XS1-L4A-64-TQ48 Test board

XCore Project reviews, ideas, videos and proposals.
User avatar
Folknology
XCore Legend
Posts: 1274
Joined: Thu Dec 10, 2009 10:20 pm

Re: XS1-L4A-64-TQ48 Test board

Postby Folknology » Mon Jan 27, 2014 9:54 pm

On Xmos's L1-48TQFP DIP board they tap the output of the first APX803 as a 3v3 Power Good signal which they in turn use to enable the 1V switcher, you might want to consider going that proven route given you have the signal.

regards
Al
User avatar
AtomSoft
XCore Addict
Posts: 135
Joined: Mon Dec 14, 2009 3:02 pm

Postby AtomSoft » Mon Jan 27, 2014 10:18 pm

For the links i will leave for DESIGN #2

This design i removed the extra reset IC and used one... basically... in theory... im using a buck ic for the 1v, and the reference is .65v i think.... so one it hits ... 1.65v we should have 1v @1A available on the output... but the time it hits 3.3v 1v has all the juice it needs and then reset will be released on the chip and all lines will have a nice (decent) power source.

I was thinking about adding a SOLDER JUMPER and regulator to allow powering from the XTAG itself. Thoughts?
User avatar
AtomSoft
XCore Addict
Posts: 135
Joined: Mon Dec 14, 2009 3:02 pm

Postby AtomSoft » Mon Jan 27, 2014 10:19 pm

Folknology wrote:On Xmos's L1-48TQFP DIP board they tap the output of the first APX803 as a 3v3 Power Good signal which they in turn use to enable the 1V switcher, you might want to consider going that proven route given you have the signal.

regards
Al

nice! This i can use!

EDIT: Updated...
You do not have the required permissions to view the files attached to this post.
User avatar
AtomSoft
XCore Addict
Posts: 135
Joined: Mon Dec 14, 2009 3:02 pm

Postby AtomSoft » Tue Jan 28, 2014 12:09 am

Making this PCB 4 layers will raise the price by $3 - $5... but then i can add the xSYS with some Solder Jumpers and you can use it optionally or make the board a bit longer and have a 74CBTLV3126 added and using 1 jumper to enable or disable the xScope portion. But when in use you must not use those lines... I guess i can add the XLB info on the pin label itself. for this its X0D16:X0D19 using XLB... since these pins are easy to get to .
User avatar
AtomSoft
XCore Addict
Posts: 135
Joined: Mon Dec 14, 2009 3:02 pm

Postby AtomSoft » Tue Jan 28, 2014 2:29 am

OK im going to order the pcb today without the xSCOPE on the xSYS connector. If anything the next set will use the 128pin version of this chip so i can use the xSCOPE without loosing valuable pins.
User avatar
Folknology
XCore Legend
Posts: 1274
Joined: Thu Dec 10, 2009 10:20 pm

Postby Folknology » Tue Jan 28, 2014 9:14 am

You should include the second 803 for power on reset.

regards
Al
User avatar
AtomSoft
XCore Addict
Posts: 135
Joined: Mon Dec 14, 2009 3:02 pm

Postby AtomSoft » Tue Jan 28, 2014 2:18 pm

Hey Al, I still do not see how it will help... perhaps with a large cap to delay it might but when the first one triggers wont the second trigger immediately? In turn having no real effect?
User avatar
mon2
XCore Legend
Posts: 1679
Joined: Thu Jun 10, 2010 11:43 am

Postby mon2 » Tue Jan 28, 2014 3:17 pm

If you have not yet submitted the PCB to the shop, review the following suggestions:

a) consider to strap pin 1 (OE) of the oscillator to +3.3 volts. Granted that such devices have an internal pull-up to permit the oscillator to be active immediately upon power up, we have seen failures on our designs (were caught during internal testing of the product) where we left OE floating. After an inspection by the factory (shipped all the way back to Taiwan) - they confirmed that the 3-5 pieces (out of 80k-100k units) that failed were missing or had weak solder connections for the internal pull-up resistor. Such audits can get expensive real fast. Since then, we have changed all of our designs to be hard strapped to Vcc (+3.3 volts) on OE (Pin 1). No harm in being redundant here since it is only a single PCB trace or you could consider an external pull-up resistor (10k is fine).

b) Instead of the APX803, consider the following find from TI:

http://www.ti.com/lit/ds/symlink/tps3808g09.pdf

Also review this:
http://e2e.ti.com/support/power_managem ... 32732.aspx

As it stands, your present circuit is still not monitoring the 1.0 volt rail for stability. Technically you are releasing #RESET when only the 3.3 volt rail is considered to be stable.

I do not think there is any benefit to having an APX803 to enable the SC189C in your design since the Semtech datasheet lists that the device starts to operate from the low side threshold of 2.9 volts. This is effectively the job of the APX803 (if present). In my opinion, it is an overkill to enable the SC189 with the power rail supervisor. Instead, consider the above device which is truly able to monitor the 1.0 volt rail with a threshold of 0.9 volts AND apply a delay that can be defined to suit for the reset line. This device appears to be available in volume from assorted suppliers and you can even get free samples.
User avatar
AtomSoft
XCore Addict
Posts: 135
Joined: Mon Dec 14, 2009 3:02 pm

Postby AtomSoft » Tue Jan 28, 2014 4:19 pm

Thank you for the input, I do have vcc on the oe line since yesterday thanks and ill look into the .9v reset ic. I had some picked out yesterday, expect a update in a hr since ill have to sift through parts then create a eagle part for it and alter sch.
User avatar
AtomSoft
XCore Addict
Posts: 135
Joined: Mon Dec 14, 2009 3:02 pm

Postby AtomSoft » Tue Jan 28, 2014 4:26 pm

Take a look at these:

BU4809
BU4833

They are about 1/4 the price and are essentially the same thing... Im sure i can fit these on.

Who is online

Users browsing this forum: No registered users and 0 guests