Pass F5 Turbo V2; one more build.
- Thread starter surfatwork
- Start date
surfatwork
Well-Known Member
So I finally got my f5t to stop humming and buzzing. It was down to cable routing inside the amp. A bit puzzled though:
1. If I move the analog-in(s) even a little bit, say 5mm, there is a big increase in hum pickup. I am using shielded twin-core coax. Shield to ground. Both cores to signal.
2. If I put the cover on, there is pronounced hum pickup.
3. If the v+/0/v- wires are moved a bit, hum increases.
All of above only when I connect a common ground input. Else it is silent.
The psu is 800va 0/24 x 2, feeding Prasi's crc psu with on board diodes. One +/0/- feeding both channels. Ground is via a bridge isolator+ntc.
The back is board (temporarily) since I was experimenting with various positions of the RCA and speaker posts.
Any ideas why it is so sensitive?
Thanks
1. If I move the analog-in(s) even a little bit, say 5mm, there is a big increase in hum pickup. I am using shielded twin-core coax. Shield to ground. Both cores to signal.
2. If I put the cover on, there is pronounced hum pickup.
3. If the v+/0/v- wires are moved a bit, hum increases.
All of above only when I connect a common ground input. Else it is silent.
The psu is 800va 0/24 x 2, feeding Prasi's crc psu with on board diodes. One +/0/- feeding both channels. Ground is via a bridge isolator+ntc.
The back is board (temporarily) since I was experimenting with various positions of the RCA and speaker posts.
Any ideas why it is so sensitive?
Thanks
This configuration creates a loop between the core conductors from joint to joint and easily catch up EMI. Use only single core shielded cables in signal input sections. Also use small neo magnets to ensure the cable conductors are absolutely non-magnetic.
surfatwork
Well-Known Member
Tried single core microphone cable. Didn't improve matters.
Great work Surfatwork. If the hum gets more pronounced if the top plate is screwed on, it's acting like an antenna to attract EMO I guess.
surfatwork
Well-Known Member
Decided to revive this build. Ordered a set of F5t boards, and got some JFETs. Some questions:
1. For a standard f5t v2 with one pair of input JFETs (per channel) with no cascoding, what matching needs to be done?
2. Do the output mosfets have to be matched as well? Is that NN/PP matching of each output pair, or is NNPP matching required?
1. For a standard f5t v2 with one pair of input JFETs (per channel) with no cascoding, what matching needs to be done?
2. Do the output mosfets have to be matched as well? Is that NN/PP matching of each output pair, or is NNPP matching required?
I had build the Aleph J earlier, so based on that the F5T being another Nelson amp giving a wild guess which I am sure should be true. Yes as per the long build thread on the diyaudio forum the usual JFETs (if you have the Toshiba genuine ones) then they have to be matched very close to around 5mv I think. If not Toshiba then you can get those Linear ones which also need to be matched for good stereo separation.
Regarding the output mosfets I presume you are using the IRFP240/IRFP9240 then yes they should be closely matched as NN and PP. It is extremely difficult to get closely matched with NP and as per Nelson getting it matched as NN and PP is more than sufficient. PM me and I can give a source where I usually get my mosfets and JFETs genuine ones which are closely matched. I have brought from them for my Aleph J and now Sony VFET2 amps.
Thanks
surfatwork
Well-Known Member
thanks, Manniraj. Have PM'ed you
Could people also suggest some reliable, reasonably priced suppliers for quality resistors and small value (not PSU) caps?
Could people also suggest some reliable, reasonably priced suppliers for quality resistors and small value (not PSU) caps?
surfatwork
Well-Known Member
Delighted to report that I finally got rid of the hum in my F5T!
There seemed to have been several factors.
1. The old boards from the GB here had tarnished badly. Coupled with that, the holes were tiny, which resulted in bad solders in many places. I replaced them with new pcbs from diyaudio and rebuilt the amp reusing the same components.
2. The transformer orientation was causing issues. Rotating it helped.
3. Redid the wiring layout
4. The psu had issues, but I had sorted that earlier. Now have a rectifier fed by one pair of secondaries for each rail, with +/- shorted to generate the 0V rail. The filter boards are in crc-crc configuration - I cascaded the GB filter boards. Unfortunately, my transformer has only 2 pairs of secondaries, not 4, so cant make it a dual mono. Maybe I'll get the transformer rewound.
And now she sings. Bit more of tidying up to do, so pics tomorrow. Thanks for all the suggestions.
There seemed to have been several factors.
1. The old boards from the GB here had tarnished badly. Coupled with that, the holes were tiny, which resulted in bad solders in many places. I replaced them with new pcbs from diyaudio and rebuilt the amp reusing the same components.
2. The transformer orientation was causing issues. Rotating it helped.
3. Redid the wiring layout
4. The psu had issues, but I had sorted that earlier. Now have a rectifier fed by one pair of secondaries for each rail, with +/- shorted to generate the 0V rail. The filter boards are in crc-crc configuration - I cascaded the GB filter boards. Unfortunately, my transformer has only 2 pairs of secondaries, not 4, so cant make it a dual mono. Maybe I'll get the transformer rewound.
And now she sings. Bit more of tidying up to do, so pics tomorrow. Thanks for all the suggestions.
surfatwork
Well-Known Member
Pics. could be better finished, I know. Didnt have the energy and time to do the chassis properly. That's for next phase, now that it's working properly.
surfatwork
Well-Known Member
So today morning I hooked it Up to my main speakers (mission m53, nothing fancy). But in my room, they sound terrific paired with my peeceebee.
Long story short, I was very underwhelmed with the f5t. Disappointed, almost. Tried several songs - they sounded cluttered, poor bass. Treble was ok. But not what I was expecting after all the rave F5T reviews.
Went out for lunch in a foul mood.
Came back couple of hours later. Realized then I had not switched off the amp. Was about to, and then thought I'd listen to one more song. And my God, it had changed. Completely. Amazing Soundstage, the bass is superb, voice so sweet.
This baby needs to run HOT!!! Been listening to song after song. The same ones I had played in the morning now played like something else.
So warm up your amps, friends. It's worth the few extra bucks for electricity.
Long story short, I was very underwhelmed with the f5t. Disappointed, almost. Tried several songs - they sounded cluttered, poor bass. Treble was ok. But not what I was expecting after all the rave F5T reviews.
Went out for lunch in a foul mood.
Came back couple of hours later. Realized then I had not switched off the amp. Was about to, and then thought I'd listen to one more song. And my God, it had changed. Completely. Amazing Soundstage, the bass is superb, voice so sweet.
This baby needs to run HOT!!! Been listening to song after song. The same ones I had played in the morning now played like something else.
So warm up your amps, friends. It's worth the few extra bucks for electricity.
To Enjoy the Sound signature of Class A it needs to get Hot. Further i would suggest you to check the initial bias current when the Amp is Powered ON. In my Hiraga 30 watt Amp the initial Bias curent is 1.4 Amp and after switching on within 5 to 6 Mins it reaches to 1.6 Amps. The Sound signature of my amp changes within 5 Mins. It has an Amazing Sound.
Sadik
surfatwork
Well-Known Member
Haven't biased it to final level yet. Currently at 1.3A, should be able to take it a bit higher. Which should speed things up a bit.
I know that class A needs to warm up. What took me aback was the complete change in character - never seen such a dramatic change before.
And with the hum gone, the background is black. The sound just starts in a very unearthly fashion. Having said that, the peeceebee is almost as good. Voice is probably a bit better in the peeceebee.
Now keen to make it a Dual mono with independent supplies.
I know that class A needs to warm up. What took me aback was the complete change in character - never seen such a dramatic change before.
And with the hum gone, the background is black. The sound just starts in a very unearthly fashion. Having said that, the peeceebee is almost as good. Voice is probably a bit better in the peeceebee.
Now keen to make it a Dual mono with independent supplies.
@surfatwork : now that you've proven the build, take the bias up to the recommended amperage. Any mildly biased amp, be it A or AB, takes longer to reach operating temperature, sometimes as much as 1 hour.
And congrats on completing the build!
And congrats on completing the build!
This is interesting, the PCB's from Diyaudiostore and build guide from 6L6 mention that they need to be mounted on the same heatsinks as MOSFET's...not sure if I understood this correctly
surfatwork
Well-Known Member
To be honest, I am also not quite sure about this. Logically speaking what omishra says is correct. And there was discussion about this in the early stages on diyaudio. But that thought seems to have been abandoned at some point and most builds have the diodes and mosfets on the same sink. The diyaudio pcbs are also designed for that. I too have them like that and haven't had any issues so far. I am hoping that if there is runaway, the NTCs will help before catastrophe occurs.
Check out our special offers on Stereo Package & Bundles for all budget types.
