Head-Fi.org › Forums › Equipment Forums › Headphone Amps (full-size) › Crack;Bottlehead OTL
New Posts  All Forums:Forum Nav:

Crack;Bottlehead OTL - Page 9

post #121 of 4706

People say they can hear the difference in capacitors.  I have seen some studies that tends to support this idea, but it really isn't "solid proof" that you can hear the difference.

 

Some capacitors make a MEASURABLE difference. My Musical Fidelity XCAN v3 had some really cheap filter caps in the power supply; replacing those cheap ones with with quality capacitors with lower ESR made a slight but measurable difference in the the amp's S/N ratio, especially at the top end of the audio band.  And, removing the DC blocking cap off the input and changing the output coupling caps to non-polarized parts with higher capacitance made a measurable improvement in low frequency response.

 

Coupling capacitors - capacitors that connect the preamp stage to the amp stage, or those that connect the output of the amp to the headphones-  people say that electrolytic caps sound distorted when used in these circuit positions.  The audio is flowing directly through these capacitors, and maybe people can hear differences in sound quality between electrolytic capacitors and polyester or polypropylene film capacitors, or oil-in-paper capacitors.  Certainly these capacitors all differ in certain real, measurable properties like ESR and so on, and so maybe you'll hear differences. However, capacitors rated to handle the high voltages inside a tube amp and capable of the high capacitance values needed in coupling capacitors are expensive.  It seems kind of silly to use $600 coupling capacitors in a $200 amp.

 

Bypass capacitors can also be important, some say, and most people like film capacitors in this application. It doesn't look to me like the stock Bottlehead Crack amp uses any - or has any need for bypass caps anywhere, though.

 

Power suppliy filter capacitors should be good quality, low ESR electrolytics, and it looks like the Bottlehead Crack uses good quality parts for this already.

 

So, I'd say, build the Crack in stock form.  Then, watch the Forums to see what people do to modify them.  See if anyone finds coupling caps that improve the sound.  You could go out and BUY a few hundred dollars worth of exotic caps and experiment yourself to see if any improve the sound.... but I'd guess you have better things to do with your money.

 

For example, people in the Bijou D.I.Y. tube amp forum have found some oil-based motor starting capacitors to replace the electrolytic output coupling caps and they report that they sound better. These caps can bought off eBay from someone who is selling them to use with motors  and doesn't realize that there are audiophiles out there who think oil-in-paper capacitors are "magic" and will pay silly prices for them.

 

Another tweak you can do is trying different tubes. There are tubes currently manufactured in Russia and China, and I think the Czech Republic too; maybe you'll like the Chinese tubes and hate the Russian ones.  Also, there are "New, Old Stock"  (NOS) tubes available on the market that have been sitting on some shelf since the early 1960's when tubes were still common.  There were many manufacturers of tubes in the US, England, Holland and Germany in those days.  You might find that the RCA tubes are great, or maybe Telefunken.  Be careful, though, some tubes have gained such a reputation that the prices have shot through the roof- and so there are people out there COUNTERFEITING them!  See http://reviews.ebay.com/Amperex-Bugle-Boy-12AX7-ECC83-preamp-tubes_W0QQugidZ10000000002395429  this guy explains how to spot certain fakes.

post #122 of 4706

Would this amp work well with Beyer DT-990 600 ohm phones? I know crap about the fundamental laws of electricity but I read this amp works best with high impedance cans, and my cans are 600ohm... that seems pretty high but it doesn't make sense to me as to why a lower power amp works well with something that is harder to drive. Any help would be welcome.

post #123 of 4706
Quote:
Originally Posted by dynasonic View Post

Would this amp work well with Beyer DT-990 600 ohm phones? I know crap about the fundamental laws of electricity but I read this amp works best with high impedance cans, and my cans are 600ohm... that seems pretty high but it doesn't make sense to me as to why a lower power amp works well with something that is harder to drive. Any help would be welcome.


I'm not sure about the electrical reasons for this, but I can tell you that some tube amps have an unacceptably high noise floor with low-impedance 'phones. My SSMH, for instance, sounds terrible out of my SR225's, but sounds great with my HD580s. I'm not sure what, electrically, is making this happen, but from a practical viewpoint, one has a loud hiss in the background and the other doesn't.

I think 600Ohm is a pretty standard value for "high impedance", so you should be good.

post #124 of 4706
Quote:
Originally Posted by dynasonic View Post

Would this amp work well with Beyer DT-990 600 ohm phones? I know crap about the fundamental laws of electricity but I read this amp works best with high impedance cans, and my cans are 600ohm... that seems pretty high but it doesn't make sense to me as to why a lower power amp works well with something that is harder to drive. Any help would be welcome.


Very interested in this myself, hoping someone has an answer.

post #125 of 4706

The amp works best with higher impedance headphones because the amp is capable of developing more output voltage into a higher impedance.  Lower impedance loads, like Grado headphones, might "load it down" too much, with the result that a listener to the lower impedance headphones might not be able to play music LOUD ENOUGH.  That's the main reason that this unit is said to be more suitable to the typical 300~600 ohm "high impedance" headphones.

 

I think aristos_achaion's experience might be more likely attributed to the greater efficiency of his Grado headphones when used with a solid state oputput stage like the SSMH. Their efficiency (ability to turn voltage into sound) is high, and so the low voltages of noise on the output stage of his SSMH become audible on his Grados but not his Sennheiser 580's.  It should be noted that the SSMH uses a SOLID STATE output stage which delivers more power into a lower impedance headphone than it would into a higher impedance headphone.  In the SSMH the headphones are not connected to the tubes; the 19J6 tube is a gain stage which increases the voltage of the incoming signal, and the tube passes it's signal on to an IRF510 MOSFET - a type of transistor- which provides the power to drive the headphones.

 

In the Bottlehead Crack, there is a tube gain stage and a tube power stage.  There are no transistors at all.

 

Typically, with a tube output stage you will get good highs without solid state "glare" which comes with many transistor amplifiers. However, since the tube has a higher output impedance, it's damping factor is lower and there may be a loss of "tautness" or "control,"  especially in the low end.  This is one reason why tube amps sometimes sound a bit "wooly" on the bass.  These are gross generalizations, however, and a lot depends on the actual circuit and the headphones used with it.  A tube design which does not use an output transformer to couple the headphones to the tube often has very good extension in the bass which can temper the "wooly" lows and result in a sound that is a bit "warm" or "rich" without seeming to have sloppy midbass.  The Bottlehead Crack is such a design.  

 

 


 

post #126 of 4706

The output impedance on the amp is 120 ohms so I am wondering is this amp would pair well with the DT 48 E which has a nominal source impedance of 120 ohms although the nominal impedance is rated at 200 ohms. 


Edited by tdogzthmn - 5/11/10 at 5:25am
post #127 of 4706

 

Quote:
the DT 48 E which has a nominal source impedance of 120 ohms although the nominal impedance is rated at 200 ohms.

 

I have not listened to DT 48 E so I can't give you a subjective answer, but the "nominal source impedance" means the output impedance of the headphone amp (and in fact 120 ohms is a rather ignored standard set by the IHF many, many years ago for headphone outputs). The "nominal impedance" of 200 ohms appears to be the actual impedance of the headphones. So on paper it looks like it should work well.

post #128 of 4706

I 2/3 finished my Crack build earlier this evening.

 

One of the pieces of wood I received in my original package came with a nasty chip in it, A quick email and all was better. I built the actual "amp" in an all night pleasure fest of solder smoke and soda. The building of the amplifier was fairly uneventful. I penned in DOC's notes for the corrections necessary and got busy with the soldering.

 

Today I got the base built and the amp all tested and ready to fire up. The process was also fairly uneventful. Its really boring when things just work, so to throw me a curve ball one of my tubes is defective! Mneh, these things happen.

 

The behavior of the output tube is quite odd based on my previous experience. The tube runs well for a few minutes and then 1 section (the right channel) stops working with a soft pop. It was quite the shock the first time it happened listening to music, but cooling the amp and restarting it got the gears turning.

 

Despite my best efforts to make tracking the problem down as difficult as possible my suspicions were confirmed when the cathode voltage (pin 3) on the output tube was measured and 0V came up when it was supposed to be (and previously was!) 100V. All I needed to check this thing out was a multimeter!

 

To determine that the output tube was at fault I employed a wide selection of expensive test equipment to follow a 10Khz signal through the amp. On initial startup the amp performed as expected swinging in excess of 10vp-p into 300 ohms before clipping. After several minutes the right channel cut out and I traced the signal through the amp (I have never actually done this before so I was very excited to have a reason to do so!) and found it passed through the 12au7 properly and got to the grid of the 6080/6as7 OK but then went away. The other channel still went end to end OK. It was at about this time that I thought to do a simple voltage check which came back "really wrong". Its amazing how complicated one can make something they only needed a multimeter to figure out when they have all sorts of fancy test equipment handy.

 

Satisfied that I solved my problem and seeing how everything was already hooked up I decided to measure away. I got numbers largely in agreement with those posted by bottlehead as to the bandwidth, maximum voltage swing, and the like for the left channel.

 

My numbers were:

Note: all numbers are for 12vp-p swing, into 300 ohms.

 

Bandwidth: (-3db, volume control set to max) 10hz to a bit more than 80Khz

Voltage swing: 12vp-p before "xy" breaks a line. The sine wave still looks pretty pretty at 15vp-p.

10K square wave response: Very steep leading edge with no overshoot or ringing at all. Overall very good.

post #129 of 4706

So did you get a new tube? Or are you still unable to use the amp?

post #130 of 4706
Thread Starter 

Just give Eileen a call. If you can catch her before stuff gets shipped out, she can get it on it's way to you before the weekend.

post #131 of 4706

It's a good idea to pick up a few extra 6080/6AS7G's anyway - they are cheap as dirt, and you often have to go through a few to find one that isn't badly microphonic.

post #132 of 4706

It could well be that the tube has some sort of heat related issue and we are happy to send a replacement, just give Eileen a call or email. The tubes have all been tested on both sections, but unfortunately my tech doesn't have time to let them each sit for several minutes in the tester (we use a TV-10D/U). So that kind of heat induced short or open could sneak through the screening process. Just for the sake of being thorough I would suggest making sure the other possible causes for 0V on the cathode be covered - rewet the solder joints at both ends of the cathode resistor and the cathode pin of the tube socket, and try giving the tube pins a little polishing with something like a scotch brite pad. Those are also places where a heat related open circuit could occur.

post #133 of 4706

I resoldered all of my joints between pin3 and ground (and pin2 as well) and the tube still stopped working when it warmed up. 

 

No worries, and I agree it is an unusual problem that a tube checker would not catch. Eillen got a new tube out to me in the mail so things should be great soon.

 

For the song or so it did play it sounded awesome, and the measurements look very good too. I'l post better impressions when it gets running under full power.


Edited by nikongod - 5/15/10 at 4:14pm
post #134 of 4706
Thread Starter 

Hey nikongod,

 

A buddy of mine just got a pair of R10's.

He's got a Crack kit that he hasn't built and he's figuring on selling it since the R10's are lower impedance than it's recommended for use with this amp.

I've got a pair of ATH AD2000's that sound great on my Crack (with certain tube sets especially). They're about the same impedance as the Sonys.

When you get a chance, could you post some impressions of how the your R10s sound with your Crack?

It would be much appreciated.

post #135 of 4706

Some Pics of my crack.

 

Im about the worlds least skilled wood worker. Good news, the wood looks nice with some sanding (coarse and fine paper) and 2 coats of tung oil. Tung oil FTW. I need to put 1 more coat of tung oil on and then I will put the little feet on the amp.


aDSC_0018.JPG

 

The end bell for the transformer was begging for some hammertone love. I cleaned as much of the protective grease and oil off as I could and sanded it a tiny bit just because. Hammertone paint is awesome for stuff like this because it covers and fills little voids in the metal ooh so very nicely.

 

aDSC_0017.JPG
 

My initial impressions with the HD800 and the T1 (on loan) are that the amp has a slightly "sweet" sound. It draws me in, and it is a touch of color, but it is not syrupy like some other amps. It still has good edges on transients and keeps things from smearing together VERY well. The image placements (in 3d!) and separation are quite good.

 

Im not sure what else to say about the amp, I have surely spent more money and time on less sonicly rewarding projects in the past.

Quote:
Originally Posted by ironbut View Post

Hey nikongod,

 

A buddy of mine just got a pair of R10's.

He's got a Crack kit that he hasn't built and he's figuring on selling it since the R10's are lower impedance than it's recommended for use with this amp.

I've got a pair of ATH AD2000's that sound great on my Crack (with certain tube sets especially). They're about the same impedance as the Sonys.

When you get a chance, could you post some impressions of how the your R10s sound with your Crack?

It would be much appreciated.


I think I should note first that I tend to listen fairly quietly(60-65db peak). If your friend listens really LOUD I suppose he may run the amp out of class A which would of course be terrible.

 

On that note, at my low listening levels the crack holds up well with my R10. The sound is overall very good. Despite the fact that the crack should work better with the HD800 than the R10 a quick head to head still puts the R10 on top. IMO the biggest strength of the R10 VS the HD800 is that it just catches EVERY little detail where the 800 is a hair smoothed over still. The R10 still shines just like that.

 

I have a bass-heavy low-serial number pair of R10 (crazy! #35x) and the bass performance is good despite the fact that the calculator says that the output caps are too small for a low impedance headphone.

 

As another thought:

Although it kind of goes away from the "OTL" idea, it should not be too hard to wire up a set of transformers (or autoformers) in their own little box to put between the amp and the headphones.

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) › Crack;Bottlehead OTL