Petit Crouton & Crystal Focus V5-V6 tips and tricks

[erv]

Here's where I'll post interesting information, tips, tricks, hacks for the CFV5.

** Let's start with the question I got initially for the CFV4 : the ground spots. Where and what for.
Here's the CFV5 version




** Sound level.
As addressed in the V5 not being loud enough and despite I'm praising for quality versus sound volume, I'm publishing the little "hack" to increase the volume. Please note that I'm not responsible of what you do with that info. You're on your own.

The goal is to change the value of the audio amp gain. It will level up everything, hence reducing the headroom, leading to audible distortion/clipping if abused.
To avoid changing the resistor, which involves good soldering skills, and since it's about *lowering* the value, the simplest solution is to add another resistor in parallel with the one in place on the PCB (39k). The resistor is a 0603 footprint.



The general maths is given by the following relationship. Call "A" the  desired amplification factor.

Resistor to add = 39k / (A-1)

ex :
- a double factor => add a resistor of 39k
- a quad factor => add a resistor of 13k.

[erv]

As per July 2011 :


CF V5.x will now come with the volume hack already in place, using a 20K resistor in place of the original 39K one, hence doubling the loudness and not requiring the resistor hack anymore.

To go back to the "distortion free" level, the vol parameter on the SD card must be set to 3. But since digital volume control also brings its quantification noise (VERY SMALL with 16 bits of resolution) I'd say leave it LOUD BABE  :dft012:

(volume settings up to 11 expected on V6 maybe ?  ;D )

[erv]

More a "warning" rather than a tip but  :)

We might have figured out with MH why a (very) few people got their activation or aux. switches pad no longer working, on either CF or PC (same core chip) : each time we heard of the problem, it was not when first soldering / installing the board, but *after*.

We traced the problem down to the fact the user is then working on the pads/switches with the recharge port installed. The negative is cut with the kill key but the positive is still active. The tip of the soldering iron is often grounded and the board gets a fake powering via the soldering iron itself.

Fix : external replacement of the blown pull-up resistor(s).

Use a resistor to fix the pad ONLY if the pad is damaged

Repair can take place under the board using 0603 SMD resistors and wrapping wire, or on the top side, more likely with classic thru hole resistors (1/8w are enough).


Bottom line : if you need to rework a board, totally isolate the battery first. It might not be necessary for the *outputs* (like the accent pads) since we never EVER got the problem. Just the input pull-ups.

[erv]

SD card holders sometimes require to be cycled 4-5 times before the push-on push-off mechanism is at 100%

take the time to cycle your SD card  in the CF or PC before you install it, it's easier before than after.

[erv]

Another tip : when wiring a board (especially the first time), most people go for the "leap of faith". They solder the recharge port, then remove the kill key.
*fingers crossed* then the magic smoke.

Ideally you would use a bench PSU to make sure your wiring is correct. However, if you don't have one, here are ways to avoid problems :

- first time you power the board, have the led disconnected
- use 3 or 4 x AAA (rechargeable preferred) to power the board. That will be enough to power the audio and make sure the board boots properly but since the AAA will output a max of 900-1000 mA, you'll limit the issues if you have a short somewhere. 1A is not enough to destroy the luxeon and most of the CF parts should survive.

this is NOT the case if you start with 18650 li-ion cells as they will pull out a good 4A if you have a straight short

[erv]

color changing and direct drive using the PC & CF

PC 1.X

"Direct drive - single color". Please note that "direct drive" reflects "a driver/regulator-less " way of powering the led. Not "direct drive" like you don't use a resistor.
"always use a resistor" (tm) (unless it's negative)




variant for a color changing saber. R1 and R2 must be calculated accordingly to the Forward Voltage (Vf) of each die. Principle can be extended to more colors / combo using the right rotary switch.




Please note that in each case, the luxeon driver remains fully functional, as well as the flickering/shimmering, ramp up/down effect, for both the direct drive and the regulated output of the PC (so you can imagine combos between a main die out of the regulator + color/shade change with the direct drive).

PEx is limited to about 2A, if you plan a heavier load, you can wire 2 PEx in parallel to double the accepted max amperage. I've also made some "doubled" PEx on a single PCB in the past

PWM access point on a PC2.x (it's the same spot on the CF, if you'd ever need a CF in direct drive)
(anotated pic courtesy of fragrax, thank you)

[fragrax]

Thanks for sharing, erv! I've learned a lot on this forum  :)

[erv]

It's been on my todo list for a while, I'm slowly catching up...
here's the screen copy of the RICE real time configuration editor help that will be added to all versions of the app (PC, CF, CF LS etc), hopefully it will help new comers to understand how it works and/or how to get started with it. The app itself isn't difficult per say, but as there's hardware and software (driver install) associated with it, it's better to have everything put down on paper.

also, if you want to see the possibilities of RICE and how easy it is to configure your saber in real time, here's a great video from Genesis Custom Saber:
http://www.youtube.com/watch?v=bRUTRc_dIzY#

Click to enlarge

[erv]

Nano Biscotte v2 Deep Sleep indicator


An alternative wiring for a power on indicator that goes off when the Nano Biscotte enters deep sleep. I've enlarged the pad used for wiring the LED on the new version of the PCB that I'm ordering today but it works on the former NBv2 PCB, though it might be harder to solder.