Sensitive horns don't need a lot of signal to go loud.Thanks mcandmar. I'm very biased now.
I've moved the F4 amp to my main system on the horns (I bi-amp with a First Watt B4). Will post about sound quality later.
I was expecting to have to move to the 6db mode of my preamp from the 0db buffer mode. But no need at this point, it seems. I'm running the volume dial at 11:00 o'clock to 1 pm. Which is kind of amusing since the amp itself has no gain, of course.
My DAC puts out 2v, but I am pre-processing with Acourate DSP in software which knocks about 7db off the signal. So it isn't even a 2v max signal out.
My current amps for the horns, gainclone monoblocks and 300b, have 30db of gain so this is a big improvement just from that standpoint, and was a main thing I was looking to improve.
The amp is the quietest I've experienced on horns - just a touch of noise with the ear in the horn (and maybe that is from the tube preamp or other components).
I want to try the amp on my bass/lower midrange channel, but I'll need to finish the K&K 4P1L gain stage I'm working on first.
But do need a very quiet amplifier to avoid noise.
The easy way to meet both targets is to use the lowest gain you can manage and still be just loud enough.
I'm really enjoying my new F4 amp. Kudos to 6L6 for making the build process smooth and fun, and very much to Nelson too!
Here's my four word review summary:
More resolution, less distortion.
For example, I am noticing the upper range of violins sounds more realistic.
The amp tonality does seem to be hypersensitive to power cord and AC outlet. At some point I'll experiment with snubbers on the transformer secondaries. I'm happy with the tone now, though, with my $8 original Volex cord and Furutech outlet.
Here's my four word review summary:
More resolution, less distortion.
For example, I am noticing the upper range of violins sounds more realistic.
The amp tonality does seem to be hypersensitive to power cord and AC outlet. At some point I'll experiment with snubbers on the transformer secondaries. I'm happy with the tone now, though, with my $8 original Volex cord and Furutech outlet.
Hi Cappy,
The general consensus i got here on the class a amps when i was building mine is to not use snubbers. Apparently they have no place in a class a amp. Anyway,
I was curious as to the power cord and supply requirements. I had been having intermittent noise trouble with a preamp (when its not doing its thing its sounds stellar and every now and then i get some wierd rustling coming through one of the channels) i made for my F4 and 1 of the routes i went down was adding internal filtering units and better cables. I used some kits from mains cables are us and they are the thick beldon cable with the drain wire and the ferrits. I replaced the dac, transport, pre and power and plugged into a tacima plug board. This replaced dirt cheap standard cables and non filtered plug board.
Im not sure i can hear any difference or at least its very subtle.
What improvements did you find? Its sounds from your wording that it was quite substantial.
Cheers.
The general consensus i got here on the class a amps when i was building mine is to not use snubbers. Apparently they have no place in a class a amp. Anyway,
I was curious as to the power cord and supply requirements. I had been having intermittent noise trouble with a preamp (when its not doing its thing its sounds stellar and every now and then i get some wierd rustling coming through one of the channels) i made for my F4 and 1 of the routes i went down was adding internal filtering units and better cables. I used some kits from mains cables are us and they are the thick beldon cable with the drain wire and the ferrits. I replaced the dac, transport, pre and power and plugged into a tacima plug board. This replaced dirt cheap standard cables and non filtered plug board.
Im not sure i can hear any difference or at least its very subtle.
What improvements did you find? Its sounds from your wording that it was quite substantial.
Cheers.
Greetings Mr. Dave,
Yeah, substantial. Talking about the upper register of violins again, I can change the tone from bright and glare-y, to just right, to closed in and dark sounding just with power cord and outlet changes.
Thanks for info on Class-A amps and snubbers. I'll do more research on that. In any case, I feel no need to change anything, it's a great amp...
Yes! Im blown away with mine.
I wonder if the horns particularly reveal where the cables make a difference.
I built a pair of troels gravesen OBL15s open baffles and whilst they're very balanced they dont seem quite as crisp sounding as my pmc monitors so maybe they're not as sensitive round that particular frequency band but i certainly wouldnt describe them as dull.
Which furutech conditioner are you using?
I wonder if the horns particularly reveal where the cables make a difference.
I built a pair of troels gravesen OBL15s open baffles and whilst they're very balanced they dont seem quite as crisp sounding as my pmc monitors so maybe they're not as sensitive round that particular frequency band but i certainly wouldnt describe them as dull.
Which furutech conditioner are you using?
It's not a Furutech conditioner, it's a bejeweled Furutech GTX-D Gold receptacle. I do use a UberBuss power conditioner, and the Furutech receptacle is installed in that.
Early this year I recycled my old original Zen amp and built an F4 into the
chassis. Unable to build a new mating preamp at the time, I decided to put it
away.
I thought of it again and over the weekend, took it out, adjusted the bias
a bit and hooked it up, directly to my cd player and speakers. On some tracks it was
actually just loud enough though most definitely could use some gain. (For
almost two years I have been running without a preamp, driving amps
directly from cd output.)
So I powered up my old Sonographe sc22 (had it since early 90's) and
connected the F4 to it. It was interesting to be able to turn the volume up
pass the 12 o'clock position without harm. The sound was very lively and
on some materials seemed almost too bright.
I'll have to listen again this weekend once everything is warmed up. (The preamp
is meant to be powered on continuously.)
Minor technical details on the F4: 400VA Antec transformer, 8 x 22000 uF in
CRC power supply. Panasonic caps on the F4 itself, using Peter Daniel PCBs.
The mosfets are Fairchild FQA19N20C and FQA12P20.
Eventually a new preamp will be needed since this one does not have adequate
voltage swing for the F4 (max 10V RMS).
Cheers,
Dennis
chassis. Unable to build a new mating preamp at the time, I decided to put it
away.
I thought of it again and over the weekend, took it out, adjusted the bias
a bit and hooked it up, directly to my cd player and speakers. On some tracks it was
actually just loud enough though most definitely could use some gain. (For
almost two years I have been running without a preamp, driving amps
directly from cd output.)
So I powered up my old Sonographe sc22 (had it since early 90's) and
connected the F4 to it. It was interesting to be able to turn the volume up
pass the 12 o'clock position without harm. The sound was very lively and
on some materials seemed almost too bright.
I'll have to listen again this weekend once everything is warmed up. (The preamp
is meant to be powered on continuously.)
Minor technical details on the F4: 400VA Antec transformer, 8 x 22000 uF in
CRC power supply. Panasonic caps on the F4 itself, using Peter Daniel PCBs.
The mosfets are Fairchild FQA19N20C and FQA12P20.
Eventually a new preamp will be needed since this one does not have adequate
voltage swing for the F4 (max 10V RMS).
Cheers,
Dennis
failed attempts at bias
The cool weather made me dust off the project, hoping it would work as a space heater
I'm having a hard time remembering my mistakes.
Originally I couldn't get the first channel to heat up with a 10K resistor at R9. I don't remember if I tried to measure across one of the source resistors.
*** it is entirely possible I was dim enough to have expected the amp to get warm with a dim bulb tester in place***
I might have done something different to the other channel last year, possibly installed a 4.7K resistor at R9 without first turning P1 back to its lowest setting. I walked away for about a minute, when I returned the entire heatsink had heated up as faster than if I had put a large propane torch on it. I don't see how the 3.x amp fuse didn't blow. I put it away for almost a year.
Last week I wasn't able to get the first cold channel to heat up after putting another resistor in parallel to get it to about 6K at R9, but it I did measure 130mv across a source resistor. No heat, other than what may have been caused by touching it with my hands. It climbed from 70 to 72 F over half an hour.
With the second channel, I was not able to get half way to 130mv. Somewhere around the 60mv I see mentioned in other posts. I do not remember taking heat measurements.
I found some cheap 5.6K resistors locally, and installed them in both sides. Applied power (without a dim bulb tester), the LED lights on the boards and PSU went out instantly as I was greeted by a puff of white smoke. The LED's normally take about 45 seconds to go dark with the PSU drain resistor value I chose.
I let a couple of days go by, then replaced the blown 3.x amp fuse in the power entry module - an oddball sized / shaped surplus EMI / RFI filtering module I had picked up a pair of for the project. Still, no lights at all on powering up, so am guessing the puff of smoke was the internals of the PEM.
Currently I'm waiting on a pair of rear panels cut for a standard sized IEC inlet to arrive from Italy. Recommendations on fuse size / type for the IEC module?
An R9 solder pad has come off the board and if were as simple as knowing the correct value I would consider removing the entire board to solder it from the back side but I'll probably just fill the hole with a strand of wire and a lot of solder.
What is different from standard on my BOM?
First, an Antek AS-4220 - 400VA 20V transformer. The 300VA 18V were not available when I started building this a couple of years ago.
I'm using all DIYAudio store boards, both F4 and the V3 PSU with 60K per rail. The drain resistor value chosen wasn't the greatest choice as it takes a minute for the caps to drain and the LED's dim off.
I chose 50V capacitors for the F4 boards. Wiring is a larger diameter than required, using 16 gauge.
I placed an order with a large selection of guesstimated R9 values, ranging from 4.5K to 6.5K. At the time I had not read about the instances where R8 had to be replaced on some people's builds.
The cool weather made me dust off the project, hoping it would work as a space heater
I'm having a hard time remembering my mistakes.
Originally I couldn't get the first channel to heat up with a 10K resistor at R9. I don't remember if I tried to measure across one of the source resistors.
*** it is entirely possible I was dim enough to have expected the amp to get warm with a dim bulb tester in place***
I might have done something different to the other channel last year, possibly installed a 4.7K resistor at R9 without first turning P1 back to its lowest setting. I walked away for about a minute, when I returned the entire heatsink had heated up as faster than if I had put a large propane torch on it. I don't see how the 3.x amp fuse didn't blow. I put it away for almost a year.
Last week I wasn't able to get the first cold channel to heat up after putting another resistor in parallel to get it to about 6K at R9, but it I did measure 130mv across a source resistor. No heat, other than what may have been caused by touching it with my hands. It climbed from 70 to 72 F over half an hour.
With the second channel, I was not able to get half way to 130mv. Somewhere around the 60mv I see mentioned in other posts. I do not remember taking heat measurements.
I found some cheap 5.6K resistors locally, and installed them in both sides. Applied power (without a dim bulb tester), the LED lights on the boards and PSU went out instantly as I was greeted by a puff of white smoke. The LED's normally take about 45 seconds to go dark with the PSU drain resistor value I chose.
I let a couple of days go by, then replaced the blown 3.x amp fuse in the power entry module - an oddball sized / shaped surplus EMI / RFI filtering module I had picked up a pair of for the project. Still, no lights at all on powering up, so am guessing the puff of smoke was the internals of the PEM.
Currently I'm waiting on a pair of rear panels cut for a standard sized IEC inlet to arrive from Italy. Recommendations on fuse size / type for the IEC module?
An R9 solder pad has come off the board and if were as simple as knowing the correct value I would consider removing the entire board to solder it from the back side but I'll probably just fill the hole with a strand of wire and a lot of solder.
What is different from standard on my BOM?
First, an Antek AS-4220 - 400VA 20V transformer. The 300VA 18V were not available when I started building this a couple of years ago.
I'm using all DIYAudio store boards, both F4 and the V3 PSU with 60K per rail. The drain resistor value chosen wasn't the greatest choice as it takes a minute for the caps to drain and the LED's dim off.
I chose 50V capacitors for the F4 boards. Wiring is a larger diameter than required, using 16 gauge.
I placed an order with a large selection of guesstimated R9 values, ranging from 4.5K to 6.5K. At the time I had not read about the instances where R8 had to be replaced on some people's builds.
Other variations on my build are:
Q1 2SK170BL
Q2 2SJ74BL
Q11 TL431BQLPR
The important part is probably R8 22.1K
Earlier today Zen Mod answered the same question for another member. I'm posting it here so if someone runs across this in a search or is following this thread.
Based on the reply in the F4 power amplifier thread #4273 R8 27K is the next thing to try (?).
"which exact schematic you're using ?" was probably the best question possible. Zen is not going to like my answer I wasn't using one, but my version of despised public BOM. My comprehension of the circuit is embarrassing. For reference I will say the schematic on the first page of this thread which includes R8 at 22.1K
DIYAudio F4 Schematic
For reference, fourth page from the bottom of the First Watt Operation and Service Manual has a schematic. There is where I see R8 27.4K
I'll try to find if there is a short between the MOSFET's.
My sloppy confined space rework on the R9 resistor was likely to have had bad solder joints.
Back to my ignorance of the circuit, as I've been forcing myself to draw by hand and understand as I draw I'm stumped by the first line I'm drawing. Not knowing what "T" abbreviations are for on the schematic..., make me guess and I will throw out Terminal, termination point? Then I get to LEDH1 with a "1" and "2" in numbering surrounding the LED1. I can't guess as to what LEDH1 is. For starting my drawing from the top, that's not a lot of ink before I don't understand something.
I spent an hour yesterday trying to write a response thanking Zen Mod for his service to the community... but in the interest of keeping things on topic, the short version is - thank you sir! Same goes for everyone else who is doing their part.
Q1 2SK170BL
Q2 2SJ74BL
Q11 TL431BQLPR
The important part is probably R8 22.1K
Earlier today Zen Mod answered the same question for another member. I'm posting it here so if someone runs across this in a search or is following this thread.
Based on the reply in the F4 power amplifier thread #4273 R8 27K is the next thing to try (?).
"which exact schematic you're using ?" was probably the best question possible. Zen is not going to like my answer
I wasn't using one, but my version of despised public BOM. My comprehension of the circuit is embarrassing. For reference I will say the schematic on the first page of this thread which includes R8 at 22.1KDIYAudio F4 Schematic
For reference, fourth page from the bottom of the First Watt Operation and Service Manual has a schematic. There is where I see R8 27.4K
I'll try to find if there is a short between the MOSFET's.
My sloppy confined space rework on the R9 resistor was likely to have had bad solder joints.
Back to my ignorance of the circuit, as I've been forcing myself to draw by hand and understand as I draw I'm stumped by the first line I'm drawing. Not knowing what "T" abbreviations are for on the schematic..., make me guess and I will throw out Terminal, termination point? Then I get to LEDH1 with a "1" and "2" in numbering surrounding the LED1. I can't guess as to what LEDH1 is. For starting my drawing from the top, that's not a lot of ink before I don't understand something.
I spent an hour yesterday trying to write a response thanking Zen Mod for his service to the community... but in the interest of keeping things on topic, the short version is - thank you sir! Same goes for everyone else who is doing their part.
Last edited:
Hi Jeffs- Since you've been working on the amp for some time, and made some resistor substitutions, it may not be a bad idea to do a spreadsheet of all your resistor values ( for the purpose of comparing to the F4 BOM/schematic ). Measuring all of them will let you rule them out.
I'll follow that by saying I'm no pro at diagnosing these things and please put everyone else's advice ahead of mine. Just seems like resistor inventory could help.
I'll follow that by saying I'm no pro at diagnosing these things and please put everyone else's advice ahead of mine
. Just seems like resistor inventory could help.
That was the idea - view the schematic, check against various purchase orders, draw the circuit, fill in values until I began to understand how things were working.
My resistor inventory has been growing, but another $8 in shipping for Mouser will be in order if I end up changing the value of R8.
If you want to export and send, it would be another way to look at it.
I see you are in Austin, it would be nice to actually listen to one of these if you have time between the holidays and other obligations.
My resistor inventory has been growing, but another $8 in shipping for Mouser will be in order if I end up changing the value of R8.
If you want to export and send, it would be another way to look at it.
I see you are in Austin, it would be nice to actually listen to one of these if you have time between the holidays and other obligations.
It looks like the store BOM is based on a slightly older schematics dated 4/17/07 (see attached.)
The differences between that and the current 6/4/07 one seem to be
1) lower degeneration resistor at R3, R4 (from 22ohm to 10 ohm)
2) gate resistors went from 100 ohm to 150 ohm
3) R8 from 22.1K to 27.4K.
Cheers,
Dennis
The differences between that and the current 6/4/07 one seem to be
1) lower degeneration resistor at R3, R4 (from 22ohm to 10 ohm)
2) gate resistors went from 100 ohm to 150 ohm
3) R8 from 22.1K to 27.4K.
Cheers,
Dennis
