| 00:05.40 | *** join/#edev flo_lap (n=fuchs@g227195073.adsl.alicedsl.de) |
| 00:06.25 | *** join/#edev fuchs__ (n=fuchs@f048040202.adsl.alicedsl.de) |
| 00:20.15 | *** join/#edev florian (n=fuchs@f048002100.adsl.alicedsl.de) |
| 00:25.33 | *** join/#edev redblue (i=star@ppp200.108-253-207.mtl.mt.videotron.ca) |
| 02:44.24 | *** join/#edev miknix (n=miknix@gentoo/developer/miknix) |
| 02:50.21 | MonMotha | by pure chance, I managed to come up with "1234568690" |
| 02:50.23 | MonMotha | close enough? |
| 02:52.36 | ds2 | anyone opinions on how bad it is to use those higher current charge pumps instead of an inductive switcher to get 5V @500mA from a Li cell? |
| 02:52.46 | ds2 | anyone have... |
| 02:58.33 | MonMotha | 500mA is kinda high for a charge pump |
| 02:58.39 | MonMotha | I'd expect it to have some serious ripple |
| 02:59.10 | MonMotha | well, either that or require caps so big that you could fit an inductive boost converter in |
| 02:59.49 | ds2 | to be fair, 500mA is from paralleling 2 250mA pumps; the mfg claims it is doable with ~10uF caps |
| 03:00.19 | MonMotha | parlleling charge pumps doesn't always behave as one might hope. Does the mfg have an app note |
| 03:00.21 | MonMotha | ? |
| 03:00.49 | ds2 | yes, this is the famous maxim pumps; they claim it can be paralleled |
| 03:00.56 | MonMotha | ah, yes, the Maxim stuff |
| 03:01.07 | MonMotha | I'd expect that with two charge pumps and lots of 10uF caps you might be able to fit an inductive solution in the same PCB space |
| 03:01.18 | *** join/#edev linac (n=lin@124.21.240.37) |
| 03:01.26 | ds2 | I like charge pumps cuz getting a proper cap seems to involve less voodoo then getting a proper inductor |
| 03:01.40 | MonMotha | eh, depends on the app notes your SMPS controller vendor provides |
| 03:01.55 | ds2 | I am about 10 failed attempts at inductive switchers to 1 success so.... |
| 03:02.01 | MonMotha | the ON and National app notes tend to provide pretty good info on selecting the inductors |
| 03:02.06 | MonMotha | the LT ones seem a little voodoo-ish |
| 03:02.24 | MonMotha | wow, all of mine have at least worked, though some whine a little |
| 03:02.40 | ds2 | it seems I have to use the exact inductor. matching just the inductance is not sufficient |
| 03:02.50 | MonMotha | then again, I'll spend a good day or two tweaking the design to match an inductor I can buy |
| 03:03.04 | MonMotha | well, L isn't the only parameter on a coil |
| 03:03.15 | MonMotha | R is very much present, and there's also usually a parasitic C you have to watch out for |
| 03:03.17 | ds2 | my understanding is a poor inductor will just be less efficient but apparently I am very wrong |
| 03:03.39 | MonMotha | biggest issue is that the parasitics (which can include layout issues) can cause control instability |
| 03:03.50 | MonMotha | basically, keep the inductor as ideal as possible, and keep the layout tight |
| 03:04.04 | MonMotha | your freewheeling diode selection is also important |
| 03:04.09 | ds2 | I am basically dead bugging a one off proto |
| 03:04.20 | MonMotha | oh, heh...yeah, don't try an SMPS dead bug style |
| 03:04.30 | MonMotha | the parasitics will eat you alive at frequencies above a few dozen kHz |
| 03:04.35 | ds2 | dead bug seems to work at RF |
| 03:04.40 | MonMotha | yeah, it's strange |
| 03:04.51 | MonMotha | I think RF benefits from the ground plane |
| 03:05.03 | MonMotha | SMPSes are usually not high enough frequency for that to be of much effect, I think |
| 03:05.10 | MonMotha | but they're still high enough to cause problems |
| 03:05.11 | ds2 | SMPS doesn't? I am dead bugging on doublesided PCBs |
| 03:05.31 | ds2 | large copper planes with the polyamide tape as an insulator |
| 03:05.33 | MonMotha | what frequencies were you using with your SMPS? They vary a lot |
| 03:05.46 | ds2 | the highest is 1MHz |
| 03:06.03 | MonMotha | some of the modern stuff is several MHz (to get the passives small and transient response better), but some is maybe 100kHz or less for off-line types |
| 03:06.31 | MonMotha | I dunno what to tell you. All of my stuff has worked reasonably well. Just some audible noise sometimes. |
| 03:06.45 | MonMotha | Sometimes the control is a little unstable, but I just end up with some unexpected ripple rather than it not regulating. |
| 03:07.08 | ds2 | is there some digikey sourceable chips with onboard diode + switch fet that is relatively fool proof? |
| 03:07.28 | MonMotha | I'm not aware of any with onboard diode |
| 03:07.31 | ds2 | this is a power supply for a battery operated beagle board |
| 03:07.44 | MonMotha | the NCP3163 seems very forgiving control-wise, but it needs an external diode |
| 03:07.49 | ds2 | I see.... what if I drop the onboard diode requirement? |
| 03:08.33 | MonMotha | NCP3163. LT1776, but it involves a little more voodoo. |
| 03:08.46 | MonMotha | the 3163 has a nice design spreadsheet (does need Excel, though) |
| 03:09.09 | ds2 | works in OpenOffice? |
| 03:09.11 | MonMotha | and it's an old, proven design. It was the MC3163 many years ago before Moto spun off ON. |
| 03:09.22 | MonMotha | I don't think it works in OO.org. It's macro-heavy. |
| 03:09.29 | ds2 | oh |
| 03:09.42 | MonMotha | what's sad is that it need not be |
| 03:09.50 | MonMotha | you can try it. You can download the design tool from the ON website. |
| 03:09.52 | ds2 | Hmmm... 16 pin |
| 03:10.08 | MonMotha | yes. It has an onboard switch, so that adds some pins |
| 03:10.14 | MonMotha | it also has a current limit function |
| 03:10.25 | MonMotha | and a power-good indicator |
| 03:10.42 | ds2 | does it have self low voltage shutdown? |
| 03:10.49 | ds2 | (looking at block diagram now) |
| 03:10.53 | MonMotha | I don't remember |
| 03:11.10 | MonMotha | I think it may rely on the LVI flag (which I commonly tag onto the system reset line - it's open collector) |
| 03:12.05 | ds2 | I see that |
| 03:12.08 | MonMotha | I unfortunately can't get quite the same level of FAE as I used to. The FAE I knew at ON got laid off :( |
| 03:12.21 | ds2 | but the design looks complicated for a mostly digital person |
| 03:12.28 | MonMotha | it's pretty simple, actually |
| 03:12.32 | ds2 | got til Sunday to finalize on a power supply |
| 03:12.54 | ds2 | idiot proof, but possibly bad design pump or doing it right with a switcher |
| 03:12.58 | MonMotha | check out page 12 |
| 03:13.07 | MonMotha | that's the buck topology |
| 03:13.14 | MonMotha | I guess you want boost, which I've not used that chip in |
| 03:13.30 | MonMotha | only 9 external components, though |
| 03:13.30 | ds2 | *nod* |
| 03:13.48 | ds2 | but those are value sensitive resistors if I read that right |
| 03:13.58 | MonMotha | well, R1 and R2 set the output voltage |
| 03:14.17 | MonMotha | basically you divide the output down to 1.25V using a voltage divider and that forms the feedback loop |
| 03:14.42 | ds2 | *nod8 and that makes then value sensitive |
| 03:14.49 | MonMotha | true, though not overly so |
| 03:14.59 | MonMotha | depends on how tight you need your output voltage to be |
| 03:15.06 | ds2 | 10% |
| 03:15.11 | MonMotha | shouldn't be a problem then |
| 03:16.11 | MonMotha | MBRS140T3 is a good diode to use |
| 03:16.15 | MonMotha | though it's surface mount |
| 03:16.29 | ds2 | plan is to surface mount |
| 03:16.49 | MonMotha | ok |
| 03:17.07 | MonMotha | for inductors, the coiltronics DR73 or DR127 (Depending on your output current needs) seems to work well |
| 03:17.28 | MonMotha | for buck topologies, the spreadsheet seems to sometimes come up with higher than expected inductances. It runs, but you get some extra ripple. |
| 03:17.29 | ds2 | is that from experience or is that from a page I missed? |
| 03:17.34 | MonMotha | experience |
| 03:18.00 | MonMotha | for the RSC, use a metal film inductor, not thick/thin film |
| 03:18.35 | ds2 | aren't metal film more inductive? |
| 03:18.48 | ds2 | i know my friend the RF guy prefers carbon |
| 03:19.22 | MonMotha | yes, but this isn't really RF |
| 03:19.27 | MonMotha | you're talking 300kHz max |
| 03:19.35 | MonMotha | you can certainly use a carbon, but htey'r ehard to find in surface mount |
| 03:19.46 | ds2 | then what's the reason for metal over thin film? |
| 03:20.10 | MonMotha | the thick/thin film can fall apart on large pulses |
| 03:20.30 | ds2 | as in heating due to current? |
| 03:20.31 | MonMotha | if you short the output, that resistor will (for a very brief period of time) conduct quite a chunk of current, until the control catches up |
| 03:20.51 | MonMotha | yes, but it's not even long enough to heat up the whole case. Just the little film trace inside |
| 03:21.06 | ds2 | I see. |
| 03:21.24 | ds2 | I just literally follow the schematic and I guess that's why my stuff don't usually work :/ |
| 03:21.54 | MonMotha | well, the schematic is right. It's just that there's more to a resistor than resistance, more to a capacitor than capacitance, and a LOT more to an inductor than inductance. |
| 03:22.02 | MonMotha | for the output cap, use a low ESR type |
| 03:22.18 | ds2 | low ESR only matters for elect.'s right? |
| 03:22.30 | MonMotha | yes |
| 03:22.37 | MonMotha | well, and tantalums |
| 03:22.47 | MonMotha | though tantalums tend to be a lot better than electrolytics anyway |
| 03:23.08 | ds2 | trying to use ceramics |
| 03:23.27 | MonMotha | those can actually cause instability due to the overly low ESR |
| 03:23.39 | ds2 | DOH |
| 03:23.41 | MonMotha | but I've not tried it on a 3163 |
| 03:23.42 | MonMotha | may work |
| 03:23.50 | MonMotha | the control loop on that thing is VERY forgiving |
| 03:24.04 | ds2 | time for me to go battle traffic... thanks for the hints |
| 03:24.11 | MonMotha | np |
| 03:24.14 | MonMotha | that'll be $100 :) |
| 03:24.31 | MonMotha | FWIW, I know of a company that offers engineering services :) |
| 03:24.58 | ds2 | this is for a one-off demo |
| 03:25.01 | MonMotha | gotcha |
| 03:25.11 | MonMotha | if you need help on the final product, I know of a place you can ask |
| 03:25.28 | ds2 | *nod* |
| 03:50.18 | *** join/#edev kergoth (n=kergoth@ip68-3-163-239.ph.ph.cox.net) |
| 04:46.40 | *** join/#edev ZeZu (n=null@c-98-227-56-17.hsd1.in.comcast.net) |
| 04:46.40 | *** join/#edev Chocobo (n=Ask@strongbad.eece.maine.edu) [NETSPLIT VICTIM] |
| 04:46.41 | *** join/#edev dfarnsworth (n=dfarnswo@65.200.49.156) [NETSPLIT VICTIM] |
| 04:46.41 | *** join/#edev file (n=file@asterisk/developer-and-muffin-lover/file) |
| 04:46.41 | *** join/#edev ccole (n=cole@66.218.140.242) |
| 04:46.41 | *** join/#edev gandhijee (i=akp@host-66-202-34-165.spr.choiceone.net) |
| 04:46.45 | *** join/#edev ChanServ (ChanServ@services.) |
| 04:46.45 | *** join/#edev agaffney (n=agaffney@gentoo/developer/agaffney) [NETSPLIT VICTIM] |
| 04:46.45 | *** mode/#edev [+o ChanServ] by irc.freenode.net |
| 05:36.25 | ds2 | hmmm |
| 06:21.16 | *** join/#edev DrZimmerman (n=theo@77-56-242-245.dclient.hispeed.ch) |
| 06:34.03 | *** join/#edev redblue (i=star@ppp039.108-253-207.mtl.mt.videotron.ca) |
| 08:34.36 | *** join/#edev Omegamoon (n=egoe@demolitionman.xs4all.nl) |
| 08:57.01 | *** join/#edev lyakh (n=lyakh@p57BD0E7F.dip0.t-ipconnect.de) |
| 09:30.16 | *** join/#edev damunix (n=damunix@61.153.204-77.rev.gaoland.net) |
| 09:30.49 | damunix | morning |
| 09:33.48 | mnemoc | morning damunix |
| 09:33.52 | *** join/#edev sleipnir (n=Sleipnir@p4FDD4984.dip.t-dialin.net) |
| 09:37.52 | *** join/#edev likewise (n=chatzill@82-171-51-231.ip.telfort.nl) |
| 09:45.49 | *** join/#edev onnoval (n=sam@123.252.146.52) |
| 10:25.29 | *** join/#edev methril|laptop (n=methril@169.73.220.87.dynamic.jazztel.es) |
| 10:25.53 | *** join/#edev methril|laptop (n=methril@169.73.220.87.dynamic.jazztel.es) |
| 10:28.33 | *** join/#edev methril|laptop (n=methril@169.73.220.87.dynamic.jazztel.es) |
| 11:06.02 | *** join/#edev MonMotha (n=monmotha@72.86.127.225) |
| 11:07.38 | *** join/#edev redblue (i=star@ppp086.108-253-207.mtl.mt.videotron.ca) |
| 11:32.20 | *** join/#edev likewise (n=likewise@82-171-51-231.ip.telfort.nl) |
| 11:51.47 | *** join/#edev g1powermac (n=g1powerm@unaffiliated/g1powermac) |
| 11:52.21 | *** join/#edev flameman (n=toor@net-93-65-2-164.cust.dsl.vodafone.it) |
| 11:54.27 | *** join/#edev DrZimmerman (n=theo@77-56-242-245.dclient.hispeed.ch) |
| 11:56.12 | *** join/#edev florian (n=fuchs@g227116144.adsl.alicedsl.de) |
| 11:57.55 | *** join/#edev pirho (i=pirho@gateway/gpg-tor/key-0x2CEEC9CB) |
| 12:02.57 | *** join/#edev g1powermac_PB (n=g1powerm@unaffiliated/g1powermac) |
| 12:36.02 | *** join/#edev demirtious (n=dima@91.146.33.143) |
| 12:36.31 | *** part/#edev demirtious (n=dima@91.146.33.143) |
| 12:46.18 | *** join/#edev likewise (n=likewise@82-171-51-231.ip.telfort.nl) |
| 13:57.04 | *** join/#edev cjameshuff (n=cjameshu@pool-71-97-181-172.aubnin.dsl-w.verizon.net) [NETSPLIT VICTIM] |
| 13:57.04 | *** join/#edev butterflyback (n=lou@c-68-53-134-58.hsd1.tn.comcast.net) |
| 13:57.04 | *** join/#edev joe_bleau (n=joe_blea@adsl-69-152-201-70.dsl.fyvlar.swbell.net) [NETSPLIT VICTIM] |
| 13:57.29 | *** join/#edev roxfan2 (n=dunno@204.183-244-81.adsl-dyn.isp.belgacom.be) |
| 14:26.12 | *** join/#edev GPSFan (n=kenm@64.92.145.112) |
| 14:26.12 | *** mode/#edev [+o GPSFan] by ChanServ |
| 14:33.27 | *** join/#edev methril|laptop (n=methril@169.73.220.87.dynamic.jazztel.es) |
| 14:34.17 | *** join/#edev DrZimmerman (n=theo@77-56-242-245.dclient.hispeed.ch) |
| 14:41.45 | *** join/#edev methril|laptop (n=methril@169.73.220.87.dynamic.jazztel.es) |
| 14:53.25 | *** join/#edev armin76 (n=armin@gentoo/developer/armin76) |
| 14:59.31 | *** join/#edev g1powermac_PB (n=g1powerm@unaffiliated/g1powermac) |
| 15:19.39 | *** join/#edev kergoth (n=kergoth@ip68-3-163-239.ph.ph.cox.net) |
| 15:35.13 | *** join/#edev methril|laptop (n=methril@169.73.220.87.dynamic.jazztel.es) |
| 17:35.33 | *** join/#edev flavioribeiro (n=avaty@189.71.5.77) |
| 18:15.18 | *** join/#edev DrZimmerman (n=theo@77-56-242-245.dclient.hispeed.ch) |
| 19:59.10 | *** join/#edev flameman (n=toor@net-93-65-2-164.cust.dsl.vodafone.it) |
| 20:13.22 | *** join/#edev florian (n=fuchs@g227116144.adsl.alicedsl.de) |
| 20:40.34 | *** join/#edev JoeLlama (n=snork@unaffiliated/joellama) |
| 21:05.25 | *** join/#edev lyakh (n=lyakh@p57BD0E7F.dip0.t-ipconnect.de) |
| 21:24.46 | *** join/#edev likewise (n=likewise@82-171-51-231.ip.telfort.nl) |
| 22:34.44 | *** join/#edev brolin (n=brolin@190.157.13.75) |
| 23:35.48 | *** join/#edev redblue (i=star@ppp074.108-253-207.mtl.mt.videotron.ca) |