Looks better 
Now that you have huge groundplane surrounding everything you will need to make sure that you have enough room between traces that are NOT ground and the ground plane itself. I have been trying to find the webpage that told me how much clearance to give the traces so there would be not interaction between them and the plane...
Ah, here it is
http://www.desmith.com/NMdS/Electronics/TraceWidth.html
I just looked like for a possible 10Amps you might be putting them a bit close ... and I am assuming a 30degree rise above ambient is all that you would want. So you might want to plug some numbers in and make sure you dont get shorts or burns.
Uriah
Now that you have huge groundplane surrounding everything you will need to make sure that you have enough room between traces that are NOT ground and the ground plane itself. I have been trying to find the webpage that told me how much clearance to give the traces so there would be not interaction between them and the plane...
Ah, here it is
http://www.desmith.com/NMdS/Electronics/TraceWidth.html
I just looked like for a possible 10Amps you might be putting them a bit close ... and I am assuming a 30degree rise above ambient is all that you would want. So you might want to plug some numbers in and make sure you dont get shorts or burns.
Uriah
Half Power
Hi,
Is is possible to change the bias so that the F5 runs at half the power? Would it still be able to achieve the superior sound that it was originally designed for at full power?
I'm in a small room and probably don't need more than 10watts. My speakers are rating at 6 ohms, 90dB Sens. I'm running a
Trends Audio TA-10.1 and even that is enough power for me.
My goal is to conserve power since the electric bills keeps going up
What parts and values would need to be changed?
Thx.
Hi,
Is is possible to change the bias so that the F5 runs at half the power? Would it still be able to achieve the superior sound that it was originally designed for at full power?
I'm in a small room and probably don't need more than 10watts. My speakers are rating at 6 ohms, 90dB Sens. I'm running a
Trends Audio TA-10.1 and even that is enough power for me.
My goal is to conserve power since the electric bills keeps going up
What parts and values would need to be changed?
Thx.
The answer is no.
I suggest you read the paper "Leaving Class A" currently
featured at www.passlabs.com
I suggest you read the paper "Leaving Class A" currently
featured at www.passlabs.com
Re: Half Power
If 10W is enough, what about reducing the power supply voltage and keep the bias current to save energy...
Bengali said:Hi,
Is is possible to change the bias so that the F5 runs at half the power? Would it still be able to achieve the superior sound that it was originally designed for at full power?
I'm in a small room and probably don't need more than 10watts. My speakers are rating at 6 ohms, 90dB Sens. I'm running a
Trends Audio TA-10.1 and even that is enough power for me.
My goal is to conserve power since the electric bills keeps going up
What parts and values would need to be changed?
Thx.
Nelson Pass said:The answer is no.
I suggest you read the paper "Leaving Class A" currently
featured at www.passlabs.com
If 10W is enough, what about reducing the power supply voltage and keep the bias current to save energy...
Re: Half Power
Ummm.....
Why do you want to build a class A or A/B amplifier then? I used to own a 35w class A stereo integrated that easily heated a small house. I didn't listen to it during summer months because of it.
A Hypex, Tripath, or ICEPower solution is better for being "green".
Your Trends piece is very efficient yes? (tripath)
Best,
Chris
Bengali said:
Ummm.....
Why do you want to build a class A or A/B amplifier then? I used to own a 35w class A stereo integrated that easily heated a small house. I didn't listen to it during summer months because of it.
A Hypex, Tripath, or ICEPower solution is better for being "green".
Your Trends piece is very efficient yes? (tripath)
Best,
Chris
Thank you for your reply. I read the article Leaving Class A and have a slightly better understanding of why it's not possible.
This mainly due to the characterics of the FET and to avoid distortion requires the higher bias.
Since there has a been a Mini Aleph, I was hoping to see something similiar with the F5 using less power.
Do you have any plans to design some lower power amplifers, 5-15watt range? A Green Passlabs amp
I understand those who can afford the F series amp don't worry about their electric bill
This mainly due to the characterics of the FET and to avoid distortion requires the higher bias.
Since there has a been a Mini Aleph, I was hoping to see something similiar with the F5 using less power.
Do you have any plans to design some lower power amplifers, 5-15watt range? A Green Passlabs amp
I understand those who can afford the F series amp don't worry about their electric bill
Bengali,
Saving money on your energy bill will not be something you will notice with your choice of amps. Its likely that you will save more money by going ClassD simply because the heatsink will cost so much less and nothing else. If you pay lets say a whole lot for electricity then I think it might cost about 20 cents for you to run the F5 for 8 hours. For me I think it would cost about 13 cents to run it for 8 hours. About $15 for a full month of music.. nonstop at full blast. So depending on your utility costs it might cost 1.5 cents to maybe 3 cents per hour.
I am not judging your budget I am just guessing that there are more direct ways to impact your electric bill. There is a "significant" amount of savings if you simply turn down the heat/up the ac a few degrees. If you would put up with maybe 80 degrees instead of 70 in the summer and 65-65 instead of 70 in winter it makes a very large difference.
Of course with the F5 you may have to put up with 80 degrees in the summer anyway, but at least in the winter you can turn down the thermostat and increase the volume
Uriah
Saving money on your energy bill will not be something you will notice with your choice of amps. Its likely that you will save more money by going ClassD simply because the heatsink will cost so much less and nothing else. If you pay lets say a whole lot for electricity then I think it might cost about 20 cents for you to run the F5 for 8 hours. For me I think it would cost about 13 cents to run it for 8 hours. About $15 for a full month of music.. nonstop at full blast. So depending on your utility costs it might cost 1.5 cents to maybe 3 cents per hour.
I am not judging your budget I am just guessing that there are more direct ways to impact your electric bill. There is a "significant" amount of savings if you simply turn down the heat/up the ac a few degrees. If you would put up with maybe 80 degrees instead of 70 in the summer and 65-65 instead of 70 in winter it makes a very large difference.
Of course with the F5 you may have to put up with 80 degrees in the summer anyway, but at least in the winter you can turn down the thermostat and increase the volume
Uriah
A lower power version of the F5 would necessarily involve compromises in one or more areas...with one possible exception which I'll get to in a moment. The rails are currently specified as being +24V. That number happens to be important in terms of the behavior of the output stage in that MOSFET Gate capacitance (a major contributor to distortion) increases with decreasing Vds. That figure generally decreases with increasing rail voltage until you reach around 25V or so, at which point the curve begins to level off. Can you run the MOSFETs at lower rail voltages? Yes. But you should do so knowing that you're making a tradeoff.
High bias current is a major contributor to better sound quality. You can reduce the bias without too much trouble, but it won't sound quite as good.
The only way I can think of to lower the dissipation without running into the Gate capacitance problem is to get rid of the Gates. Use bipolar output devices. That would involve a bit of redesigning, but it might do the trick if you're willing to accept any other compromises that might crop up.
Grey
High bias current is a major contributor to better sound quality. You can reduce the bias without too much trouble, but it won't sound quite as good.
The only way I can think of to lower the dissipation without running into the Gate capacitance problem is to get rid of the Gates. Use bipolar output devices. That would involve a bit of redesigning, but it might do the trick if you're willing to accept any other compromises that might crop up.
Grey
Thanks Uriah.
It does not seem much but adds up very quickly overall. It would be nice to have a smaller power Class A amp. Having a small
room, there is no need for that extra power. Wasted energy and heat. And yes, heat is a big factor more than the electric bill.
smaller heatsinks, smaller transformer, the lower cost of the amp would certainly be better. This will be a winter amp for me
It does not seem much but adds up very quickly overall. It would be nice to have a smaller power Class A amp. Having a small
room, there is no need for that extra power. Wasted energy and heat. And yes, heat is a big factor more than the electric bill.
smaller heatsinks, smaller transformer, the lower cost of the amp would certainly be better. This will be a winter amp for me
Re: Re: Re: Half Power
I meant use the same bias current as the original F5 but reduce the power supply voltage. Since Power = voltage * current, if voltage is reduced than Power dissipation too. I do not understand why keeping the same bias current and lowering the voltage (example +/-15Vdc and re-ajusting some resistors) would affects negatively the sound?
Nelson Pass said:
Certainly you can bias the output stage as you please, but the criterion
was "still be able to achieve the superior sound that it was originally designed".
You could bias it higher, and it would sound a little better yet.
I meant use the same bias current as the original F5 but reduce the power supply voltage. Since Power = voltage * current, if voltage is reduced than Power dissipation too. I do not understand why keeping the same bias current and lowering the voltage (example +/-15Vdc and re-ajusting some resistors) would affects negatively the sound?
Re: Source resistors
I am trimming the JFET source resistors (10R) -- I put in 12R//68R =10.2R.
NP found a more dramatic influence trimming the MOSFET source resistors -- in a couple of weeks I will get around to doing this. I set up the PCB's so I could do both --- belt and suspenders I guess.
Right out of the box you should be able to get pretty remarkable THD% and listening pleasure if your design is well thought out, etc., etc.
mithomas said:
I am trimming the JFET source resistors (10R) -- I put in 12R//68R =10.2R.
NP found a more dramatic influence trimming the MOSFET source resistors -- in a couple of weeks I will get around to doing this. I set up the PCB's so I could do both --- belt and suspenders I guess.
Right out of the box you should be able to get pretty remarkable THD% and listening pleasure if your design is well thought out, etc., etc.
Re: Re: Re: Re: Half Power
It's them ol' Gate capacitance demons. Go get a datasheet for a power MOSFET and somewhere back in the charts section you're likely to find a graph of how Gate capacitance changes with the voltage across the device. The short version being that lower rails mean higher (and nonlinear) capacitance at the Gate. It decreases as the rail voltage increases, leveling off around 25V or so.
I used 15V rails in the Mini-A, but I did it with my eyes open.
Grey
fab said:
It's them ol' Gate capacitance demons. Go get a datasheet for a power MOSFET and somewhere back in the charts section you're likely to find a graph of how Gate capacitance changes with the voltage across the device. The short version being that lower rails mean higher (and nonlinear) capacitance at the Gate. It decreases as the rail voltage increases, leveling off around 25V or so.
I used 15V rails in the Mini-A, but I did it with my eyes open.
Grey
Re: Re: Re: Re: Re: Half Power
Thanks for the info.
I am used to deal with lateral mosfet where this data is not disclosed. Is it different with laterals?
I have the IRF244 dadasheet and what it shows is that each mosfet capacitance is effectively reduced with higher voltage. Higher VDS voltage is shown to be better, so 50V is even better than 25v. However, 25V is not a threshold per say. In fact there is not a dramatic change between let say 15V and 24V. 18V should not be that bad for rail voltage according to this criteria or am I missing something here?
GRollins said:
Thanks for the info.
I am used to deal with lateral mosfet where this data is not disclosed. Is it different with laterals?
I have the IRF244 dadasheet and what it shows is that each mosfet capacitance is effectively reduced with higher voltage. Higher VDS voltage is shown to be better, so 50V is even better than 25v. However, 25V is not a threshold per say. In fact there is not a dramatic change between let say 15V and 24V. 18V should not be that bad for rail voltage according to this criteria or am I missing something here?
Re: Half Power
It is possible to reduce bias current by paralleling additional resistors with R3 and R4 with relays, for background listening. You can put on a front panel a toggle-switch with RED LED indicating ''critical listening - global warming'' mode.
No other ill side effects than reduced sound quality.
I would even not think about reducing supply voltages.
Bengali said:
It is possible to reduce bias current by paralleling additional resistors with R3 and R4 with relays, for background listening. You can put on a front panel a toggle-switch with RED LED indicating ''critical listening - global warming'' mode.
No other ill side effects than reduced sound quality.
I would even not think about reducing supply voltages.