Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

Current source to simulate load transients

Status
Not open for further replies.
C

CAMALEAO

Full Member level 4
Joined
Jul 29, 2016
Messages
201
Helped
2
Reputation
4
Reaction score
2
Trophy points
18
Activity points
1,868
Hi all,

Can someone tell me what's the best way to have a current source to perform load transient simulations? I could use idc source but that's an ideal source and that could lead to massive voltage at its terminals.

Regards.
 

Why don't you use non-ideal current source ?? Use Norton equivalent ( a high valued parallel resistor )
 

Thanks for your kind reply. I thought about that but how big can that resistance be? I thought about using a current mirror as well - for example, but here we would be limited by it's bandwidth.

In addition to this the current source impedance should change with the current itself as well isn't it?

Do you implement your current sources like that?
 

I like a simple switched FET with a drain resistor to set
the current. "Active loads" (E-load, or multi-device
current source / sink) can add their own time- and
frequency-domain (mis)behaviors and these can be
hard to de-embed from the response. This of course
pertaining to the real world / bench.

A pulsed current source has infinite impedance. This
is unlike almost any real load. A switch and resistor
(or poly vccs) could be more realistic. What do you
know about the application load-space? Maybe you
want to "box it" with the ipulse and a switched
resistor, these being the max and min impedances
that a load of a given current could present.
 

You've told us what you don't want but no details about what you do want.

What frequency response do you need?
What exactly are the transient test conditions you need to generate (voltage, transient duration, pulse shape, current, etc.)?

Can't make good suggestions without knowing that information.
 

Sorry about that, you are right.

I intend to do load transients to an LDO. I want to make a load step at the output for several current value and combinations.

What you mean by pulse shape? If it is a pulse you are I directly saying that it it a pulse, like square-wave.

What you mean by transient duration duration? You mean the step duration? Like rise/fall time? I would say 10ns, 100ns,1us.

In addition to this I want to do the same to an amplifier (two stage) and to class AB amplifier.
 

CAMALEAO said:
..............
I intend to do load transients to an LDO. I want to make a load step at the output for several current value and combinations.

What you mean by pulse shape? If it is a pulse you are I directly saying that it it a pulse, like square-wave.

What you mean by transient duration duration? You mean the step duration? Like rise/fall time? I would say 10ns, 100ns,1us.

In addition to this I want to do the same to an amplifier (two stage) and to class AB amplifier.
Click to expand...
By pulse duration I mean how long is the load applied?

Do you really need variable rise and fall times?
If this is for a voltage regulator, the rise and fall times do not have to be extremely fast, since the response time of a regulator is generally relatively slow.

What type of amplifiers do you have that need a transient load test?
Normally that's only done for power supplies that see varying loads.
 

CAMALEAO said:
Thanks for your kind reply. I thought about that but how big can that resistance be? I thought about using a current mirror as well - for example, but here we would be limited by it's bandwidth.

In addition to this the current source impedance should change with the current itself as well isn't it?

Do you implement your current sources like that?
Click to expand...

A typical internal parallel resistor of a practical current source can be considered as 50k-100k Ohm.
Try with these values.
 

crutschow said:
Do you really need variable rise and fall times?
If this is for a voltage regulator, the rise and fall times do not have to be extremely fast, since the response time of a regulator is generally relatively slow.
Click to expand...

As a sometime designer of LDO ICs I'll disagree with this.
An interest in just how fast load-slew a LDO can maintain
regulation against, is common for things like FPGA I/O
supplies and the sensitive high speed serial resources'
supplies. Small signal analysis is not good for large signal
load steps. You need to challenge the regulator with
"realistic++" to see what it's capable of, but hyperabrupt
steps will just drive you crazy trying to fix an unfixable
"problem".
 

Só hat would be the best solution then? As suggested by onde person I can use a resistor in parallel. Is this tbe best approach bigboss?
 

I'd be happy to help if you answer my questions in post #5 about the voltage and current values of the tests which, for some reason, you seem to be reluctant to state.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top