I need to measure Hfe of a NPN power transistor with 5A of collector current. I found a lot of of circuits to measure Hfe but no one has this characteristic.
Hfe is Ic/Ib so you need to bias the transistor so it passes 5A (Ic) then measure the base current (Ib).
For a more accurate result, increase the bias current until Ic increases to say 5.1A, measure the base current again and use (delta Ic/ delta Ib).........
Details are very important:
You said "hfe which is AC current gain. Maybe you mean hFE which is DC current gain.
You also said "5A" but did not say 5A DC or 5A AC.
You will probably want to do this pulsed-mode or well-heat-sunk
(very) if you want both this Ic and higher Vce. For simplicity's
sake I think I'd use base drive for control and a resistor load,
AC couple it and look for the 0-Ic(set) difference perturbation.
You can find curve tracers / parameter analyzers which offer
a pulsed mode, but not all.
Perhaps you can see the graph of Ic vs Ib in the datasheet. This is not a straight line and the slope varies from point to point.
Locate the point corresponding to Ic =5A in the graph and measure the slope of the tangent at that point. That will give you the hfe at that value of Ic and Ib.
It is often tricky to measure slopes from a small size graph in a datasheet but you can enlarge and print it and take the slope manually. Perhaps a small error will not matter.
The graphs in a datasheet are for a "typical" device that you cannot buy. The hfe of hFE could be better or worse than typical and the printed spec's say the limits.
Do you want hfe AC current gain or hFE DC current gain??
Most of the devices you buy are considered "typical".
Parameter variations, although the distribution is not specified, are normally "normally distributed". This distribution has two tails and both tails are cut off by the quality control checks. These two cut off points are often specified in the same way: max and min. Usually the typical values are in the middle of the max and min values but formally they are the peak of the distribution.
Most of the devices you buy are "typical" and the stated parameter values will be close to the value specified and in any case lie between the max and min range. Critical parameters have tighter range (max-min= range) specifications. Less critical parameters, of course, have wider range values.
Measuring the slope manually is not accurate enough and that itself is subject to some error. But the values are to be used as a design guide only. Do you need hfe accurate to two decimals? A design that critically depends on a given datasheet parameter is a bad design.