F5 Headamp ?

[EUVL]

The F5X-HA is singing. (Still proper wiring, etc. to finish off.)
Many congratulations to a brave man taking this on as alpha tester.




Patrick

.

 

[cwtim01]

F5X-HA (Balanced Headamp) Prototype

The wiring are still rather messy but the sound is top notch, those who like the F5-HA will love the F5X-HA , if you have a balanced source.

Thanks to Patrick who designed this spectacular amp

This amp is probably not for the first time builder, though.

 

[blackdod]

Very nice.

 

[EUVL]

It only makes sense if you have a balanced source AND a 4-wire headphone.
And definitely not a project for people who want to just solder and switch on, and hope for no smoke.


Patrick

 

[EUVL]

Just for completeness, and by no means new invention, herewith a F5-HA Circlotron.

As the 2SJ74s are becoming real unobtainium, a circlotron can avoid the P-JFET altogether.
The disadvantages are higher Zout and requiring floating supplies.


Patrick

.

 

[ripson]

One set of V2 PCBs

blackdod Spain 2
BMF USA 2
ripson UK 1
hipcheck USA 1
Ryssen Sweden 1
Dennis Hui Canada 1
pchw USA 1
Xen team internal 2
CeeVee Portugal 1
pegasus21 Singapore 1
Fabian85 Germany 1

Running total 14

Plus 1 Set of Case

pegasus21 1 set
bkdog 1 set
BMF 1 set
s610adam 2 sets
Fabian85 1 set

Sorry guys I had to change my order from 2 sets to 1, hope it doesn't change anything.

 

[EUVL]

No problem.
We understand peronal circumtances change with time.


Patrick

 

[Hikari1]

P2P f5 headamp almost complete. Iam having difficulty finding complementary pairs of Fairchild mosfets. Out of 40, I cannot find a n p set with a vgs closer than .4v apart when tested at 19v with 50 ohms resistance.

What are the consequences of less than ideally matched outputs? Higher even order harmonics but will it also effect the stability of DC offset?

 

[EUVL]

Not related to DC offset. Only even harmonics.
I would not lose sleep over that.


P.

 

[Hikari1]

Not related to DC offset. Only even harmonics.
I would not lose sleep over that.


P.

Thanks Patrick. Here is where Iam at progress-wise. Regulators done, building the amplifiers now. Case is 2107. Transformer is shielded antek. The Rean jack puts signal ground to the case. Are all headphone jacks like this? Certainly not desirable with rca connections. Psu ground is separated from case by cl60 but should I worry?

 

[EUVL]

You can get fully inulated jacks. I would use e.g. Neutrik NRJ6HF1AU.
NRJ6HF-1-AU - Neutrik

Also suggest you mount the amp PCB to the front of the heatsink, as much away from the Tx as possible.

Patrick

 

[EUVL]

I have been asked by PM about the difference between Stixx's Cap Muiltiplier (post #232) and the 317_337 based CM that we use.

Essentially not much difference.
The 317_337 is more convenient because it has all the features of a commercial regulator.
And you can easily set the voltage at will.
The Stixx version uses 2x LM329 + 1x LED to set the voltage, so not as flexible.

Both of them are not very tight regulators, and the output voltage will vary with current.
See post #52.
If you need tightly controlled output voltage (for whatever applications), then you need a regulator with voltage feedback.
Examples are the famous Jung_Didden regulator, or LT3042, ...., etc.
And you want to use one with a high bandwidth for digital.


Patrick

 

[xrk971]

Let's mix it up somewhat and save those Toshiba JFETs, without comprimising SQ.
P1/P2 set the Iq of the output stage to about 50-100mA as well as DC offset at the output (start adjusting with maximum resistance of the pots).

Did anyone build this yet? Seems that maybe 100pF between input and ground might be a good idea for RF suppression as this circuit is very fast and may oscillate with RF induced pickup. Also, would a 100nF and 10R to GND on output also be a good idea as well to ensure stability at HF?

Although looking at this circuit and how simple it would be to make on veroboard, had me intrigued. Can it work as drawn? Also, can it work with 2SK170BL and 2SJ74BL for the inputs?

 

[EUVL]

> Can it work as drawn?

I have not built one myself using 2SK246/2SJ103, if that was the question.
And of course you need to trim R3, R4 for bias and DC offet.

> Also, can it work with 2SK170BL and 2SJ74BL for the inputs?

Please read the articles.


Patrick

 

[Hikari1]

[/ATTACH]

You can get fully inulated jacks. I would use e.g. Neutrik NRJ6HF1AU.
NRJ6HF-1-AU - Neutrik

Also suggest you mount the amp PCB to the front of the heatsink, as much away from the Tx as possible.

Patrick

Good advice. One channel is done.

 

[Hikari1]

One channel up and running. Offset is very, very stable and the amp is very cool. Things are going very well, can't wait to listen.

 

[EUVL]

No cross feed ?


Patrick

 

[Hikari1]

No cross feed yet. I am not used to listening w/crossfeed and thought it might be wiser to build without it first. Crossfeed + cap multipliers will be phase 2.

I am surprised how cool this amplifier is temp wise. Running 150ma bias but only one channel the amplifier is barely warm. At first I thought the mosfets were not turning on it is so cool. The 2107 case definitely is good choice for those who wish to keep the amplifier cool, if it's on a desk at work, etc. it would be a good choice. Bias could definitely be increased with this case if desired.

DC offset is so stable. Perhaps due to the massive sinks. I have never seen an amp or preamp with such stable DC tracking. I see a 30mv spike on turn-on/off and it stabilizes in a second to less than 3mv. And stays there. I can touch the mosfets (changing the temp) with no change in DC offset.

 

[EUVL]

Yes, the case is over-dimensioned for the purpose.
You can probably get away with a case without the sinks, but then I like the thermal stability with the 2107.




Patrick

 

[Hikari1]

Yes, who knows how stable the tracking will be with less heatsink. I will be using the amp at work, my coworkers will appreciate the lack of heat.

Any concerns about unplugging/plugging in the headphone jack when the amp is on? I don't have any current limiting on my perfboard build. I am concerned about a shorted output while unplugging/plugging in the amp but I may be wrong.