Very pleased with my Szekeres
Aug 4, 2002 at 7:07 AM Thread Starter Post #1 of 21

TimSchirmer

Repelling digital infidels. (Would that be called the Digifadah?)
Joined
Nov 22, 2001
Posts
3,246
Likes
82
I'm currently using a DIY Mosfet amp that I got from Thomas. It is a great little thing, The sound is very neuteral and flat. The mosfet has become my new reference headphone amp because of this, in fact. It leads me to believe that minimalism is the correct path to hi-fi bliss. Seriously...the more I upgrade my source, the better my rig sounds. It is very nice knowing that a 120$ headphone amp is not the weakest link in my system. I'm currently going to keep upgrading my analog rig (untill I have an origin live RB250 and Accuphase AC-3 phono cartridge. All of this will be fitted onto a CJ Walker quad-suspension turntable. Once I have this setup, I'll start auditioning headphone amps again, and honestly, I have yet to hear anything under 400$ that competes with this amp (havn't heard META42 yet) Eventually, I (sometime within the next 5 years) I want to have an Oracle turntable, with the OL tonearm, with a koetsu cartridge, all connected to a SET tube phono stage, then an earmax and HD600's.
 
Aug 4, 2002 at 9:19 AM Post #2 of 21
the szekeres amp is truly a good DIY amp. any details as to how thom built it? dc coupled? gain stain stage? or just a plain vanilla one?
 
Aug 4, 2002 at 5:03 PM Post #4 of 21
I'm glad you're enjoying your amp, i'm using a very similar amp myself for my home setup....
This is from what i remember, i built this amp a LONG time ago
smily_headphones1.gif


The amp is basically stock, with a few minor modifications...

The powersupply uses an external transformer, with rectification, regulation, and filtering inside the main enclosure...

Regulated Powersupply:
Over 10 000uF capacitance (including Black Gates and Nichicon Muse Electrolytics and SMT ceramic caps) , LM317

Amp:
- No gain stage, AC coupled.
- Holco resistors
- Wilma Polypropylene coupling caps (output cap is nichicon muse bypassed with wilma)
- Alps Blue pot
- jacks- Gold/tefolon RCAs, Neutrick Silver plated/locking jack
- Steel enclosure

Meier Crossfeed: Circuit is built right onto the DPDT switch:
Polypropylene caps, Vishay-draloric resistors

The whole circuit had a custom layout, and is wired point to point with CAT 5 (signal path) and 22 ga copper (power)


The szekere's s my favourite amp, by far the warmest and smoothest solid state amp i've heard, yet is still very detailed. I prefer it over the META for every headphone, though the meta with AD8610 may have a slight edge in the treble and is by far the best portable/opamp based amp i've heard. I strongly recommend the DIY'selvers here who haven't tried it to built one. There are no premade circuit boards, but the amp is not difficult to build.

(BTW, I am NOT selling building amps for other people any more)
 
Aug 5, 2002 at 12:19 AM Post #5 of 21
so the input of the meier xfeed is unbuffered? i'm guessing it was placed with the pot at the output of the xfeed.

nice parts. excellent power supply, how many volts is it running at?
 
Aug 5, 2002 at 2:03 AM Post #6 of 21
The Crossfeed was right at the inputs, followed by pot and the rest of the amp. No buffering, so its effectiveness varies with the output impedance of the source driving it...

It was running off 12V AC (rms), regulated to about 15V DC Wasn't an exact value since the regulator was adjustible, just set to the highest voltage where ripple/noise did not get through...
 
Apr 17, 2003 at 9:25 PM Post #7 of 21
I just received Tim's amp. Cute little blue and white thing.

I don't have a decent source here right now but the improvments are outstanding. The main apport is a feeling of ease and clarity. I was concerned about volume level with the HD580. To give you an idea, out of the line out of a ****ty expanium 103, the knob is set at the middle point for normal listening level (with crossfeed engaged) and even going to the end of the pot isn't really deafening. But I don't really care since it allows an extremely accurate tracking of the volume in the useful range.

In fact, I see exactly now how an headamp is not so much a volume booster. I had a time an airhead and it was just useful to increase the volume. This szekeres doesn't. But it brings alive the cans.

OOOOHH I LIKE IT.
biggrin.gif



PS : time to spend on the source now
 
Apr 18, 2003 at 7:58 AM Post #9 of 21
Hey,

Just so you know, you might have problem driving high impedance headphones since a bare MOSFET will start to hit non-linearity for larger voltage swing. Details drops because of this. On the other hand, it is a good one for Gradoes. Powerful current drive.

T
 
Apr 18, 2003 at 5:14 PM Post #10 of 21
Yes, I know. I've tried it out of a pitiful line-out and put in the cd player a few cds with quite a lot of intensity. With the HD580, I heard one or two problems but nothing to be really concerned about right now. I can live with till I receive an Art DIO which will take care of the gain (7V in output should do the trick).

I also try it on the cdp of a friend which output about 2V, most of the problems disappear. But it sounded like ****, because the cdp was ****. This amp is absolutly not euphonic, just the truth even if it's not pleasant.
 
Apr 18, 2003 at 6:05 PM Post #11 of 21
Hey,

I am not talking about gain. You see, this design cannot drive a load to high voltage swing even if you feed large signal. It is an unfortunate side effect of our favorite MOSFET characteristcs. It suit us good for low voltage swing but as soon as voltage swing gets too far, we will hit too much of non-linearity.

T
 
Apr 18, 2003 at 6:59 PM Post #12 of 21
I'm wondering... on headwize , you stated that you drove a pair of K240M (600ohms) with it. Could you give me an idea of these variations in linearity ?

I've tried this morning to push the amp to the limits with some large symphonic pieces but cannot really hear problems (other than bad sources). Till now, no clipping which is good news. Does the fact that i'm listening usually to really low levels (no kidding, I've excellent ears, at least in terms of volume, musical ears are another problem) play a role in this ?

Edit : I also found this you wrote some times ago on headfi :

"In my experience, Szkeres amps perform well with high impedance headphones. I used to use my Szkeres amp with AKG K240M with very good results. Theoretically, higher impedance headphones will give more linearer responses since the current draw by headphones is less. "

I'm a newbie in electronic and really confused right now
confused.gif
 
Apr 19, 2003 at 2:29 AM Post #13 of 21
Hi,

I realized that I was missing out of details when I built my current amp.

Bare MOSFET amps can't have too large voltage swing although it can offer a large current drive.

T
 
Apr 19, 2003 at 6:32 AM Post #15 of 21
Hey,

When I was experimenting with Szkeres amp and K240M. I noticed that by adding constant current source, I get more accurate sound. It took me few years to realize that that is because CCS is letting me have more linearity over the voltage swing.

You will notice missing depth in sound when you test META-like opamp based amp. It isn't the gain but this type have more linear response when it is operating at large voltages. But then the surging warmth isn't there. My THS4022 opamp based derivative is just plain transparent with hint of sweetness. I don't mind having any warmth.

Don't get me wrong. If I had Grado headphones or any 32 ohm cans, there is no doubt I would go back to Szkeres amp. Szkeres amp will offer all the accuracy I crave since it has to swing only 0.7V. It will stay very linear for that narrow range. PLUS, I get all that current these low impedance monsters need. (I ain't talking about low impedance cheapo's.)

Tomo
 

Users who are viewing this thread

Back
Top