Power Supply

[Chet]

Teleno
Thank you for the explanation. I have smoked two NE555 drivers on my Mini Pulse Plus when things when wrong. Maybe Green’s capacitor coupling or an additional driver added after a monostable multivibrator would work?
Have a good day,
Chet

[green]

Teleno
Thank you for the explanation. I have smoked two NE555 drivers on my Mini Pulse Plus when things when wrong. Maybe Green’s capacitor coupling or an additional driver added after a monostable multivibrator would work?
Have a good day,
Chet

Hi Chet, reminds me I have damaged a couple TLC555's, capacitor coupling to the trigger. You need to pull up the trigger to the + PS with a resistor. The trigger input gets zapped when the input goes high if the drive current capability is high enough(another 555 output). Limit current if needed with a resistor in series with capacitor, pull up resistor >5 times series resistor.

[ODM]

Local bypassing of any/all high-current ICs is a good idea. It's also important to note that the bipolar versions of 555 have essentially a short circuit in the output during switching which leads to high current spikes. It's a good idea to locally bypass a 555 with industry standard practices of a decent 100n cap.

When it comes to phase shifts and such in audio design, it's usually considered critical for signal transfer and not necessarily for power supplies (unless wanting to pile up fancy words for kooky "high-end" scam design). What we want in a power supply is a steady voltage, current transients properly bypassed. Don't be afraid of individual bypasses for any generic opamp and logic IC with 100n ceramics, if your circuit only works with a shifty supply impedance you should create it with individual components instead of expecting battery impedance to be constant.

The point with star routing designs is minimizing common paths for currents, but sneaky reactances can easily creep into the design when actually laying it out as a board. A solid ground plane is usually a more successful first run tactic as long as long as local alternating current consumption is tamed by bypassing. And for high speed circuits it is generally a requirement, not a bonus!

Since nobody takes good ideas at face value on a forum, here's some good reading for anyone wanting to design ... well, anything, really! http://cds.linear.com/docs/en/applic...ote/an47fa.pdf

[Teleno]

Local bypassing of any/all high-current ICs is a good idea. It's also important to note that the bipolar versions of 555 have essentially a short circuit in the output during switching which leads to high current spikes. It's a good idea to locally bypass a 555 with industry standard practices of a decent 100n cap.

When it comes to phase shifts and such in audio design, it's usually considered critical for signal transfer and not necessarily for power supplies (unless wanting to pile up fancy words for kooky "high-end" scam design). What we want in a power supply is a steady voltage, current transients properly bypassed. Don't be afraid of individual bypasses for any generic opamp and logic IC with 100n ceramics, if your circuit only works with a shifty supply impedance you should create it with individual components instead of expecting battery impedance to be constant.

The point with star routing designs is minimizing common paths for currents, but sneaky reactances can easily creep into the design when actually laying it out as a board. A solid ground plane is usually a more successful first run tactic as long as long as local alternating current consumption is tamed by bypassing. And for high speed circuits it is generally a requirement, not a bonus!

Since nobody takes good ideas at face value on a forum, here's some good reading for anyone wanting to design ... well, anything, really! http://cds.linear.com/docs/en/applic...ote/an47fa.pdf

While it's a good idea to bypass subcircuits locally as needed, using a main electrolythic capacitor is a bad idea for a monocoil PI. The high current puse depletes the capacitor and the voltage drops. It then starts to recover slowly at a time when the tiny target signal is being measured. This overlays a spurious ramp on top of the signal, a shifting baseline akin to a changing EF.

Instead, without the big capacitor the voltage recovers in a matter of nanoseconds after the pulse and the target signal is not affected by the capacitor's charging curve.

[ODM]

You will see a similar drop in battery voltage as well, especially if you scope it in AC coupled mode. Stuff that is sensitive to power supply noise is often decoupled with a controlled impedance like filters you would see in a properly designed audio amplifier's preamp stage to prevent the buffer stage from coupling to it by shared current paths from the PSU. Most (well perfomring) single coil PI detector designs that specify a large cap for the coil drive usually decouple it with a resistor, unless they use other means. Look at the Hammerhead for example.

Due to the differential nature of amplifiers you will see them directly coupled to both ends of the coil in a single coil PI detector. To not couple the amplifiers reference directly to the "cold end" of single coil would mean a mutual impedance path with something else and that would introduce an error voltage.

My intention was not to argue anything, just to point out that for instrumentation circuits it is wise to consider bypassing any and all individual ICs supply voltage. Open up any western oscilloscope, and you will see all amplifiers and logic properly decoupled. The shortest supply paths from the individual ceramic/polystyrene caps will stop the ICs internal currents from being coupled by mutual impedance or other emi conduction. Whether you bypass to the opposite supply or the power ground is determined by following the component's current paths.

I've observed EMC tests where a badly bypassed 10MHz low-level clock crystal oscillator made it couple to the ground of an output cable, and made a product nearly fail. Tombstoned bypass ceramic. Other test units had been reflowed without fault fortunately.

[Old cart]

Local bypassing of any/all high-current ICs is a good idea.

Since nobody takes good ideas at face value on a forum, here's some good reading for anyone wanting to design ... well, anything, really! http://cds.linear.com/docs/en/applic...ote/an47fa.pdf

This is an old, but still relavent AP note. I especially appreciate how it devotes lots of time to making good measurements which is an lost art for many engineers and designers, especially from the school of simulating concurs all!

[Sean_Goddard]

NOW I understand why Fisher metal deteor power supplies were ridiculously complex.

[Sean_Goddard]

But I still can't understand WHY I can't spell detector!

[WM6]

Your power supply is under high stress.

[Sean_Goddard]

[ripsdevala]

Good day all,

I like to use 3 x 18650 batt in series to get 11v, is there an easy way by using charge module or create one instead?, but the charger should use standart 5v phone charger, is this possible?

Or i can use boost module with parrarel batt instead?