Level Shifter of sorts?

Thread Starter

MikeJacobs

Joined Dec 7, 2019
226
So I built this thing which may or may not be common.

I wanted some feedback on the concept of this schematic. The parts choices , voltages , resistor sizes etc etc were just a quick choice to get the circuit functioning. I fully understand if you were to use this thing a different part number here or there and better biasing is needed. I just wanted to talk about the concept.

Essentially want to take a source that has very little drive capability and at a smaller voltage and give myself the ability to drive several small loads.
When I say small loads, I am simply referring to something like a very low current. Perhaps a pin on a chip that needs to be held high.
Maybe you have hundreds of pins from hundreds of discrete IC's that need to be held high at once. Say these hypothetical pins only draw uA's each but you want to drive 50 of them at once.

So you are doing two things with this.
You are using a 3v drive to create a 5v level shift and giving your self the sourcing current that the p channel mosfet can supply.
Also, I will add, I specifically did not want the inversion.

Just was hoping to get some conceptual thoughts.
Thanks
 

Attachments

Last edited:

jpanhalt

Joined Jan 18, 2008
11,087
There's a lot on level shifting. This is one of the simpler bi-directional ones:
1576900569342.png
Source: SparkFun

Unidirectional are even simpler, basically a diode and resistor.

What advantages does your shifter offer?
 

Thread Starter

MikeJacobs

Joined Dec 7, 2019
226
There's a lot on level shifting. This is one of the simpler bi-directional ones:
View attachment 195002
Source: SparkFun

Unidirectional are even simpler, basically a diode and resistor.

What advantages does your shifter offer?
I don't know the answer to that which is why I wanted to discuss it.
There may not be any advantages lol. I simply wanted to hear if people think it will work
 

Thread Starter

MikeJacobs

Joined Dec 7, 2019
226
I know there are a million other ways to level shift.
There are hundreds if not thousands of IC's all with certain disadvantages like source current limitations.
but what I am hoping to do is stimulate some thought on this particular configuration
 

jpanhalt

Joined Jan 18, 2008
11,087
I don't know the answer to that which is why I wanted to discuss it.
There may not be any advantages lol. I simply wanted to hear if people think it will work
Have you tried simulating it? LTSpice is free and pretty good.

If you are just going 3.3 to 5V at very low current and don't mind an inversion, why do you need the mosfet?
 

Thread Starter

MikeJacobs

Joined Dec 7, 2019
226
Have you tried simulating it? LTSpice is free and pretty good.

If you are just going 3.3 to 5V at very low current and don't mind an inversion, why do you need the mosfet?
I can't have the inversion that's why I did it this way
Yes, simulated it. It seems to work well In the sim as of now but I see the sim sometimes not demonstrate things that real if bread boarding does
 

Thread Starter

MikeJacobs

Joined Dec 7, 2019
226
It works but you can do it with less current by using an N channel instead of the BJT.
Interesting, so your thinking N channel on the left instead of the BJT.

Your thinking the base current of a bit is greater then the gate current on a mosfet ?

I thought the load of the gate on an n Chanel was microamps also?
Im curious to hear more from you.

Thanks for the input thus far
 

Sensacell

Joined Jun 19, 2012
3,432
Every shifter configuration has it's advantages.

In reality, you match the configuration to the situation at hand, based on things like:

Cost
Complexity
Size
Propagation Delay - speed /symmetry
Power consumption
Invert/no invert
Noise immunity (how well does it meet the voltage specs of the things it's connecting
Bi-directional?
 

atferrari

Joined Jan 6, 2004
4,764
Prior coming here I would breadboard it and be able to add my conclusions to the discussion.

Will try to retrieve my own version that I used cannot recall where.

/Edit
Sorry; you built it!
Edit/
 
Last edited:

MisterBill2

Joined Jan 23, 2018
18,176
I have used an IC level shifter, the CD4009, and then a whole lot more the CD4049, which both are inverters. For one standard product of ours I used two transistors as a level shifter to switch the negative 5 volts of a small laser with a positive 5 volt signal from a CD4013B.
To design any level shifter the first requirement is to know both the load and the driver. Do they sink current, or do they source current, and what voltages they require. And always research as to what IC devices are available to do the shift you require. And if it is for an item to be produced for a while, always verify that there is reliable sourcing, which usually means multiple sources.
 

Wolframore

Joined Jan 21, 2019
2,609
The mosfet vs BJT efficiency depends on what your final use is. If you’re spending much of your time in one state or other, it makes a huge difference. If you’re switching a very active serial interface perhaps not as much... it will depend on your final application. I favor mosfets. You will have miller capacitance on the BJT as well due to the CE bias at the BC junction. Do you have a switching frequency and other details?
 

jpanhalt

Joined Jan 18, 2008
11,087

Thread Starter

MikeJacobs

Joined Dec 7, 2019
226
The mosfet vs BJT efficiency depends on what your final use is. If you’re spending much of your time in one state or other, it makes a huge difference. If you’re switching a very active serial interface perhaps not as much... it will depend on your final application. I favor mosfets. You will have miller capacitance on the BJT as well due to the CE bias at the BC junction. Do you have a switching frequency and other details?
thanks for the thoughts
The switching frequency will be very infrequent
Say one discrete switch for extended periods of time. Hours even

other then that this type of design would just stay high or low
 

jpanhalt

Joined Jan 18, 2008
11,087
So does anyone have have any specific concerns about the schematic I posted?
It's a lot of parts to accomplish one-way level shifting. In post #2, I mentioned other resistor diode designs. Here is what XBee uses in its products:

1576941819382.png

The same circuits are also in a Tips 'n Tricks application note from Microchip. For the 3V to 5V translation, R1 needs to be smaller than the input impedance of the 5V device.

However, if it works for you, and you have enough space, use it. None of us have enough information to answer that question, as we do not know your application.
 
Top