> This amp is basically ready to go in a box - sounds excellent.
For such a low-cost scratch build, I am pleased to hear that you are still happy with the results. 🙂
Patrick
For such a low-cost scratch build, I am pleased to hear that you are still happy with the results. 🙂
Patrick
Due to the simplicity of the design - P2P wiring is perfectly suitable. The sound quality from about $25 in parts is quite astounding. Well, literally it will blow many much more expensive and complex head amps out of the water. It is better than my JLH head amp. The use of JFET inputs and low impedance MOSFET outputs gives this amp an immediacy and quickness that is not there in the JLH. The low noise floor is also unmatched as it is not audible when inputs are shorted or when no music playing. I think Ineould like to have a DPST switch to enable/disable the cross feed filter.
This may be a more suitable case:
http://m.aliexpress.com/item/556829834.html
Last edited:
This is the same 2107 case that everyone has been using.
You can get them without all the knobs and handles.
Here is how I made use of a standard case; no custom XEN mechanical parts.
http://www.diyaudio.com/forums/head...portable-headphone-amplifier.html#post4758783
Patrick
You can get them without all the knobs and handles.
Here is how I made use of a standard case; no custom XEN mechanical parts.
http://www.diyaudio.com/forums/head...portable-headphone-amplifier.html#post4758783
Patrick
Last edited:
The minimum I would recommend you to change are all the ceramic caps in the XFD to MKP or FKP.
The next thing I would consider are Fairchild MOSFETs.
Patrick
The next thing I would consider are Fairchild MOSFETs.
Patrick
Do you cut/drill your own holes for jacks and knobs? The price on this unit with cutouts and knobs is actually less than some other "blank" ones.
The IEC inlet box is the biggest pain to cut if you don't have a milling machine.
I will probably mount the little amp boards directly to the heatsink side walls rather than use a heat spreader.
Will have to drill a single hole in front for headphone jack though if I get this particular unit.
I plan on film caps as soon as they arrive. Had to use ceramics just to test if it makes sound last night and to attenuate volume as way too loud on my headphones and also lets me turn volume down to connect/disconnect inputs.
Film caps for XF filter on order. Upon more careful reading of the cross feed writeup, I see that the point of the 12k resistor was to bring the 10k pot to a 5.5k overall impedance to be an ideal match for the Linkwitz crossover filter. Since I am using a 50k pot, a 6.2k resistor is what I need. Currently running a 47k there so the XF filter is probably not working as designed - still sounds quite good. So will correct this as well.
I do notice that the length of thin signal wire between the XF board and the main amps picks up noise if I get my fingers near it. Looks like it may be that the 100k input impedance of the amp is too high and results in easy stray noise pickup. But that impedance allows the 5.5k XF filter impedance to work without a buffer I think.
Maybe bring down impedance resistor on F5 HA down to say 47k or 22k to reduce noise pickup?
I do notice that the length of thin signal wire between the XF board and the main amps picks up noise if I get my fingers near it. Looks like it may be that the 100k input impedance of the amp is too high and results in easy stray noise pickup. But that impedance allows the 5.5k XF filter impedance to work without a buffer I think.
Maybe bring down impedance resistor on F5 HA down to say 47k or 22k to reduce noise pickup?
Hi X,
Which version of the linkwitz crossover filter are you referring to? The one on page 9 of the F5-HA Description V1.4 or from https://tangentsoft.net/audio/mlxfeed.html ?
Which version of the linkwitz crossover filter are you referring to? The one on page 9 of the F5-HA Description V1.4 or from https://tangentsoft.net/audio/mlxfeed.html ?
I am using the one Patrick is using:
http://www.diyaudio.com/forums/atta...f5-headamp-f5-ha-using-passive-cross-feed.pdf
Patrick.
I searched but could not find the schematic for the F5X HA. Is that available as a schematic or veroboard layout, or only as a PCB from xenaudio?
Thanks,
X
I searched but could not find the schematic for the F5X HA. Is that available as a schematic or veroboard layout, or only as a PCB from xenaudio?
Thanks,
X
WOW!! I wished I hadn't thrown mine in the trash now.
Hmmm...I'm noticing a pattern with your reviews.
Seems the newest design you build always seems to sound the best until the next one comes along.
Not sure what you and others find so great about the sound of MOSFET output stages.
Quickness and immediacy are certainly not terms I would use to describe their sound.
Those terms are far more applicable to BJT output stages than they are to the slooooow and boring sound of MOSFETs.
The fact that two or more people can hear the same design and describe its sound differently is what makes this hobby so great.
> I searched but could not find the schematic for the F5X HA.
> Is that available as a schematic or veroboard layout, or only as a PCB from xenaudio?
Not yet ready for publication.
But if you compare the schematic of the F5X and the F5-HA, you can easily work it out yourself.
Patrick
> Is that available as a schematic or veroboard layout, or only as a PCB from xenaudio?
Not yet ready for publication.
But if you compare the schematic of the F5X and the F5-HA, you can easily work it out yourself.
Patrick
My last amp, the Pass M2 did not get favorable accolades from me for SQ. So no, just because I make an amp doesn't always mean it gets a great sound quality review. 🙂
Your F5 HA may have had some damaged parts? This amp really does sound very good and that's not just hyperbole of the most recent amp syndrome.
My experience with MOSFETs vs BJTs is that MOSFETs have better sound quality from bass standpoint and about the same for mids and highs.
Test case - a paper box to check fit
I received my 50VA Antek toroidal transformer yesterday. I installed it and the rest of the amp plus all the ancillary stuff needed for a real standalone headamp. RCA jacks, headphone jack, volume knob, EIC wall plug inlet with fuse/switch.
I put it all in a small cardboard box to check for fit to see the minimum size box I will need. My PSU board is rather large so not sure if the usual 2107 aluminum cases are going to work here. Although in the 2107, the actual amp boards will be screwed to the side walls where the heat sinks are so that frees up some space.
I also fixed the output impedance of the cross feed module to get 5.5kohm, and I added film bypass caps to bring the 20nF up to 22.1nF. I also added my traditional practice of a 330pF 1000v ceramic RF suppression cap (it's on all of my amps thus far) at the input to the amp - which, in the case is the input of the cross feed module.
Stuffing it all into the box, it all barely fits and I try to keep the AC line stuff far away from the sensitive amp modules which have a 100k input impedance. I also tied the metal shell of of the volume pot to ground with a 10R resistor to prevent hum from touching volume knob. Amp ground is connected to earth ground via star at center of PSU cap bank.
Well, the short story is even in a cardboard box which affords no grounding or RF suppression, the amp still is dead silent when no signal is present. No noise pickup issues with the lines between the cross feed module and amp. I absolutely cannot tell it is even on if not for the LED glowing. The sound is fantastically clear and free of any audible distortion. Test tracks with bass that I am familiar with all play well with excellent reach and articulation. No lack of bass in this headamp at all.''
The clarity and dynamics are stunning. I think the correct cross feed resistor and now with film bypass caps (3x 680pF 100v per side + 20nF 50v ceramic), the soundstage and imaging are excellent. I have much more confidence now that this will all work well when stuffed into an aluminum box.
I received my 50VA Antek toroidal transformer yesterday. I installed it and the rest of the amp plus all the ancillary stuff needed for a real standalone headamp. RCA jacks, headphone jack, volume knob, EIC wall plug inlet with fuse/switch.
I put it all in a small cardboard box to check for fit to see the minimum size box I will need. My PSU board is rather large so not sure if the usual 2107 aluminum cases are going to work here. Although in the 2107, the actual amp boards will be screwed to the side walls where the heat sinks are so that frees up some space.
I also fixed the output impedance of the cross feed module to get 5.5kohm, and I added film bypass caps to bring the 20nF up to 22.1nF. I also added my traditional practice of a 330pF 1000v ceramic RF suppression cap (it's on all of my amps thus far) at the input to the amp - which, in the case is the input of the cross feed module.
Stuffing it all into the box, it all barely fits and I try to keep the AC line stuff far away from the sensitive amp modules which have a 100k input impedance. I also tied the metal shell of of the volume pot to ground with a 10R resistor to prevent hum from touching volume knob. Amp ground is connected to earth ground via star at center of PSU cap bank.
Well, the short story is even in a cardboard box which affords no grounding or RF suppression, the amp still is dead silent when no signal is present. No noise pickup issues with the lines between the cross feed module and amp. I absolutely cannot tell it is even on if not for the LED glowing. The sound is fantastically clear and free of any audible distortion. Test tracks with bass that I am familiar with all play well with excellent reach and articulation. No lack of bass in this headamp at all.''
The clarity and dynamics are stunning. I think the correct cross feed resistor and now with film bypass caps (3x 680pF 100v per side + 20nF 50v ceramic), the soundstage and imaging are excellent. I have much more confidence now that this will all work well when stuffed into an aluminum box.
Attachments
Last edited:
I have spent another week with the F5 headamp after fitting the cap multiplier.
I do not believe there is anything wrong with this amp although I have not put it on anything other than a PC based scope. A frequency sweep by ear sounds normal. Let's assume the amp is good to go.
My thoughts have changed a bit...probably due to a few things.
I believe this amp this broken in...a lot. Probably due to the Silmics in the power supply. It has been hot for 100+ hours now and I've probably listened for 30 of those hours. It has changed. The cap multiplier did nothing to background noise but the amp seems a little (not a lot) more round and weighty. A positive change.
Here is thing, this amp is fairly sensitive to the headphones being used. I use AKG 272 at work, a closed back headphone with a light, airy sound. No bass at all. The F5 hates these headphones. Sounds bleached and dry. I much prefer the Hifiman EF2 amp with these headphones.
Switch to a pair of Grado 125 and the F5 is suddenly transformed.
Compared to the hifiman amp the F5 is fast and composed. As the volume goes up things become more apparent. The F5 is smooth and clean..the hifiman amp is becoming compressed sounding. Bass and mids are becoming blurred on the hifiman amp. The f5 is holding locations of instruments. Vocals are more present and natural on the F5, the hifiman amp is driving the vocals downward. And the Grados are supposed to be bright and closed-in. Seems like most would prefer the tube hifiman amp but The F5 is clearly superior.
Unfortunately, I built this amp to use at work with the AKGs. It is simply not a good combo.
I am not an experienced headphone person. I have a good home hifi setup (Rega P5, Pass f5t, KEF R700) and I understand sound...I am just not that enthuisiastic about headphones.
I now realize the hifiman amp is colored and the F5 headamp is not. The F5 has less apparent bass than the hifiman amp. I have confirmed this by ear with test tones. Not deeper, but stronger.
But paired with a headphone that is more colored (Grado) it became apparent that the F5 is superior. Paired with a flat, sterile sounding headphone (AKG 272) and the f5 was boring and sounded dry and washed out. It seems to be all about the pairing...
Ultimately (and keep and mind I don't have a lot of experience with headphones) I think it's a good amp. I would love to compare to more amps and apologize that my comparisons are so limited.
Headphones vary so much in sound signature (much more than speakers I think) that I can see why some people love this amp and some people seem not to.
But if you like fast and clean with a big soundstage I don't see why you wouldnt like this amp. If you want lush and dark it's probably not for you.
Now all I need is a new set of closed back phones...how would the F5 do into say a set of 250 or 600 ohm beyers?
I received my 50VA Antek toroidal transformer yesterday. I installed it and the rest of the amp plus all the ancillary stuff needed for a real standalone headamp. RCA jacks, headphone jack, volume knob, EIC wall plug inlet with fuse/switch.
I put it all in a small cardboard box to check for fit to see the minimum size box I will need. My PSU board is rather large so not sure if the usual 2107 aluminum cases are going to work here. Although in the 2107, the actual amp boards will be screwed to the side walls where the heat sinks are so that frees up some space.
I also fixed the output impedance of the cross feed module to get 5.5kohm, and I added film bypass caps to bring the 20nF up to 22.1nF. I also added my traditional practice of a 330pF 1000v ceramic RF suppression cap (it's on all of my amps thus far) at the input to the amp - which, in the case is the input of the cross feed module.
Stuffing it all into the box, it all barely fits and I try to keep the AC line stuff far away from the sensitive amp modules which have a 100k input impedance. I also tied the metal shell of of the volume pot to ground with a 10R resistor to prevent hum from touching volume knob. Amp ground is connected to earth ground via star at center of PSU cap bank.
Well, the short story is even in a cardboard box which affords no grounding or RF suppression, the amp still is dead silent when no signal is present. No noise pickup issues with the lines between the cross feed module and amp. I absolutely cannot tell it is even on if not for the LED glowing. The sound is fantastically clear and free of any audible distortion. Test tracks with bass that I am familiar with all play well with excellent reach and articulation. No lack of bass in this headamp at all.''
The clarity and dynamics are stunning. I think the correct cross feed resistor and now with film bypass caps (3x 680pF 100v per side + 20nF 50v ceramic), the soundstage and imaging are excellent. I have much more confidence now that this will all work well when stuffed into an aluminum box.
That's probably not gonna fit in a 2107 chassis. The 2107 is almost exactly 4cm tall internally on the heatsink side which means no downward facing mosfets on the 4x6 boards. I had to mount the mosfets side to side to mount the 4x6 boards on the heatsinks. And you have much more PSU size than I do . Better go with the 2109 chassis. It's taller. Here's a pic of mine before the cap multipiers in a 2107 case.
Attachments
Hikari1: The 2107 is supposed to be 62mm high on the heatsink side..? (4mm top and bottom plates) - And im quite certain my 2109 have 82mm tall sinks..
> Paired with a flat, sterile sounding headphone (AKG 272) and the f5 was boring and sounded dry and washed out.
It is your subjective impression, to which you are fully entitled.
But among the group of people who have built and listened to the amp, all being headphone enthusiants except myself,
we have tested with a great variety of phones, including
Senn HD250 Linear, HD600, HD650, HD800 & 800S, AKG K501, K701, Hifiman HE500, Fostex THX00, TH-900, ......
Their impressions have already been posted here before and I shall not repeat.
But certainly not "boring, dry and washed out".
True that we have not tested with the K272, but one can see some measurements of its frequency response here :
http://www.innerfidelity.com/images/AKGK272HD.pdf
and then the comparison to the Grado 125 :
http://www.innerfidelity.com/images/GradoSR125i.pdf
> If you want lush and dark it's probably not for you.
This I agree.
But anyone who knows the F5 should know that already.
For lush and dark, you should not have an amp with such high bandwidth.
Or should use MOSFETs with high tranconductance and high capacitances instead.
> how would the F5 do into say a set of 250 or 600 ohm beyers?
It is much happier with 32 or 50 ohm, like in most modern planar phones, though users of the HD600, 650, 800 have no complaints.
Patrick
It is your subjective impression, to which you are fully entitled.
But among the group of people who have built and listened to the amp, all being headphone enthusiants except myself,
we have tested with a great variety of phones, including
Senn HD250 Linear, HD600, HD650, HD800 & 800S, AKG K501, K701, Hifiman HE500, Fostex THX00, TH-900, ......
Their impressions have already been posted here before and I shall not repeat.
But certainly not "boring, dry and washed out".
True that we have not tested with the K272, but one can see some measurements of its frequency response here :
http://www.innerfidelity.com/images/AKGK272HD.pdf
and then the comparison to the Grado 125 :
http://www.innerfidelity.com/images/GradoSR125i.pdf
> If you want lush and dark it's probably not for you.
This I agree.
But anyone who knows the F5 should know that already.
For lush and dark, you should not have an amp with such high bandwidth.
Or should use MOSFETs with high tranconductance and high capacitances instead.
> how would the F5 do into say a set of 250 or 600 ohm beyers?
It is much happier with 32 or 50 ohm, like in most modern planar phones, though users of the HD600, 650, 800 have no complaints.
Patrick
Last edited: