Correct location for VBE bias transistor on CFP output amplifier

[AndrewT]

i am using two pairs , but i have faced the same problem with one pair also.. might be I am using 45V rails and a 4E load.. what should be the voltage across R9 ..?? and how do i increase the bootstrap current

R9+R10 set the VAS current.
Ivas = Vrail/ (R9+R10)
note that VAS current changes with supply rail voltage !
As soon as you start to push some output current into your load the supply rail voltage will fall and that reduces the VAS current.

If your 4ohms load demands current, the output devices will place a demand on the driver stage. That places a demand on the VAS stage, since you have no pre-driver.
Can your VAS stage supply sufficient current to meet the current demands of a reactive 4ohms speaker load?

 

[AndrewT]

I calculated same way, as the current consumption calculated in class B amplifier (bias current neglected). My idea is that the emitter resistors passing same current as the amplifier get from one power rail. Simple rms calculation is no applicable for this.

Sajti

The currents in the supply rails are not the same as the current in the load.
I don't think you can apply averaging of supply rail currents to determine supply rail resistor dissipation and then compare that to sinewave current passing to the load via the two emitter resistors.

 

[AndrewT]

Can SMD resistor 2512 package be used for current sharing purpose , I have used four 1E resistor in parallel to form the 0.25E in the emitter. does SMD resistors fails in long term usage in high temp situation ??? Mostly people used ceramic encased wire wound resistor for this type of application , is it only for cost reasons or due to the higher pulse power rating of them ....and aint the Wire wound resistors inductive in nature.. I have also noticed in high power amps when you do sine wave testing the resistors start producing a audible noise i.e the wire wound resistors start vibrating at the frequency of the test frequency ...

There has been some discussion on overload current capabilities of smd vs round ceramic metal film resistors.
It seems that the flat plate ceramic smd resistors have a lower tolerance to overload than does the helical metal film adhering to the tubular ceramic substrate.

The average dissipations in the emitter/source resistors can be quite low and on that basis one would think that maximum dissipation is never a concern. Unfortunately the peak transient current gives a peak instantaneous dissipation that can effectively be an overload situation. You would really need to delve into the datasheets to find which smd resistor would tolerate those transients.

 

[sekhar]

R9+R10 set the VAS current.
Ivas = Vrail/ (R9+R10)
note that VAS current changes with supply rail voltage !
As soon as you start to push some output current into your load the supply rail voltage will fall and that reduces the VAS current.

If your 4ohms load demands current, the output devices will place a demand on the driver stage. That places a demand on the VAS stage, since you have no pre-driver.
Can your VAS stage supply sufficient current to meet the current demands of a reactive 4ohms speaker load?

hi andrew ,
i tried lowering the resistor value to 2.7k on both R9 & R10 ..but the same happens ... I also tried lowering the DC supply rail to 35V and it still had the weird artifact , i even tried one output pair ...but its all the same...

 

[sekhar]

The differential voltage across R9 should be about 22V and steady across the audio range for 45V rails.

Before we proceed, can you measure DC coupled Vout on CH1 and the DC coupled voltage of the negative rail on the same scale on CH2.

hi Mt490,
i measured the Dc rails its +-45V no load and +-40 V loaded and across R9 its 20V at load and 22V at no load ...

 

[AndrewT]

hi andrew ,
i tried lowering the resistor value to 2.7k on both R9 & R10 ..but the same happens ... I also tried lowering the DC supply rail to 35V and it still had the weird artifact , i even tried one output pair ...but its all the same...

how do i increase the bootstrap current

I answered your question on VAS current and how to adjust.

The glitch as the amp comes out of clipping is a different problem. You need to look at the AC behaviour to investigate that.