We still have to see one of these beating AN-1645 with 2SK1530/2SJ201 even without output stage.
Why is national using this prototype-phobic package.
Why is national using this prototype-phobic package.
yes, but these mosfets have other drawbacks like thermal coefficient and are pretty nonlinear at the crossover region.A Anyways, you will never get a faked IRFP.
Two separated driving paths were designed in LME49830. It seems to be particularly used for driving MOSFET. 
I would like to ask what is the value of BiasP-BiasM voltage of LME49810. The datasheet states that the value is 10V. However, I can only measure about 4V 😕
My circuit schematic follows the figure 1 in the datasheet.
Thanks for your answer🙂
My circuit schematic follows the figure 1 in the datasheet.
Thanks for your answer🙂
10V is the max voltage. The voltage you will measure depends on the vbe multiplier and it is (aprox) 0.7*(Rb1+Rp+Rb2)/(Rb1+Rp) (aprox 3.2 V for the values on the typical application. The actual value will be slightly higher)
Try sticking just a 10K resistor or so from the bias pins and see what voltage you can measure. This is not the bias voltage you probably want to use as the Q current will be very high in the output stage but if you want to see how high the voltage CAN go then try this.
-SL
-SL
Thx the answer from SpittinLLama & ionomolo. 😀
According to ionomolo derivation when Rb1 and Rp are adjusted to zero, the value should be 10V. However, the result is not same as the calculation. The voltage is about 4V when I set Rb1 and Rp to be 0.
I guess there is so much current passing through the biasing pins causing the voltage drop. Therefore, I replace Rb2 by 33k and Rp by 100k then the biasing voltage will be same as the calculation.
Does anything wrong here? 😕
I think the biasing network can be a Mosfets driver for some large Vgs Mosfets so when the large Vgs Mosfets are used large resistor values should be used in biasing network to obtain enough biasing voltage.
According to ionomolo derivation when Rb1 and Rp are adjusted to zero, the value should be 10V. However, the result is not same as the calculation. The voltage is about 4V when I set Rb1 and Rp to be 0.
I guess there is so much current passing through the biasing pins causing the voltage drop. Therefore, I replace Rb2 by 33k and Rp by 100k then the biasing voltage will be same as the calculation.
Does anything wrong here? 😕
I think the biasing network can be a Mosfets driver for some large Vgs Mosfets so when the large Vgs Mosfets are used large resistor values should be used in biasing network to obtain enough biasing voltage.
An externally hosted image should be here but it was not working when we last tested it.
If you adjust Rb1 and Rp to zero another factor comes to play. Current from Sink to source is constant and equal to aprox 3 mA. So 1.21K * 3 mA = 3.63 V.
If you are going for those high voltages then probably you are planning to use a mosfet output stage. If you choose mosfets that don't require thermal compensation (namely 2SK1058 - 2SJ161) you may use a resistor to set the bias voltage (V = IBiasP-BiasM*R and check for maximum voltage as described by SpittinLlama
If you are going for those high voltages then probably you are planning to use a mosfet output stage. If you choose mosfets that don't require thermal compensation (namely 2SK1058 - 2SJ161) you may use a resistor to set the bias voltage (V = IBiasP-BiasM*R and check for maximum voltage as described by SpittinLlama
ionomolo Thanks for your reply 😀
Now, I am using mosfets to act as the output stage do your think LME49810 can drive Mosfet directly?
Here is my prototype 😀
Now, I am using mosfets to act as the output stage do your think LME49810 can drive Mosfet directly?
Here is my prototype 😀
An externally hosted image should be here but it was not working when we last tested it.
ionomolo
Thanks for your reply
Why the feedback loop is required to be short? Would you mind to explain more?
Thanks😀
Thanks for your reply
Why the feedback loop is required to be short? Would you mind to explain more?
Thanks😀
Two objectivist reasons:
If it's long it's more prone to pick noise at a high-gain input (there is no feedback at the other pin so the noise will be amplified at full open-loop gain. Instead of noise it may also couple parasitically to input/output/supply and make it start motorboating
It will also be more inductive and will introduce phase shift at high freqüencies which will increase distortion and decrease stability (with 20 MHz gain-bandwith product this is really important)
One subjectivist reason:
Others who have been working with the gainclone know that this affects the sound. The gainclone is a copy of an $3000 amplifier whose parts costed less than $100. The reason the designer told to sell it at that price was, among others, to have the world's shortest feedback path.
What i don't know about the LME49810 driving mosfets is wheter the baker clamp/clipflag works. The led in mine turns red when severely overdriven but maybe slightly after the onset of clipping.
If it's long it's more prone to pick noise at a high-gain input (there is no feedback at the other pin so the noise will be amplified at full open-loop gain. Instead of noise it may also couple parasitically to input/output/supply and make it start motorboating
It will also be more inductive and will introduce phase shift at high freqüencies which will increase distortion and decrease stability (with 20 MHz gain-bandwith product this is really important)
One subjectivist reason:
Others who have been working with the gainclone know that this affects the sound. The gainclone is a copy of an $3000 amplifier whose parts costed less than $100. The reason the designer told to sell it at that price was, among others, to have the world's shortest feedback path.
What i don't know about the LME49810 driving mosfets is wheter the baker clamp/clipflag works. The led in mine turns red when severely overdriven but maybe slightly after the onset of clipping.
LME49810 kit or pcb
Can forum members recommend a source with a quality circuit design for the LME 49810 (either kit with PCB and components or PCB with parts list capable of 2 ch and at least 100W/ch into 8 ohms when appropriately powered)?
Can forum members recommend a source with a quality circuit design for the LME 49810 (either kit with PCB and components or PCB with parts list capable of 2 ch and at least 100W/ch into 8 ohms when appropriately powered)?
Re: LME49810 kit or pcb
I am finalizing the design of a kit using the LME 49810 with Onsemi Thermaltraks scalable in power with two pcb's per compact heatsink ( forced air cooled) and pic controller supervision and alarms.
If interested, send me your email I can provide you with details on the performances, packaging and design philosophy behind
The initial objective is a group of 10 amps between 100 and 150 w / 8 ohms each to be used with Linkwitzlab Orion/Thor
the design is now more general and scalable in power
jpvanderreydt@skynet.be
JPV
marthaman said:
I am finalizing the design of a kit using the LME 49810 with Onsemi Thermaltraks scalable in power with two pcb's per compact heatsink ( forced air cooled) and pic controller supervision and alarms.
If interested, send me your email I can provide you with details on the performances, packaging and design philosophy behind
The initial objective is a group of 10 amps between 100 and 150 w / 8 ohms each to be used with Linkwitzlab Orion/Thor
the design is now more general and scalable in power
jpvanderreydt@skynet.be
JPV
I'm Interested
I e-mailed you earlier this morning also.
I would like to check out the boards you have available.
Thanks,
Donovan
I e-mailed you earlier this morning also.
I would like to check out the boards you have available.
Thanks,
Donovan