Time to design a all-round 5V metal detector?

[Teleno]

Li-ion power bank charger/regulator based on RT9480

Output peak is 2.5A at 5V

[kt315]

i did buy one with solar panel. but my idea was to fix another power bank that given me in repair just i can find nothing on last.

this i did buy https://ru.aliexpress.com/item/Unive...e-cb4842f9e359

[kt315]

this pcb i would have to fix, in result i found a cause.

http://storage8.static.itmages.ru/i/...9c16e21009.jpg
http://storage5.static.itmages.ru/i/...0c72c7c3d8.jpg

[homefire]

Why limit your self to 5Vdc ? LiIon packs can be set up in a number of voltages. With a Base Voltage of 3.7V you can do plenty of combinations.

3.7 x 2 7.4v works fine for most 9v applications.

3.7 x 3 11.1 works fine for most 12v application.

LiIon check out the LiIon battery discharge curve and you will find that they are quite flat and well within the Discharge curves of other common battery's for 6 , 9 and 12 volt requirements. Alkaline AA rated at 1.5V actually start out quite a bit higher then that. 1.62 volts individually so a single LiIon cell at 4.2-3.7v would be reasonable to replace a 3v AA pack.

6v battery pack could be replaced with 2 times 8.4 - 7.4v 4.2-3.7v LiIon again within safe limits I think.

12v battery pack could be replace with 2 time 4.2-3-7 12.6 - 11.1v LiIon again within most other battery discharge curves.

]Characteristics of Rechargeable Batteries - Texas Instruments

www.ti.com/lit/an/snva533/snva533.pdf


Just a Thought.

[Teleno]

Why limit your self to 5Vdc ?

For simplicity: Logic-level MOSFETs, 5V microcontroller, 5V Op-amps (TLC082, etc...), all powered by a USB rechargable battery.

[kt315]

guys take attention on mosfet A1SHB p-channel.
it was burnt, but i ordered them and wait
http://www.ebay.com/sch/i.html?_from...A1SHB&_sacat=0

[kosacid]

i did build a pinpointer using smt fets but not a full scale md so i don't know what depth it will go, i should test it

[bklein]

I've experimented with the RT9480 as well as other Richtek parts. What is nice about them is the fuel gauge. Believe me, you don't want to get into the quagmire of TI fuel gauges. It is a total nightmare.
But for your 5V project, this doesn't matter at all because you will be best off just buying power banks off ebay or Alibaba and connecting them to your detector - which will be your first design problem:
Such a power bank can get down to 4.3V or lower with rated output loads. You need more voltage margin. The TLC082 is rated to 4.5V Vmin. You will also want a more reliable field connection than a USB cable (think SAND).

I think you will want to stay with higher voltage battery sources; like a 11.1V lipoly RC battery.
You can use a 5V to 12V boost converter (Richtek has a good one) but it doesn't make sense to me to sway from just using a higher voltage battery.
If a PI circuit you won't use a logic level mosfet...

[Teleno]

Such a power bank can get down to 4.3V or lower with rated output loads.

According to the datasheet, the chip uses boost converters that can maintain 2.1A at 5V. Should be enough.

If a PI circuit you won't use a logic level mosfet...

I'll be using a logic level MOSFET in cascode configuration with a High-voltage one (see this thread) . Energy recovered in a capacitor provides the medium voltage required to bias the cascode top gate.

[bklein]

I've played with the chip and the RT9481 as well - using their reference boards as well as my own solar charger board.
I might be conservative with my 4.3V worst case but I'm sticking to it given not all power banks use the Richtek chip.
The RT9480 boost converter gets really hot boosting to 5.1V at high current load. Unless you have good heat sinking it will shut down at lesser than rated loads. You need good PCB design here.
Add to this that there is no sensing at your load point and you have losses at two connectors and the thin wiring in the cabling. For a detector likely the loading never gets this high so perhaps you have no issue?
I just don't agree with the premise that a detector using a 5V power bank is better than one using a dedicated battery. I think the parts count will be higher and detection no better. The circuit you refer to for the coil drive still uses a 12V source so I'm confused - are you going to add a 5V to 12V boost for this?

[Teleno]

I just don't agree with the premise that a detector using a 5V power bank is better than one using a dedicated battery.

I would be more efficient since Power = V x I. For the same I, a given number of cells last 2x longer.

I think the parts count will be higher and detection no better.

Part count needs not be larger because (1) no 5v regulators needed for the uC, (2) direct driving of logic level MOSFET, (3) the converter is part of the battery.

The circuit you refer to for the coil drive still uses a 12V source so I'm confused - are you going to add a 5V to 12V boost for this?

The circuit I refer to is a cascode with the energy recovery part removed. It's this part (basically a diode in series with the coil and a cap) that generates the 12V for the gate of the top MOSFET. In other words: the coil itself is the boost circuit that supplies the higher voltage.

Here's the 5V circuit:

[6666]

If you have a look at DaveJ's PI its almost all +5v.

[Teleno]

If you have a look at DaveJ's PI its almost all +5v.

As I see it, it's almost all +12V.

[6666]

Sorry for trying to help you.

[Teleno]

Sorry for trying to help you.

I'm sharing my solutions with you, people. Whining doesn't help anyone.