Head-Fi.org › Forums › Equipment Forums › Headphone Amps (full-size) › Creek OBH-11 Test Results (OBH-1 vs. Elpac)
New Posts  All Forums:Forum Nav:

Creek OBH-11 Test Results (OBH-1 vs. Elpac)

post #1 of 19
Thread Starter 
Received my Elpac 24v 330mA linear regulated wall wart today (WM080-1950-760), plugged it into the Creek OBH-11 amp (non-SE) and did some testing. The male jack on the Elpac wall wart is center positive and of the correct size, and so the wall wart is a direct swap-out replacement for the Creek OBH-1 (note -- the Elpac has a 3-prong grounded plug).

First, a note on my philosophy: I feel that ears + measurements are often better than ears or measurements by themselves. Ears always come first, but measurements can be good pointers or indicators toward problem areas, and how they're likely to affect sound quality. In areas where they apply, they can also serve as a sort of objective verification of audible improvements, so there's less chance that we're imagining them or even missing a potential improvement somewhere.

Output is from the Creek's headphone jack to the analog line-ins of my M-Audio Audiophile 24/96 soundcard. Volume was set at minimum to test residual noise only, with nothing connected to the RCA input jacks of the Creek -- however, I did mess with the volume control to verify that nothing much changed at maximum volume either (negligible hiss or general noise floor increase).

Measurements were taken using Cool Edit Pro v1.2a software and its Analyze/Statistics function on a 10-second recorded sample of the Creek's residual noise (I'm not much of a RMAA fan, and prefer doing things my own way). The values below are not absolute, as the sound card has a noise floor itself and also adds a little low-level hiss called dither, which is normal and required in digital recording. However, the values are excellent for comparative purposes since the level of dither doesn't change.

Here are the results (interpretation below):

---------------------
Using Creek OBH-1 PSU:

Left / Right

Peak Amplitude:-60.94 dB-61.21 dB
DC Offset:-.001 -.001
Minimum RMS Power:-94.92 dB-95.12 dB
Maximum RMS Power:-61.76 dB-61.93 dB
Average RMS Power:-81.1 dB-81.2 dB
Total RMS Power:-77.1 dB-77.25 dB


Using Elpac 24v Linear Regulated PS:

Left / Right

Peak Amplitude:-80.61 dB-81.01 dB
DC Offset:0 0
Minimum RMS Power:-95.74 dB-95.5 dB
Maximum RMS Power:-91.93 dB-92.91 dB
Average RMS Power:-94.84 dB-94.62 dB
Total RMS Power:-94.82 dB-94.61 dB
---------------------

Interpreting the results:

Look at the following when comparing:

(1) Peak Amplitude -- this delineates the loudest the Creek's noise floor got during the time measurement was taking place. As these are in negative dB values, a higher number is better. The Elpac bettered the OBH-1 by nearly 20dB! This is a huge improvement that should translate directly to better sound.

(2) Total RMS power -- this measures average levels over time (in this case 10 seconds), rather than the highest residual noise peak that occurred. The Elpac bettered the OBH-1 by nearly 18dB, again a very large improvement.

(3) Minimum and Maximum RMS power -- look at the difference between the two values with each power supply, rather than comparing one PS to the other. This shows fluctuations and instability in the noise floor (often due to EMI and/or RFI as well as flaky power). With the OBH-1 it's bouncing all over the place like a kangaroo on crack -- confirmed visually using Cool Edit Pro's digital VU meters. With the Elpac it's quite steady (also confirmed visually).

The differences are stunning, and far greater than I'd expected. The test results show that the standard Creek OBH-11 is more sensitive to bad power and more responsive to improvement using a clean power source than is commonly believed, and that the OBH-1 is a horrible power source that adds a great deal of low-level noise and noise floor instability when used with the OBH-11. The grounding with the Elpac wall wart probably helps as well.

Preliminary Sonic Impressions: As I suspected (from previous experience with delivering cleaner power to amplification equipment), treble has improved greatly (biggest improvement) and there is a general unveiling, firming up and improved clarity throughout the entire sonic spectrum. Soundstaging has improved. Even bass seems to have lost some of the 'loose' character it had using the OBH-1 supply. I like the sound well enough now that I'm not going to attempt a capacitor upgrade and risk damaging the unit.

Final Impression: The Creek sounds like a completely different and much better amp. Get rid of that OBH-1 as soon as you can and get the Elpac (much cheaper) or an OBH-2.

Update (30 Aug 03): I realized this morning with some dismay that I had performed these tests with the wall warts plugged into my PC's surge protector! After banging my head on the wall, I quickly repeated the tests with the OBH-1 plugged into two different wall outlets, and with both the wall wart and the amp itself located well away from any computer gear. Thankfully I can report that the results didn't vary significantly. It did make me realize that your mileage may vary a little on the amount of improvement you see -- you may get less (or even more) than I did depending on how clean your wall power is in the first place, and what's located nearby the OBH-11.
post #2 of 19
Now this is the kind of info that this stie needs! Thanks for doing the leg work and providing us with this info fewtch. As a current owner of the little wonder I was curious as to how much the wall wart change would affect the sound of the amp. I guess we now know.

Thanks for pioneering this effort that will help all of us Creek users out there. If this does improve things that dramatically it would prove to cause a little more interest in getting a Creek if all it takes is a wall wart swap to make it sound even better.
post #3 of 19
What's the price for this PS, and what is a good source?
post #4 of 19
You'll have to get a check on which one specifically goes with this amp, but www.mouser.com has a ton of Elpac power supplies, and several of them in the 24V rating.
post #5 of 19
Thread Starter 
Also www.alliedelec.com -- the Allied Stock # there is 879-1080 (they seem to currently be out of stock).

Thanks go to a previous poster for locating it (unfortunately I didn't note their handle). Anyway, the PS that comes up if you enter the above number is the one I bought from mouser.com and used in the comparison. The Elpac part number is WM080-1950-760.

BTW, these should also work well with Tigger's (Headsave) Meta42's with a DC jack option installed -- 24v is what he recommended when I asked him about it. He didn't say anything about the polarity of the jack, however, but that's easy to change by cutting off the tip on the Elpac and re-splicing the wires reversed.
post #6 of 19
22 of them in stock at Mouser at $28 each.

http://www.mouser.com/index.cfm?hand...e_pcodeid=6801
post #7 of 19
Thanks much!
post #8 of 19
Is anyone else here thinking group buy, or is it just me? If you buy 5 of them you get like $4 off each one, so is anyone interested in buying one or more so we can get a total of 5 ordered to get the price break?

If so count me in for one.
post #9 of 19
Quote:
The male jack on the Elpac wall wart is center positive and of the correct size
Correction: it's compatible, but not exactly the right thing. Creek uses a 5.5/2.1mm jack, and the Elpacs use a 5.5/2.5mm plug. This means you're probably only contacting the center pin of the Creek's jack when you push the plug all the way in. With the proper plug, you'd still get solid contact with the plug almost completely pulled out. I recommend cutting the plug off the Elpac and putting a Switchcraft S760 plug on it. It'll cost you about $3 extra and be a bit of soldering work, but I think it's worth it. Just be sure you get the polarity right?

Quote:
www.mouser.com has a ton of Elpac power supplies, and several of them in the 24V rating.
Most of those are not what you want. You want the WM080, the only one in Elpac's line that is a linear regulated 24V DC power supply. The others are switching regulated supplies, and I don't believe any of them are isolated so they may not even work properly with the Creek.
post #10 of 19
Thread Starter 
Quote:
Originally posted by tangent
Correction: it's compatible, but not exactly the right thing. Creek uses a 5.5/2.1mm jack, and the Elpacs use a 5.5/2.5mm plug. This means you're probably only contacting the center pin of the Creek's jack when you push the plug all the way in.
An easy 10 second verification -- no. The power LED lights up with the plug less than 50% in. The plug on the OBH-1 is actually too big for the jack on the OBH-11, and sticks partway out (it does help when people actually have this stuff before providing corrections... ). Anyway, I don't know if this ever changed between revisions... I'd assume probably not. The Elpac works fine without alteration, in any case.
Quote:

Most of those are not what you want. You want the WM080, the only one in Elpac's line that is a linear regulated 24V DC power supply. The others are switching regulated supplies, and I don't believe any of them are isolated so they may not even work properly with the Creek.
Agreed that it's important to be careful to get a linear regulated and not a switching supply (especially after having read your webpage on power issues with op-amps ). As long as people stick to Elpac part number WM080-1950-760, they should be able to duplicate my results.
post #11 of 19
I do have an OBH-11, and it does use a 5.5/2.1mm jack, and the Elpac does use a 5.5/2.5mm jack. I suppose the weight of the cord is pulling it down so that the pin in the Creek's jack is contacting the inner side of the barrel.

How about you do some homework of your own: look at the plugs on your OBH-1 and compare it to the plug on the WM080.
post #12 of 19
Thread Starter 
Quote:
Originally posted by tangent
I do have an OBH-11, and it does use a 5.5/2.1mm jack, and the Elpac does use a 5.5/2.5mm jack. I suppose the weight of the cord is pulling it down so that the pin in the Creek's jack is contacting the inner side of the barrel.

How about you do some homework of your own: look at the plugs on your OBH-1 and compare it to the plug on the WM080.
Yes, it's a little shorter. No, the Elpac plug has no problems with contact when plugged into my OBH-11, even wiggling it around in the jack.

Please tangent, let's stop this here. People who buy OBH-11's are generally not DIYers and may be confused and put off by a suggestion to pull out the soldering iron, hunt around online for another plug, and replace the plug on the WM080. Your recommendation is posted here now, in case anyone really needs to do it.
post #13 of 19
Quote:
Originally posted by fewtch


BTW, these should also work well with Tigger's (Headsave) Meta42's with a DC jack option installed -- 24v is what he recommended when I asked him about it. He didn't say anything about the polarity of the jack, however, but that's easy to change by cutting off the tip on the Elpac and re-splicing the wires reversed.
All of the Headsave amps have the DC jack wired centre pin positive. The DC jack accepts a barrel connector of 5.5mm OD x 2.5mm ID and I don't recommend any power supply other than the Elpac WM-080-1950-760. Do NOT change the plug on the Elpac for use with any of my amps.
post #14 of 19
Thread Starter 
Quote:
Originally posted by tigger
All of the Headsave amps have the DC jack wired centre pin positive. The DC jack accepts a barrel connector of 5.5mm OD x 2.5mm ID and I don't recommend any power supply other than the Elpac WM-080-1950-760.
I'm not surprised, it really appears to be a darn clean power source. I'm still surprised at the degree of improvement it made with the Creek -- I was expecting some, but nowhere near that much.

BTW, for anyone who may be interested (not related to Tigger's amps) and ever needs 12vdc for something, the WM060-1950-760 is the 12 volt version of the WM080, and supplies 500mA (in all other respects it appears to be identical). Elpac's WM-series in general looks good, and if I ever need another voltage that's what I'll be researching.
post #15 of 19

Yeah...

...Tangent, Fewtch, others and I had a whole thing on the WM-080 with the OBH-8 in the Do-It-Yourself forum where Tangent specifically said to go for the WM-080, as it is a linear (non-switching) regulated power supply, therefore less noisy. There are other 24V Elpacs which have higher mA ratings, but the WM-080 is the one you want.

Just to back up what Fewtch said, the WM-080 PS has made a very appreciable difference in:

1. Soundstaging (wider, freer)

2. Bass solidity (non-boomy and very musical, deeper, more dynamic and punchy)

3. Dynamics (more finer shades as well as greater, punchier peaks)

4. Tone/Timbre (more texture)


5. Front-to-back imaging (there actually is some now...plus, the enhanced inner detail helps with this)

with my OBH-8 Phono Preamp.

At $28, the WM-080 is an absolute steal for Creek owners. Truly one of the great audio bargains out there.

For instance, I am now enjoying just about every rock 'n roll record I have, since I now have some punch to the rhythm section, instead of the compressed mush I had with the OBH-1. Bass has actually "backed off," but in a very good way: I get the rippley bass-string texture now, instead of just the wooly pressure. There is better front-to-back imaging (all of this stuff is contained in the aforementioned post:

http://www4.head-fi.org/forums/showt...ighlight=creek

). Rock records have gone from just about uniformly uninteresting to tons of fun with the mere switch of a power supply.

I have a feeling Mouser is going to have a run on these things.

Regards,
Sir Mister Matt
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Headphone Amps (full-size)
Head-Fi.org › Forums › Equipment Forums › Headphone Amps (full-size) › Creek OBH-11 Test Results (OBH-1 vs. Elpac)