I have a very cool question to ask from you:
For some time im interested about RAM. More as a curiosity.
I've read this page: Magnetic-core_memory and also watched a couple of youtube movies to better clarify for me. And basically, one BIT of memory, is made of 1(one) ferite ring(shape), 2 wires for address at 90degree crossing through the donut hole, 1 diagonal wire that is Sense and Inhibit in the same time (sense when read, inhibit when write). The (magnetic) polarity of the ferrite gives the state of the Bit, if its right is On, if is left is Off. This is also a destructive readout = when the read action is performed, it is erased in the same time. When not being read or written, the cores maintain the last value they had, even if the power is turned off. Therefore they are a type of non-volatile memory.
But this is A TYPE of memory bit and is OLD technology. I want to know how Modern Memory are made of. Understand my question well. Im not interested in how is accessed/addressed, but how is built and what material really stores the electric charge? From my general knowledge, in modern microcips are transistors, diodes, resistors and capacitors, all in order of millions inside the cip circuit . But my question is not the circuit, but what is keeping the charge? I always presumed, because i didnt find anything on this subject good enough, maybe its the circuit itself? I always assume there may be some capacitors that keep the charge inside the circuit of the memory cip. How would look the simplest circuit ? (if it turns out to be the circuit or the capacitor holding the charge as i imagine). The basic circuit.
Thank you !

 

There are several different types of memory and thus several different ways of storing the state information. Some memories use active circuits such as flip-flops to hold the information. A simple form of this is a pair of cross-coupled inverters with one strong inverter and one weak inverter. To change the state the output of the weak inverter is overdriven by a stronger input signal. Other memories store information on a capacitor and then sense whether the voltage is above or below certain thresholds. The read operations on this, like core memory, tend to disturb the value stored and so they need to be refreshed. They also tend to degrade over time and need to be periodically refreshed whether they've been read or not. Other memories use charge that gets driven into a insulating layer by a higher than normal voltage and it thus gets trapped there. This trapped charge is then sensed through some means -- and there are many. All you need is to be able to detect some property of the material that is different when it is in one state verses the other.

 

There are at least two fundamental types of random access semiconductor memory (RAM); they are called static and dynamic. They are both volatile; when the power goes off they lose their contents. A static RAM cell is basically a flip-flop that has two stable states. It takes about six transistor equivalents to make one memory cell. A dynamic RAM cell is much simpler in terms of transistor equivalents and it basically traps some amount of charge on a device, but this charge has to be periodically refreshed or it will "leak" away.

https://en.wikipedia.org/wiki/Static_random-access_memory
https://en.wikipedia.org/wiki/Dynamic_random-access_memory
https://www.allaboutcircuits.com/te...duction-to-dram-dynamic-random-access-memory/

 

Very cool answers, thank you both !

 

what is keeping the charge?

Charge on a gate. This method is used in DRAM, EPROM, EEPROM, and FLASH. SRAM use a 6 transistor memory cell that's essentially two antiparallel inverters. NVRAM (like X2210) use SRAM backed up by EEPROM.

DRAM, EPROM, EEPROM, and FLASH use one transistor per bit. In FLASH, they now have the capability to store 2 or 3 bits of information on a single gate by controlling the number of electrons that are stored. Early EPROMs used to just store a lot of charge on the floating gate. Later iterations used a closed loop system to insure program margin.

 

I have a basic 1bit memory circuit SRAM. This is more a didactic example.
What i don't understand is the part where he says: "The voltage on the collector of transistor (A or B) is less than 0.6V". Is like it is inhibiting somehow. Can you explain this part, please? Also, super answers so far.

 

What i don't understand is the part where he says: "The voltage on the collector of transistor (A or B) is less than 0.6V". Is like it is inhibiting somehow.

The voltage is actually lower than 0.6V, but he's trying to explain, in an awkward way, that the transistors are off if the voltage on the base is less than 0.6V. Though this isn't strictly true. Transistors will start turning on at 0.6V.

There aren't many memory cells that use an LED... Yet another "so-called" expert...

 

What he means to say is the voltage at the collector of transistor B that is fully conducting presents a voltage that is below the base turn-on voltage at transistor A when switch A is released. This is positive feedback which causes the flip-flop to stay latched in a stable state. This circuit is also known as a Set-Reset flip-flop.

 

I very much agree with everything you said so far.
Thank you very much - you really helped me clarify this problem.