Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › My take on the entech
New Posts  All Forums:Forum Nav:

My take on the entech

post #1 of 20
Thread Starter 
Modding the Entec 203.5 Dac

The aim was to raise the performance of this pint sized unit that already sounds pretty good. My budget was about £20 but ended up going a bit over it …that was the fault of the opamp I chose.

Opening the case its clear that the unit is reasonably well built with relatively good quality components.

The solder used was 4% silver lead free …so the label stated.

D1 and D2 caught my eye …..these are just standard 1n4001 rectifiers so out they came and in went a pair of 11QD10’s

C48 and C49 these are the reservoir caps not a make I recognise…so out they came and in went a pair of pannasonic fc [1200uf 63v ].

All the Muse caps are better than what I had to hand …so I left them alone apart from bypassing each with a 0.1uf 63v wima film cap. You need to be careful as there’s little room under the pcb so they need to be slim.

Next up came the opamp opa 2134 is pretty good in my book …..but there are newer units now out there. I desoldered the opamp and fitted a high quality turned pin socket. Looking through my stock of opamps I found a opa2227 so what the hell …lets try it

Looking at the signal path its clear there are not caps in the output .. hooking up my oscilloscope thought I saw a trace of dc …..that’s easy to sort I added a pair of 4uf 150v paper in oils these are the size of three Jacobs cream crackers …with the case work fitted I was able to just squeeze them in .

There are two anonymous resistors in the output these were swapped for the same value dale vishay rn55’s .

Then finally I removed the nasty pcb mounted rca sockets replacing them with bulkhead mount types ….these were then hard wired with silver in ptfe from the pcb.

The results are very good , it seems far more musical than before with a sense of ease to the sound that belies its budget.

Well worth the couple of hours work involved.

Again these are only suggestions I'm certain there are many other similar solutions.....get experimenting
post #2 of 20
10x for sharing.

Any issues with the PCB...like flimsy traces or discolored/burned board...?

I'm surprized you opted 2227 for 2134 - the latter sound better to me... I'd eventually go with 2107 there...what's on the power pins of that one, have you measured by any chance

The DC offset on mine is 1.4/2.5mV so at 2.5Vpp max I would't bother putting any caps there, but that's just me...

Looking at the board, seem to me all Muses are already bypassed with those yellow 1J100 rectangular caps, or I'm wrong...I guess I'll go for the diodes, powercaps and the opamp...
post #3 of 20

Entech Mods

Yes, I think the power supply is pretty good with generous use of film caps and regulators. The outboard transformer has a ferrite filter on the cord so I doubt there will be much ringing on the diodes.

My first try with the op-amp was OP275, an audio-specific device which AD says has a "Butler" front end (jfet-bipolar combo?). I thought it provided more detail and tighter bass but not a great improvement.

I began to look at the output stage in terms of requirements. Low offset is needed since there are no output caps. The impedances are balanced and low enough in the LPF, < 10K ohms at both inputs.

So next I tried the LM6172, a bipolar amp with a lot of bandwidth and slew rate. It sounded very good, a lot of body and a liquid feel to the treble, no sense of strain on peaks. (I checked the offsets and found 2mv and 4mv. On the original OPA2134 I had 0mv and 5 mv.)

OK, get out the costly devices. A couple of AD825 SOIC singles on a Brown Dog was next. Humm, disappointment. The sound is very detailed and effortless but thin and dry, not much bass impact. Maybe the wide open-loop bandwidth (~10Khz) and fast settling time (~80ns) are not enough in this circuit, where the higher frequencies are attenuated by the LPF feedback.

So I put the LM6172 back in and was again impressed. I decided to order some OPA2227 and OPA2107 amps for more testing. Maybe I should get a 78L12 and 79L12 to change the voltages so I could use the AD8620 and AD8066. I generally like the AD sound if not too edgy.
post #4 of 20
Quote:
Originally Posted by joe_cool View Post
Yes, I think the power supply is pretty good with generous use of film caps and regulators. The outboard transformer has a ferrite filter on the cord so I doubt there will be much ringing on the diodes.

My first try with the op-amp was OP275, an audio-specific device which AD says has a "Butler" front end (jfet-bipolar combo?). I thought it provided more detail and tighter bass but not a great improvement.

I began to look at the output stage in terms of requirements. Low offset is needed since there are no output caps. The impedances are balanced and low enough in the LPF, < 10K ohms at both inputs.

So next I tried the LM6172, a bipolar amp with a lot of bandwidth and slew rate. It sounded very good, a lot of body and a liquid feel to the treble, no sense of strain on peaks. (I checked the offsets and found 2mv and 4mv. On the original OPA2134 I had 0mv and 5 mv.)

OK, get out the costly devices. A couple of AD825 SOIC singles on a Brown Dog was next. Humm, disappointment. The sound is very detailed and effortless but thin and dry, not much bass impact. Maybe the wide open-loop bandwidth (~10Khz) and fast settling time (~80ns) are not enough in this circuit, where the higher frequencies are attenuated by the LPF feedback.

So I put the LM6172 back in and was again impressed. I decided to order some OPA2227 and OPA2107 amps for more testing. Maybe I should get a 78L12 and 79L12 to change the voltages so I could use the AD8620 and AD8066. I generally like the AD sound if not too edgy.

Try the LM4562 you may be very, very surprised I am. Just awaiting feedback from Nigel (fellow Head-Fier / audiophile) in Liverpool (I posted out his chip yesterday).... as far as I'm concerned from initial auditions this is the chip. Tried it in the Graham Slee Solo and a few others and, yes YES YES.............
post #5 of 20

LM4562 And Entech 203

LM4562NA (DIP8) and LM4562MA (SO8) on order (from Singapore?). Three of each for ~$20.

How long does it take to get samples? TI and AD are very fast.
post #6 of 20
Thanks for the info on your mods zanash, bit advanced for me at the moment, but have retained your post for future use.

Will just try changing the opamp first.

Cheers - Harryo
post #7 of 20
Thanks to Pink Floyd for the amp recommendation (LM4562). It took about 2 weeks to get chips but it was worth the wait. I tried OPA2134, OP275, AD825, OPA2227, OPA2107, LM6172 and even NE5532. The LM4562 is better for this DAC by a significant amount.
post #8 of 20
Quote:
Originally Posted by joe_cool View Post
I tried OPA2134, OP275, AD825, OPA2227, OPA2107, LM6172 and even NE5532. The LM4562 is better for this DAC by a significant amount.
That is good to know, I have six LM4562NAs en-route from Digi-Key, no doubt they are tied up somewhere in the Christmas post!

One is for use in the Entech DAC the other five are for my Audiotrak Prodigy 7.1 Hi-Fi soundcard.

Harryo
post #9 of 20
Quote:
Originally Posted by PinkFloyd View Post
Try the LM4562 you may be very, very surprised I am. Just awaiting feedback from Nigel (fellow Head-Fier / audiophile) in Liverpool (I posted out his chip yesterday).... as far as I'm concerned from initial auditions this is the chip. Tried it in the Graham Slee Solo and a few others and, yes YES YES.............
Thanks... I mean THANKS..............
post #10 of 20
where can one purchase a single Dip-8 LM4562 ?
post #11 of 20
Quote:
Originally Posted by NiceCans View Post
where can one purchase a single Dip-8 LM4562 ?
If you go to www.national.com , you can have a few samples sent to you by paying a small fee (at least this was for me).
post #12 of 20
I got into my ENC 203.2 tonight, and swapped 2134 for BB2107/BB2132/AD823/AD8620. Can't hear much of a difference really...2107 seem to be smoother, but maybe that's because I know 2107 is smoother so...even with AD823 DAC sounds almost the same to my ears......finally I went back to 2132.

I'm looking forward to try that 4562; I can't justify messing up with the rest (diodes, caps etc)

I finally got LM4562. I could confirm that it's the best sounding chip in ENC so far - ultra-detailed to the point of being unnatural especially in the mid-bass area, with smooth ...well maybe slightly veiled top (823 best here), and milky vocals. Not BB, not smooth chip, that's for sure...kinda like more deatiled 843

Mercsylessly exposes any miniscule flaw of the recording though. Probably's gonna be an overkill for anything but Senn
post #13 of 20
Bumping an old thread... Finally got around to socketing the OP amp of my 205.
I must have messed up soldering the socket, or maybe I bumped something on the PCB. At first my left channel was all static-ey and intermittent?? So I went back in and re-flowed all my solder joints. While I was at it I reflowed all the solder joints for all the caps on the output side. PROBLEM SOLVED!! I'm not sure what I did, but evidently either I bumped something on the PCB or I had a cold solder joint somewhere.

I haven't tried that many OP amps yet I have a 2107 in it right now and it is warmer than the 2227, 2132 and the original 2134. Better detail than the 2123/4. Not as airy as the 2227. I have a JRC4556 I might play around with.

Half the members of this thread have been banned Andrea and pinkfloyd are both Senn HD fans so, I'm wondering how that LM4562 OP amp will sound with my RS1 and A250? The 2107 sounds very good... nice and warm. Good synnergy with my Larocco PPA / RS1.

If anyone can recommend some more warm, bassy yet open/detailed OP amps??... I'm all ears!!

Anyone know if I can plug my adapted 8610 or OPA627 right in? I'm not sure what +/- voltage there is at the OP amp, if it matters at all the unit runs off a 16V/DC power supply.

Thanks!!!
Garrett
post #14 of 20
Just wait for andreas fourth return to headfi

I saw him trying to come back two times now under a new name, but being banned again right away.

Four times is the charm perhaps^^
post #15 of 20
I'm not 100% sure but I think ENC 203.2 runs at 9V, not 12V...there's a LM7809 in the lower left corner of the PCB...or is it 7812...

Anyway, yes you can fit single or dual BrownDog there - I rolled several chips and 8620 was working fine...I wasn't able to get much of a difference from most of the chips except LM4562.

My guess is 2107 should be an excellent choice for Grado...try 823 for smth different. After 2107 4562 will be bright and probably harsh
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › My take on the entech