Hi all,
I have build a amplifier like this one
http://www.aussieamplifiers.com/lateral.htm
but i wonder why it is neccesary to use lateral mosfets ?
As i understand, the lateral mosfet is different from the vertical ones in the way they are build.
So can anyone explain why this amp dont work with ordinary, cheap mosfets ?
Kind regards,
Stormo
I have build a amplifier like this one
http://www.aussieamplifiers.com/lateral.htm
but i wonder why it is neccesary to use lateral mosfets ?
As i understand, the lateral mosfet is different from the vertical ones in the way they are build.
So can anyone explain why this amp dont work with ordinary, cheap mosfets ?
Kind regards,
Stormo
The main difference between these two types is the V-I characteristic. Lateral types starts to conduct at 0.2-0.4 V and other types at approx. 3 volts.
The other big difference is the gm, tranconductance, much higher in vertical types.
and ... lateral types has positive temp coeff above 200-400 mA (which you can use) and vertical has 5 A or so, useless!
If you use vertical mosfet's you must have a Vbe multiplier (mosfet or BJT) with 5-7 volts or so.
It's OK to use vertical types but beware of their higher cpacitance's which can cause you trouble in form of unstability. This will also reduce the speed of the amp is you don't increase maximum current from the driver.
As a conlusion: The amp WILL work with normal vertical mosfet's but you have to redesign a little.
The other big difference is the gm, tranconductance, much higher in vertical types.
and ... lateral types has positive temp coeff above 200-400 mA (which you can use) and vertical has 5 A or so, useless!
If you use vertical mosfet's you must have a Vbe multiplier (mosfet or BJT) with 5-7 volts or so.
It's OK to use vertical types but beware of their higher cpacitance's which can cause you trouble in form of unstability. This will also reduce the speed of the amp is you don't increase maximum current from the driver.
As a conlusion: The amp WILL work with normal vertical mosfet's but you have to redesign a little.
Hi Per-Anders,
thank you for your answer, it was most helpfull.
I now understand the issue different V-I characteristics, and i know how to use a VBE multiplier to solve the problem.
I also understand the temp coeff problem, but i dont understand how you get those information from the datasheets.
Now, if you place the VBE multiplier on the heatsink it will reduce the Bias current with rising temperature,- yes ? But if you run the amp very hard (and hot) how can this (little) change in bias current affect the temperature of the amp ?
Suppose it turns down bias a lot, would'nt it just introduce cross-over distortion, but dont save the amp from overheating ?
Kind Regards,
Stormo
thank you for your answer, it was most helpfull.
I now understand the issue different V-I characteristics, and i know how to use a VBE multiplier to solve the problem.
I also understand the temp coeff problem, but i dont understand how you get those information from the datasheets.
Now, if you place the VBE multiplier on the heatsink it will reduce the Bias current with rising temperature,- yes ? But if you run the amp very hard (and hot) how can this (little) change in bias current affect the temperature of the amp ?
Suppose it turns down bias a lot, would'nt it just introduce cross-over distortion, but dont save the amp from overheating ?
Kind Regards,
Stormo
Stormo said:
Vbe is a real tweaking thing with a real technical purpose, many smart solutions can be made. If the temp coeff is similar to the fet's (around -8 to-10 mV/deg C) you will get constant bias but do you use a BJT you could get class B when the heatsink is very hot.
The temp coeff can sometimes be read from the V-I graph. Normallly at least three curves are drawn, cold, normal and hot but the temp coeff in numbers are never presented. This you have to measure yourself
> Now, if you place the VBE multiplier on the heatsink it will reduce the Bias current with rising temperature,- yes ? But if you run the amp very hard (and hot) how can this (little) change in bias current affect the temperature of the amp?
It hardly affects the temperature of the amp.
But your amp idles warm, you play it LOUD and it get HOT from the power, then you turn it down. The transistors are still hot. What is the idle current?
If the bias voltage is fixed, the hot transistors will now have a very high idle current.
If the bias voltage is temperature corrected to drop when hot the same amount that the transistor Vb-e (or Vg-s) drops at desired idle current, then it idles about the same hot or cold.
Also, when idle current rises all the heat goes into the transistors. That is not the same sa playing loud, when most of the heat goes into the speakers. It is possible (though now rare) for an uncorrected circuit's idle current to increase, which makes it hotter, which makes the idle current increase more. "Run-away".
It hardly affects the temperature of the amp.
But your amp idles warm, you play it LOUD and it get HOT from the power, then you turn it down. The transistors are still hot. What is the idle current?
If the bias voltage is fixed, the hot transistors will now have a very high idle current.
If the bias voltage is temperature corrected to drop when hot the same amount that the transistor Vb-e (or Vg-s) drops at desired idle current, then it idles about the same hot or cold.
Also, when idle current rises all the heat goes into the transistors. That is not the same sa playing loud, when most of the heat goes into the speakers. It is possible (though now rare) for an uncorrected circuit's idle current to increase, which makes it hotter, which makes the idle current increase more. "Run-away".
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.