Last night i had time to fiddle around with the amp some more. I reset the transformer to its original position flush with the base plate of the amp and instead switched the monolithic diode bridges for the breakaway boards from the diyaudiostore universal PSU with 250V/40A schottkys. Fast forward to the result - no change in the FFT nor noise. My current impression now is that we are back to a case of bigger is better - more distance between the power and input transformers is probably the way to go. Unfortunately, the leads of the PT are too short for me to put it anywhere else at the moment, and the chassis is too small to offer many more options, so i cannot really test it. I have bought some mu-metal samples from a german company and will try to make some cans for the Jensens with that and see if it makes a difference. So a lesson from that would be, if going with the 4U store chassis, its probably worth getting the deeper version for this amp if absolute minimal noise is the goal.
Bear in mind though, that this noise is still so low that i have to put my ear close to the speaker to hear it - no big deal really.
I will keep this in mind for the next build i plan to try (M2) and go for a bigger chassis and possibly a shielded PT.
Bear in mind though, that this noise is still so low that i have to put my ear close to the speaker to hear it - no big deal really.
I will keep this in mind for the next build i plan to try (M2) and go for a bigger chassis and possibly a shielded PT.
IMO 'something' ( basic ?) Is amiss in your build.
My F6.. built exactly as per 6L6's guide, being too much of an amateur to adlib Anything.. is DEAD silent.
Vol pot to max.. ears pressed as close as I can get them to the HF compression dome(s) of my Tannoy Golds ...Zero noises.. period.
Same as when the amp is unplugged .. no difference.
Astounded me.. initially, now I simply take it for granted
Revisit your assembly perhaps ?
There is a Small possibility that your mains power my be extremely noisy.
Once had a Video monitor (cad station) that would flicker badly But only in one specific building.
Subsequent Electrical utility testings found that for reasons they wouldn't admit to.. the power being supplied was full of extraneous noises.
My F6.. built exactly as per 6L6's guide, being too much of an amateur to adlib Anything.. is DEAD silent.
Vol pot to max.. ears pressed as close as I can get them to the HF compression dome(s) of my Tannoy Golds ...Zero noises.. period.
Same as when the amp is unplugged .. no difference.
Astounded me.. initially, now I simply take it for granted
Revisit your assembly perhaps ?
There is a Small possibility that your mains power my be extremely noisy.
Once had a Video monitor (cad station) that would flicker badly But only in one specific building.
Subsequent Electrical utility testings found that for reasons they wouldn't admit to.. the power being supplied was full of extraneous noises.
Last edited:
Well, as I said above, I’m leaning towards radiated noise from power transformer into input transformers and on the way to testing it. The build is faithful to the guide as well, within stated specs. Just because you follow the guide doesn’t mean everything will be exactly the same every time though, especially considering different component brands/series etc will not have identical performance.
Shielded pt was recommended specifically by 6L6 in the original F6 thread for the smaller deluxe chassis but I only just found that mention recently.
More filtering may help, the only way to know is to try, but I’m testing the shielding idea first.
Shielded pt was recommended specifically by 6L6 in the original F6 thread for the smaller deluxe chassis but I only just found that mention recently.
More filtering may help, the only way to know is to try, but I’m testing the shielding idea first.
Last edited:
So, what is more effective, CRCRC or CRC, assuming same total capacitance/resistance? I can get 100mF/40V caps that are actually reasonably priced here (somewhat at least), so was wondering if 100mF-0.12R-100mF for each side might be worth doing (over stock, 30mF-0.12R-30mF)? Or would it be better to run 30mF-0.12-30mF-0.12-30mF or something similar?
Bonus question, what determines the size of the R? i tried putting 0.12 vs 0.24R into PSUD2 and got much lower ripple with the higher R. do we simply go for the preferred tradeoff between voltage drop (lower R better) and smoothing (higher R better), or is that too simple? Can i compensate for higher R by using a transformer with higher secondary voltage?
more effective is by far CLC , more capacitance for reservoir ( the last caps ) and the shorter wire as possible , and one single point for all 0V ( per channel assumig your going dual mono ), the lowest esr for the second row of caps .
for example my F6 have : toroid trafo 2 x 220 va , then 16 x STPS30SM120ST, then 4 x 33000 uf then 4 x inductors 10 mH/0.47 ohm then 4 x 69000 uf
the all thing is dead silent on 15 Tannoys
because my trafos and inductors are near the boards i did that with 0.1 mm mumetal :
it won't harm for sure
.
for example my F6 have : toroid trafo 2 x 220 va , then 16 x STPS30SM120ST, then 4 x 33000 uf then 4 x inductors 10 mH/0.47 ohm then 4 x 69000 uf
the all thing is dead silent on 15 Tannoys
because my trafos and inductors are near the boards i did that with 0.1 mm mumetal :
it won't harm for sure
.
Last edited:
Haha that looks like a serious amp. Are those tubes for cooling fluid??
I ordered some mu-metal sheets so i will try my hand at making boxes for the transformers. How did you make those neat boxes? Cut and fold - are they welded shut (is that necessary)?
OK follow up: I just ran your numbers into PSUD2 to get a feel for them, and surprisingly 66mF-5H/0.22R-66mF gives me lower ripple than 66mF-5H/0.22R-140mF. Approx 100-fold difference, from 0.15 to 15mV. Any reason this might be?
I agree, the ripple reduction looks quite impressive. Is there any danger of radiation from chokes coupling into input transformers?
I ordered some mu-metal sheets so i will try my hand at making boxes for the transformers. How did you make those neat boxes? Cut and fold - are they welded shut (is that necessary)?
OK follow up: I just ran your numbers into PSUD2 to get a feel for them, and surprisingly 66mF-5H/0.22R-66mF gives me lower ripple than 66mF-5H/0.22R-140mF. Approx 100-fold difference, from 0.15 to 15mV. Any reason this might be?
I agree, the ripple reduction looks quite impressive. Is there any danger of radiation from chokes coupling into input transformers?
Last edited:
Yes the output are liquid cooled
To do the box , you may have the less cut as possible and the more fold , to ensure the magnetic conduction , no way to weld that stuff !!!
and the mumetal box may not be in contact with the input trafos core
psud2 is a great tool , but in real life ( measuring ) you find better result
regarding the chokes coupling into input trafos , i havent measured anything that prove anything , with or without mumetal protection, for me those protection were the cherry on the top of the cake
.
To do the box , you may have the less cut as possible and the more fold , to ensure the magnetic conduction , no way to weld that stuff !!!
and the mumetal box may not be in contact with the input trafos core
psud2 is a great tool , but in real life ( measuring ) you find better result
regarding the chokes coupling into input trafos , i havent measured anything that prove anything , with or without mumetal protection, for me those protection were the cherry on the top of the cake
.
Last edited:
As much as i agree with you that this is a quality amp, i am unfortunately not flush enough to provide it with such a premium supply! However, i find that using 4x33mF (per rail) caps in the DIYAS PSU with 4x1.88R (0.47R) should give below 20mV ripple and keep the voltage above 22V. Practical question: The power dissipation across those four parallel resistors is approx 3.5-4W (assuming 3A draw). Is it still safe to use 3W resistors (since they share the load), or should i go to 5W or higher?
I tried my best to read this thread before asking questions however did not find it on F6's owner's manual too i.e. what is the power consumption when it's on idle (assuming one build's it as per guidelines and with same bias figures as illustrated on the 1st page) ? Secondly how long does one need to keep this amp on before playing music or if there is even any such thing with this amplifier ? I ask these because I wish to build something which I will use on a daily basis and should be able to use in weather conditions here in India without having to worry about AC or room temperature control etc.
Was concerned with all the Warm up talk too.
Welll... I'm lazy and just as often I turn it and enjoy.
Stone cold it still sounds better than ANY amp I've ever owned (been at this audio sillyness since '68)
Sometimes i will turn it on then come back a half hour later to listen.
Truth is the differences (yes there are some.. are Subtle tbh ) as in I find it far more enjoyable to warm it in use rather than waiting for it's best possible sounds
PS; the heatsinks don't seem to get more than barely warm.
Mine is a Poor room heater
Welll... I'm lazy and just as often I turn it and enjoy.
Stone cold it still sounds better than ANY amp I've ever owned (been at this audio sillyness since '68)
Sometimes i will turn it on then come back a half hour later to listen.
Truth is the differences (yes there are some.. are Subtle tbh ) as in I find it far more enjoyable to warm it in use rather than waiting for it's best possible sounds
PS; the heatsinks don't seem to get more than barely warm.
Mine is a Poor room heater
Last edited:
If you dont bias too hot it should be fine (depending on chassis of course). I also rarely let mine warm up, and have not had the feeling anything is lost because of it. It only takes 20-30 mins or so before it will be warm anyways, so that time is quickly spent if the music is good, like Bare said
http://www.firstwatt.com/pdf/prod_f6_man.pdf
States 180W power draw.
http://www.firstwatt.com/pdf/prod_f6_man.pdf
States 180W power draw.