrw8a4J

chouster
Offline
Last seen: 3 months 1 week ago
Joined: 02/20/2014 - 15:05
Posts: 746
Location: germany

Very nice Pablo! Thank you very much for sharing. Only thing I’d like to see changed is the 0603 cap, I would prefer 0805 for an OTC.

pilotdog68
pilotdog68's picture
Offline
Last seen: 1 month 3 weeks ago
Joined: 05/30/2013 - 23:31
Posts: 6422
Location: Held against my will in IOWA, USA

chouster wrote:
Very nice Pablo! Thank you very much for sharing. Only thing I’d like to see changed is the 0603 cap, I would prefer 0805 for an OTC.

+1
I think if you moved the FET pull-down resistor to right above the FET, you would have room for 0805 pads right at the top of the board.

My Favorite Modded Lights: X6R, S8 , X2R , M6, SP03

Major Projects:  Illuminated Tailcap, TripleDown/TripleStack Driver

Mike C
Mike C's picture
Offline
Last seen: 2 days 18 hours ago
Joined: 01/22/2014 - 08:03
Posts: 2554
Location: Sweden

Or rotate R1 90 degrees, then you can fit all three resistors side by side closer to the center, leaving plenty of room for a 0805 OTC pad.
But myself, I wouldn’t bother. I use 0603 on all my boards.

Pablo E.
Pablo E.'s picture
Offline
Last seen: 4 years 3 months ago
Joined: 08/08/2015 - 13:54
Posts: 234
Location: Spain (GMT+1)

Don’t you consider soldering a 0805 cap on the 0603 pads?, I don´t know how difficult it is, but it fits (like in the right diagram)

Do you prefer the 0805 option?, if so it could be done performing both of your suggestion, i see some potential clearance problems but i think they could be solved.

wight
Offline
Last seen: 1 year 3 months ago
Joined: 11/27/2013 - 16:40
Posts: 4969
Location: Virginia, USA

Yet another untested monstrosity… This implements what is intended to be a snubber for 1s applications, plus a bleeder for pilotdog68’s tailcap LED system, plus an OTC pulldown, plus a gate pulldown, plus a 3mm via for the BAT+ wire. There was no space for a BAT+ pad on top, so it’s bypass or go home. At first I thought that I was going to be forced to increase my normal 1mm physical keepout, but somehow I got away with it.

wight wrote:
This looks decent… No guarantees that it’ll work though.


https://oshpark.com/shared_projects/4nsRthrB

Stuff:

  • 3mm bat+ spring bypass wire hole.
  • Pin3 pad available for scraping. (1.4×1.8mm triangle)
  • ~1mm physical keepout around the edge of the driver
  • >0.5mm electrical keepout around the edge of the driver
  • Placement is “a little” tight.

Parts list:

  • C1 – 1uF, maybe much more depending on stability (5uF). [0603 – decoupling capacitor]
  • OTC – 1uF X7R. Maybe much more (10uF) if used w/ R5 pulldown. [0805 – off-time capacitory]
  • R1 – whatever
  • R2 – whatever
  • R3 – 200 ohm or so. [0603 – Zener load resistor]
  • R4 – 560 ohm or so. [0603 – tailcap LED bleed resistor] [not needed w/out tailcap LED circuit]
  • R5 – unknown value [0603 – OTC pulldown resistor]
  • R6 – 12k-ohm [0603 – FET gate pulldown resistor] [used to prevent flash when moving to moon mode]
  • D1 – protection / stability diode. Do not bypass. [SOD-323 Schottky Diode]
  • Z1 – snubber diode. Do not bypass. [SOD-323 Zener Diode]
  • MCU: SSU or SU w/ legs bent (eg our normal narrow MCUs, like v009 and most other BLF drivers)
  • FET: Power-SO8 (eg our normal FETs, like v009 and many other BLF drivers such as A17DD-S08)

Still fine, still on a break. One day I’ll catch up with you folks! previous wight catchup Wink
list of my drivers & variants (A17DD, FET+1 stuff, WIP stuff, etc)

Pablo E.
Pablo E.'s picture
Offline
Last seen: 4 years 3 months ago
Joined: 08/08/2015 - 13:54
Posts: 234
Location: Spain (GMT+1)

This version has been replaced by v007: http://budgetlightforum.com/comment/914429#comment-914429

This board is really dense populated! you made a great job!

I have already changed OTC pads size to 0805, i have had to remove some vias, and voltage resistors may not be easy accesible, placed that way between tall components, but i think it is done.

https://oshpark.com/shared_projects/0NAo2EmQ

Pages