Cool simple "clean" CFA

Status
Not open for further replies.
Jump to Latest
#264
Prototype

Double Decker schematic "as built".
 

Attachments

  • @CF-FET-03-Double.jpg
    @CF-FET-03-Double.jpg
    235.3 KB · Views: 468
#268
Hi Valery and Jeff

Testing my CFA-CFP boards today and I have some issues. Did either of you do a schematic with voltages or an asc. file? I'm going over my parts again one by one to make sure they are correct but it would also be useful if I had at least a few places I could measure to help locate my error.

Thanks, Terry
 
#272
Hi Guys,

False alarm. I used TL072 instead of TL071. All is well in the universe. I will give it an audio test in the morning.

Blessings, Terry

Edit:
I have about .5V offset on initial turn-on which settles to 0V once the servo gets through. Is this normal?
 
Last edited:
#273
still4given said:
Hi Guys,

False alarm. I used TL072 instead of TL071. All is well in the universe. I will give it an audio test in the morning.

Blessings, Terry

Edit:
I have about .5V offset on initial turn-on which settles to 0V once the servo gets through. Is this normal?
Click to expand...

Hi Terry - ok, cool. 0.5V before the servo jumps-in is OK for this kind of heavily-symmetric CFA topology. Different tolerances lead to certain disbalance (you never know in what direction) - that's why the servo is very recommended here. It's possible to balance-out the offset with a trimmer, instead of using a servo, but over the time it may certainly drift. With the servo you can just forget about it.

One precaution - it is not recommended to connect/disconnect the input cables with the power on. Such a high-speed, wide-bandwidth amp can have pretty evil harmonics-rich transition noises at the output. Potentially not good for the speakers, in some cases leading to ptotection trip-out.
 
#274
Hi Valery,

That is good to know. Initially I had the wrong opamp and had. 5v on both boards. I pulled the opamps and still saw the same thing. After I installed the correct servo I watched it closely and that's when I noticed that it took the servo a few seconds to pull it down to 0v.I just wanted to make sure it didn't indicate a problem. I'm looking forward to hearing it. Jeff gave it glowing reviews. It was a fun amp to build and you did a great job on the screenprint so it is kind of a no-brainer. I can see why you and Jeff were surprised when I had issues. May I ask what you set the bias at?

Thanks, Terry
 
#275
still4given said:
Hi Valery,

That is good to know. Initially I had the wrong opamp and had. 5v on both boards. I pulled the opamps and still saw the same thing. After I installed the correct servo I watched it closely and that's when I noticed that it took the servo a few seconds to pull it down to 0v.I just wanted to make sure it didn't indicate a problem. I'm looking forward to hearing it. Jeff gave it glowing reviews. It was a fun amp to build and you did a great job on the screenprint so it is kind of a no-brainer. I can see why you and Jeff were surprised when I had issues. May I ask what you set the bias at?

Thanks, Terry
Click to expand...

Hi Terry,

Optimal quiescent current for this amp is around 35mA per pair - this is a specialty of CFP output stage topology. It runs pretty cool.

Cheers,
Valery
 
#276
OK, good to know. I did read through the thread again and saw that at one point but just wanted to make sure it didn't change again along the way. I will be listening to them as soon as my soon wakes up. So that I'm clear, that will be 15.4mV across a pair of 0R22 emitters?

One more thing, what is the "BIAS" plug for in the middle of the PCB? Can I just measure across that to set the bias?

Also, I created a spice file for this amp. Something is not quite right because I had to change R9 to 47R to keep it form clipping and the THD is a little higher than I think it should be. Maybe you can take a look and see where I may have an error.

Blessings, Terry
 

Attachments

#278
still4given said:
OK, good to know. I did read through the thread again and saw that at one point but just wanted to make sure it didn't change again along the way. I will be listening to them as soon as my soon wakes up. So that I'm clear, that will be 15.4mV across a pair of 0R22 emitters?

One more thing, what is the "BIAS" plug for in the middle of the PCB? Can I just measure across that to set the bias?

Also, I created a spice file for this amp. Something is not quite right because I had to change R9 to 47R to keep it form clipping and the THD is a little higher than I think it should be. Maybe you can take a look and see where I may have an error.

Blessings, Terry
Click to expand...

Terry, you're right - it's 15.4mV over a pair of 0R22 resistors.

BIAS plug is there for connecting an over-current sensor of the control board, but you can measure the above voltage drop right there.

The amp provides the "standard" 29db voltage gain with R9 = 18R.
Actual THD is very low. I will try to play with your model to see what's wrong with it.

Cheers,
Valery
 
#279
Hi Guys,

The forum seems to be struggling today. I first hooked it up to my scope with an 8R dummy load. for first tests I'm running it at +-36V rails. Sine waves look fine but square waves look good up to about 15khz. At 20khz there is a notch at the front edge of the top. At 50khz it looks like the teeth of a saw on the top and bottom of the squares and it get very distinct at 100khz. If it's important I can take some pictures of it. I played music through it and it sounded fine. Next I hooked it up to a +-50V PSU and now I hear a constant high pitched tone, maybe around 5khz or so. The is with the input shorted. I turned my scope up high and I see a little sawtooth looking signal. I turned the rail back down to +-36V and the tone goes away.

The board is populated per the schematic in post #126.

Any ideas?

Thanks, Terry
 
Status
Not open for further replies.
Top Bottom