How does this provide constant current ?

[nigelwright7557]

How does the constant current circuit work around R8 ?
Surely R8 should be a diode ?

An externally hosted image should be here but it was not working when we last tested it.

 

[jkuetemann]

Looks to be a current mirror degenerated with 100Ω resistors and not a current source at all.

 

[jcx]

yes, its not a ccs, it is a mirror - d1 would better be another of the same Q to match to tr5

 

[nigelwright7557]

yes, its not a ccs, it is a mirror - d1 would better be another of the same Q to match to tr5

I see now, the what would be the second transistor in a current mirror has been reduced to a diode ?
And what I thought was a constant current source is simply a 100R resistor at the bottom of the drawing.
I am used to seeing LTP the other way up, probably why I got confused.

 

[jcx]

transistors used to cost more

 

[sreten]

Hi,

I can't see much to like about that circuit, and would
suggest R13 and R14 are far too low to prevent
parasitic self destruction if it is run very hard.

330R ?

rgds, sreten.

 

[nigelwright7557]

Hi,

I can't see much to like about that circuit, and would
suggest R13 and R14 are far too low to prevent
parasitic self destruction if its run very hard.

330R ?

rgds, sreten.

What I have done is take this circuit as a basis for one of my designs.
It is an old Hitachi lateral mosfet circuit where the mosfets are very rare now.
I have changed outputs to vertical mosfets.
I have added a Vbe multiplier.
The mosfet gate resistors are now 390R.

 

[wahab]

I see now, the what would be the second transistor in a current mirror has been reduced to a diode ?
And what I thought was a constant current source is simply a 100R resistor at the bottom of the drawing.
I am used to seeing LTP the other way up, probably why I got confused.

This design has a flaw that is easy to spot, the LTP current is defined by R3; hence the voltage across R4 and R5 will vary in proportion of the supply voltage as well, this will increase accordingly the current through TR3/TR4 and thus through the bias setting pot wich will increase the power stage iddle current, the 47K resitance that load the LTP emitter should be replaced by a constant current source as all the standing currents in the amplifier are dependant of the LTP current, and will vary in function, of the LTP standing current, besides using a current source will extend greatly the voltage supply range.

I guess that it s one of those Maplin amplifiers of old based on the Hitachi app note.

 

[nigelwright7557]

This design has a flaw that is easy to spot, the LTP current is defined by R3; hence the voltage across R4 and R5 will vary in proportion of the supply voltage as well, this will increase accordingly the current through TR3/TR4 and thus through the bias setting pot wich will increase the power stage iddle current, the 47K resitance that load the LTP emitter should be replaced by a constant current source as all the standing currents in the amplifier are dependant of the LTP current, and will vary in function, of the LTP standing current, besides using a current source will extend greatly the voltage supply range.

I guess that it s one of those Maplin amplifiers of old based on the Hitachi app note.

I also added 100R and 100uf on each supply rail. This was because the original design suffered a bit with hum on no signal input.

Yes the Maplin mosfet 75W/150W amplifier was based on the Hitachi datasheet. I believe a lot were sold.