http://www.diyaudio.com/forums/pass-labs/121228-f5-power-amplifier-56.html#post1740824
Increased gain from 6 to 8.5.
Won't save the day.
Patrick
Increased gain from 6 to 8.5.
Won't save the day.
Patrick
R6 and R7 bite the dust
I took out R6 and R7. Mr. Pass has a post on this and says you loose very little. It's not really clear to me what I lost vs. what I gained.
If this simple change in the feedback resistors is so helpful and has so little downside, why not make it the original design? There must be an issue or a lot more people would do it.
If someone understands the real changes in removing R6 and R7, I'd appreciate understanding it better.
Also, I'm very willing to add a preamp to my setup if that will get me to the same place.
Advice is very welcome.
I took out R6 and R7. Mr. Pass has a post on this and says you loose very little. It's not really clear to me what I lost vs. what I gained.
If this simple change in the feedback resistors is so helpful and has so little downside, why not make it the original design? There must be an issue or a lot more people would do it.
If someone understands the real changes in removing R6 and R7, I'd appreciate understanding it better.
Also, I'm very willing to add a preamp to my setup if that will get me to the same place.
Advice is very welcome.
Hi,
with the F5 gain of 6 and source maximum signal levels of 500mVac to 2.2Vac you need some additional gain.
I would look at a switchable gain stage of +6dB (2times) and +12dB (4times).
This brings the F5+Gain stage to 12times or 24times and this suits most sources you are likely to attach.
Juma's BF862 pre-amp or Perander's gain stage (smaller than a postage stamp) or Russ White's Diamante, or an opamp will all do the job.
with the F5 gain of 6 and source maximum signal levels of 500mVac to 2.2Vac you need some additional gain.
I would look at a switchable gain stage of +6dB (2times) and +12dB (4times).
This brings the F5+Gain stage to 12times or 24times and this suits most sources you are likely to attach.
Juma's BF862 pre-amp or Perander's gain stage (smaller than a postage stamp) or Russ White's Diamante, or an opamp will all do the job.
Last edited:
Thanks. I'll look at all of them.
I'm not sure about the idea of bridging, or using a balanced configuration. If it's not too repetitive, can someone explain how the gain would change under those circumstances? I'm not really sure how a balanced configuration gets one more gain for a SE source.
I'm not sure about the idea of bridging, or using a balanced configuration. If it's not too repetitive, can someone explain how the gain would change under those circumstances? I'm not really sure how a balanced configuration gets one more gain for a SE source.
The hot signal arrives with a max of 1Vac.
That gets multiplied by the gain of 6times to become 6Vac max at the hot output.
The cold signal arrives with a max of 1Vac, again multiplied to become 6Vac at the cold output.
The speaker is connected across hot to cold. The speaker sees 12Vac.
The balanced amplifier and the balanced input signal results in twice (+6dB) the voltage across the speaker.
That gets multiplied by the gain of 6times to become 6Vac max at the hot output.
The cold signal arrives with a max of 1Vac, again multiplied to become 6Vac at the cold output.
The speaker is connected across hot to cold. The speaker sees 12Vac.
The balanced amplifier and the balanced input signal results in twice (+6dB) the voltage across the speaker.
So does that mean that two F5 boards, one configured to recieve the + signal and one configured to recieve the - signal, would result in 200 watts output?
Single F5 = 50 watts out, two F5s = 50w +6dB (200w).
Nope, voltage losses go up with power, times two for bridged/balanced acts.
200W in 8 is 5A, some 2.4V loss across the source resistors.
The current runs through 2 source resistors simultaneously in bridged mode : almost 5V loss.
For peak output values, the voltage loss would go up ~40% more.
Secondly, distortion level goes up for lower impedance values.
In bridged mode, each of the amps looks at 4 Ohm instead of 8.
In theory 4 times, in practice a factor ~3.
200W in 8 is 5A, some 2.4V loss across the source resistors.
The current runs through 2 source resistors simultaneously in bridged mode : almost 5V loss.
For peak output values, the voltage loss would go up ~40% more.
Secondly, distortion level goes up for lower impedance values.
In bridged mode, each of the amps looks at 4 Ohm instead of 8.
In theory 4 times, in practice a factor ~3.
Hi,
the F5 gives about 27W into 4r0 and about 50W into 4r0. For 8r0 all the output is ClassA, for 4r0 ~25% is ClassA
If you bridge them or build as a balanced pair then you get double the power into double the impedance. This rule works for both theory and practice.
The F5+5 should give about 54W into 16R and 100W into 8r0, for 16r all the output is ClassA, for 8r0 ~25% is ClassA.
the F5 gives about 27W into 4r0 and about 50W into 4r0. For 8r0 all the output is ClassA, for 4r0 ~25% is ClassA
If you bridge them or build as a balanced pair then you get double the power into double the impedance. This rule works for both theory and practice.
The F5+5 should give about 54W into 16R and 100W into 8r0, for 16r all the output is ClassA, for 8r0 ~25% is ClassA.
Last edited:
Today I modified my Mini F5 (4 channels to drive my midrange horns and tweeters) and I replaced the IRFP9240/IRFP240 by 2SK2013/2SJ313.
I also changed the gatestoppers to 220R and use reduced feedback (100R).
Thats it! I got the missed treble back and can replace my t-amps now.
I run the Mini F5 at +-15VDC with a idle current of 0.6A.
Everything else is standard F5.
The mains transformer I use now is one 225VA type to drive all four channels. The tranny should be able to handle this 2.4ADC and still have some reserve.
And you may not believe it: I dont need to parallel the output FET's, as the power is absolutely sufficient for my high efficency speakers!
Franz
I also changed the gatestoppers to 220R and use reduced feedback (100R).
Thats it! I got the missed treble back and can replace my t-amps now.
I run the Mini F5 at +-15VDC with a idle current of 0.6A.
Everything else is standard F5.
The mains transformer I use now is one 225VA type to drive all four channels. The tranny should be able to handle this 2.4ADC and still have some reserve.
And you may not believe it: I dont need to parallel the output FET's, as the power is absolutely sufficient for my high efficency speakers!
Franz
Last edited:
In the meantime I asked and I got from Papa Nelson permission to publish the picture of my Mini F5 with chinese pirate boards:
On the right side, I already installed another set of FET's to parallel the output. But I dont think it is needed for my speakers.
Franz
P.S.
I modified this boards in some aspects to satisfy my needs. I dont encourage to buy another of this boards. Please look at the offers from Peter Daniel or cviller.
An externally hosted image should be here but it was not working when we last tested it.
On the right side, I already installed another set of FET's to parallel the output. But I dont think it is needed for my speakers.
Franz
P.S.
I modified this boards in some aspects to satisfy my needs. I dont encourage to buy another of this boards. Please look at the offers from Peter Daniel or cviller.
Last edited:
Just to show
The four channels of the Mini F5 are driving the horns (mid and tweeter)
And soon, the four FW250 midwoofers will be driven by a full F5
Franz
/Edit
This picture is some years old, before refurbishing of the speakers. Yes, it was in my livingroom...
The four channels of the Mini F5 are driving the horns (mid and tweeter)
An externally hosted image should be here but it was not working when we last tested it.
And soon, the four FW250 midwoofers will be driven by a full F5
Franz
/Edit
This picture is some years old, before refurbishing of the speakers. Yes, it was in my livingroom...
Last edited:
Troubleshooting F5 components
My apologies if this has been covered but my search has not helped.
After originally getting a good bias setting on my left channel, I made a dumb error and cooked some components. In particular, R111/R112 on cviller v2.0 board got red hot and other components seem to have died.
I am now am trying to isolate the bad components and would love to do this without unsoldering everything on the board. However, I'm getting different resistance readings between the good board and bad one in a number of places so I'm not making much progress.
Has anyone posted any guidance on this? E.g., the expected resistance values at different points or how to check the active components.
Sorry if this is a dumb question, I'm new at this.
Any help is appreciated.
My apologies if this has been covered but my search has not helped.
After originally getting a good bias setting on my left channel, I made a dumb error and cooked some components. In particular, R111/R112 on cviller v2.0 board got red hot and other components seem to have died.
I am now am trying to isolate the bad components and would love to do this without unsoldering everything on the board. However, I'm getting different resistance readings between the good board and bad one in a number of places so I'm not making much progress.
Has anyone posted any guidance on this? E.g., the expected resistance values at different points or how to check the active components.
Sorry if this is a dumb question, I'm new at this.
Any help is appreciated.