[BrokenEnglish]

Quote:

Originally Posted by spendorspain Drawbacks? Only a very minor and I hope not worrying: when I rotate the volume control fully counterclockwise (I always do that before turning the amp off or when changing interconnects) there is a faint crackling, but only at the very end, never in the useful range of pot. I now simply stop their rotation a little before it ends (anyway, at this point the sound level can be -60/70 dB, I guess).

hi jose,
nice to read from another "uncapped" wna! i described the same symptoms of crackling while turning the volco after removing the input caps. it's neither related to plastic or metal knobs, nor to incomplete grounding of the pot. most likely it's caused by dc-offset returned to the input, as dr. white told me some time ago. i was hoping to get rid of it with the 10m-resistor and 100k-trimpot mod, but couldn't test it so far. stay tuned.. i'll try it soon.

Quote:

Originally Posted by PinkFloyd to cut it short you can use the LM6171 with 10M resistor and trimpot only if you leave the input caps in place....

mike... do you think, one could use the 10m/100k "caps-free" mod in combination with output caps while leaving the input caps out? ah... i'll try it anyway...

btw... i finished my dual-voltage psu (basically 2 tweaked TREADs with electronic ripple and noise rejection, currently 2*14v DC) to bypass the wna-railsplitter. after 2hs of usage there were no audible advances. but... BUT... now i'm able to increase the rail capacitance without the risk of further transistor death. i started with 1000µf electrolytic, bypassed with 10µf tantalum, bypassed with 0.082 + 0.022µf fkps. very intense bass, nice highs. i'll give it a week of burn in before i report back.

 

[PinkFloyd]

Quote:

Originally Posted by BrokenEnglish mike... do you think, one could use the 10m/100k "caps-free" mod in combination with output caps while leaving the input caps out? ah... i'll try it anyway...

I really don't know Udo, that question will have to be posed to Dr. White... funny things were going on when I ommitted the input and output caps and fitted the 10M and trimpot.. as I turned the volco the offset changed pretty dramatically....

Without output cap / input cap AND without the 10M and trimpot (just LM6171 fitted in IC1) both LM6171's measured VERY low offset (+1mV and +4mV) so I thought "hello there, two well matched LM6171's this is good"

BUT........ when I turned the volco clockwize the offset went CRAZY... the more I turned the volco the crazier it got ( something like +150mV and -300mV) I'm glad I didn't have my headphones plugged in and I'm glad I didn't rely on the reading with the volco at minimum position.

I explained this to David and he says It was due to the input caps not being in place...... it sure seemed pretty weird to me and I trust David will explain why this was happening shortly... I'll post as soon as / if I hear anything.

Quote:

Originally Posted by BrokenEnglish btw... i finished my dual-voltage psu (basically 2 tweaked TREADs with electronic ripple and noise rejection, currently 2*14v DC) to bypass the wna-railsplitter. after 2hs of usage there were no audible advances. but... BUT... now i'm able to increase the rail capacitance without the risk of further transistor death. i started with 1000µf electrolytic, bypassed with 10µf tantalum, bypassed with 0.082 + 0.022µf fkps. very intense bass, nice highs. i'll give it a week of burn in before i report back.

Ah Nice one Udo

Now you don't have to worry about taking out the transistors and CRD's in the rail splitter

That was so funny mate, both you and I fitting crazy caps in the rail splitter and wiping out the transistors and CRD's.... my amp continued to work though and actually sounded great with the burnt out transistors..... it was only when I glanced at them one day I noticed the scorch marks on them and alarm bells started ringing

The large inrush current (using the large caps) had obviously blown the arse out of the trannies and CRD's..... what amazed me though was the fact the amp still worked and sounded absolutely FAB with half the railsplitter toasted

Do you have any pics of your new PSU Udo? I'd really love to see it mate. I'm happy It's working well and even happier that your WNA sounds better with it...... now you've got the PSU sorted out you can maybe try a MKll board with it as provision is made on either PCB1 or PCB2 to bypass the rail splitter..... in fact, there are a lot of PSU options on the MKll boards if you're into playing about with PSU configurations

I think the amp could be improved by incorporating a TLE2426 into the railsplitter circuit and, for the kit builder on a budget, it would mean they would get even better performance from a cheap wallwart PSU and it would save on a few resistors.........

First and foremost I'm searching for bang for the buck (not for me but for others new to the scene) and am a great believer in evolution rather than revolution...... My part in this is simple...... I want people to get the max at the minimum cost which is why I tend to stick to the recipe and report on a cheap ingredient that may make it taste better......... I seldom add saffron and oysters to my recipes.......

I'm starting to go off on a tangent but, in a nutshell, the WNA kit "as is" is a superb amp but adding saffron / oysters and other exotic ingredients is an option that adventurous chefs have up their sleeve and a recipe which may taste wonderful but the baked beans and toast approach I adopt is aimed at the beginner.

Mike.

 

[BrokenEnglish]

Quote:

Originally Posted by PinkFloyd Without output cap / input cap AND without the 10M and trimpot (just LM6171 fitted in IC1) both LM6171's measured VERY low offset (+1mV and +4mV) so I thought "hello there, two well matched LM6171's this is good" BUT........ when I turned the volco clockwize the offset went CRAZY... the more I turned the volco the crazier it got ( something like +150mV and -300mV) I'm glad I didn't have my headphones plugged in and I'm glad I didn't rely on the reading with the volco at minimum position. I explained this to David and he says It was due to the input caps not being in place...... it sure seemed pretty weird to me and I trust David will explain why this was happening shortly... I'll post as soon as / if I hear anything.

yeah, mike... that's the sort of behaviour one would expect from our experiences without input caps (leaving the output caps in). i'll try the 10m/100k thingy as next step. but one thing is sure: as long as i own this amp, it will be free of input caps. i'll also try the effects of decreasing the input impedance to 10k. let's wait and hear...

(oops... i think, you don't like this smilie at the moment... so... for you mike:

...sorry... couldn't resist...

)


Quote:

That was so funny mate, both you and I fitting crazy caps in the rail splitter and wiping out the transistors and CRD's....

yeah.. i'm still traumaticed... i don't like pluggin in the dt440 any more, the phones i used when i "grilled" the transistors the first time...

Quote:

Do you have any pics of your new PSU Udo? I'd really love to see it mate.

here it is: http://www5.head-fi.org/forums/showp...&postcount=564

Quote:

I'm happy It's working well and even happier that your WNA sounds better with it......

well... i'm not quite sure, if it sounds really better. i don't have my reference phones (dt880) with me and also have to wait for further progressions of burn in. but... you know... it's so much fun to have the option of playing with rail capacitance...

Quote:

now you've got the PSU sorted out you can maybe try a MKll board with it as provision is made on either PCB1 or PCB2 to bypass the rail splitter..... in fact, there are a lot of PSU options on the MKll boards if you're into playing about with PSU configurations

hehe... you devil...

...i'll join the club... as soon as i've done all the modding with the mkI that will guide me to my dream mkII....

Quote:

I think the amp could be improved by incorporating a TLE2426 into the railsplitter circuit and, for the kit builder on a budget, it would mean they would get even better performance from a cheap wallwart PSU and it would save on a few resistors.........

well... i don't think so... the tle2426 most likely will not provide enough current for the wna. eventually dr. white could think about replacing the railsplitter ic with a buffer (buf634?) to give us a little bit more juice. but maybe this is just a silly idea...

wish you a nice and painless evening, mike...
udo

 

[PinkFloyd]

Quote:

Originally Posted by BrokenEnglish here it is:

It's here too looks GREAT Udo!!:

Beautiful job Udo, truly FAB

Mike.

 

[12796]

Any update on the ad843 Pinkie? Still waiting for mine to arrive...

 

[PinkFloyd]

Quote:

Originally Posted by biovizier Any update on the ad843 Pinkie? Still waiting for mine to arrive...

Hi Nick,

I'm still listening to it and my head is a lot clearer than it was a couple of weeks ago..... I still like the AD-843 a lot. This is one of these instances where it would be ideal to have two WNA's running side by side through a comparator, one with the LM6171 on board and the other with the AD-843 on board.... by the time you remove the AD-843 and replace them with LM6171 and output caps etc. It's hard to remember what the other opamp sounded like.

Switching between two amps through a comparator would give me a much better, and more immediate, insight into the qualities of both opamps so I may well build another WNA solely for opamp evaluation.... I know this is bordering on being clinically obsessed with opamps but, what the hell, it keeps me off the streets and out of mischief

I had a peek at the AD843 datasheet earlier and snipped this little morsel out of it:

I'll have a go at the bypassing over the weekend and if I do build up a second WNA amp to compare opamps I'll ditch the dip8 sockets and solder the AD843 directly to the board as suggested in the datasheet.

One thing I can say Nick, the AD-843 and LM6171 are the only two opamps that can reproduce my reference recordings accurately.... the OPA627, AD-811, OPA134 etc. make mincemeat of the recordings and totally bastardise the pace, rhythm and timing.... they add insult to injury by seemingly placing a 500 denier veil between the headphones and my ears.... the 627 is pretty rank sounding compared to the LM6171 / AD-843.... so much so that a comparator is not necessary to differentiate between 627 and LM6171 / 843... OPA627 just sounds plain wrong in the WNA application. The LM6171 / AD-843 are a different kettle of fish and "seem" pretty close to the mark when listening to my reference tracks and you really would need to have either a prolonged listen to both or a brief switchable comparo between the two before you could make your mind up....... LM6171 V OPA627 is night and day, AD-843 V OPA627 is night and day, AD-843 V LM6171 is pistols at dawn.... hard one to call.

Having said that, as always, It's all dependant on the synergy of your system and your own ears.... I can only comment on what I hear and I'll only listen to something that I like and never rely on other peoples reviews as we all have ears which are as unique as our fingerprints (no two ears are the same) and only your ears can be the judge at the end of the day..... that's why audio equipment and opinions are so diverse, if we all had the same pair of ears and we all heard things the same way then it would be easy but we don't which is why It's so hard to find the sound that suits your ears...... that's why I DIY, I've got a certain amount of control and can tailor the device to suit my ears.... whether what sounds good to me will sound good to you is another matter..........

All the best.

Mike.

 

[12796]

thanks! I assume if you have two amps you can also additionally leave one fully optimised for the lm6171 while you experiment with tweaking for the ad843... at the moment it is working in the environment developed for the lm6171. Teeth ok now?

 

[PinkFloyd]

Quote:

Originally Posted by biovizier thanks! I assume if you have two amps you can also additionally leave one fully optimised for the lm6171 while you experiment with tweaking for the ad843... at the moment it is working in the environment developed for the lm6171. Teeth ok now?

Well no....... amp one will be tailored around the AD-843.... let's call it AD1..... amp 2 will be purely an LM6171 tailored amp...... if I am to try out composite opamps etc. then I'll need to build "amp 3" I'll be ordering the TANTALUM caps and ceramic caps for the AD-843 bypass tomorrow as Rapid have a promotion on at the moment "free delivery if you order online at the weekend during the month of May"...... so, first and foremost I want to try bypassing the 843 with the TANTs and ceramic caps...... if they bring about even 10% improvement I'll be happy and may not even bother building up a comparo WNA........ let's see what happens.

My teeth are great (gums I should say) "touchwood" I ended up going back to my own dentist (pain driven decision) and he whipped it out pretty sharpish and cleanly......... he obviously didn't want bad press / vibes and made sure that I was treated "properly" He was well aware of my phone calls to the dental Hospital and dental council etc.... remainder of tooth was whipped out after 30 minutes or so of gouging and stirring a chisel around in my mouth so a good day was had by all! I dropped 40mG of valium 20 minutes prior to the extraction so I probably wouldn't have flinched if his chisel had slipped and penetrated my esophagus!

Quote:

Originally Posted by biovizier thanks! I assume if you have two amps you can also additionally leave one fully optimised for the lm6171 while you experiment with tweaking for the ad843... at the moment it is working in the environment developed for the lm6171. Teeth ok now?

Well no....... amp one will be tailored around the AD-843.... let's call it AD1..... amp 2 will be purely an LM6171 tailored amp...... if I am to try out composite opamps etc. then I'll need to build "amp 3" I'll be ordering the TANTALUM caps and ceramic caps for the AD-843 bypass tomorrow as Rapid have a promotion on at the moment "free delivery if you order online at the weekend during the month of May"...... so, first and foremost I want to try bypassing the 843 with the TANTs and ceramic caps...... if they bring about even 10% improvement I'll be happy and may not even bother building up a comparo WNA........ let's see what happens.

My teeth are great (gums I should say) "touchwood" I ended up going back to my own dentist (pain driven decision) and he whipped it out pretty sharpish and cleanly......... he obviously didn't want bad press / vibes and made sure that I was treated "properly" He was well aware of my phone calls to the dental Hospital and dental council etc.... remainder of tooth was whipped out after 30 minutes or so of gouging and stirring a chisel around in my mouth so a good day was had by all! I dropped 40mG of valium 20 minutes prior to the extraction so I probably wouldn't have flinched if his chisel had slipped and penetrated my esophagus..... I was medically prepared for the worst case scenario.

The best part of it..... "no charge" (I should bloody well hope not given the circumstances!!!) and the socket "touchwood" is starting to heal up nicely..... a good healthy looking clot has formed and I'm rinsing with warm salt water between beer, fags and eating so I hope It'll heal up ok.

Best part is I'm "Pain Free" after almost 4 months of continual discomfort and about 11 days of "excrutiating agony"..... I can't even begin to tell you how good it is to be able to sleep, eat, concentrate and "live" again.......... Well, I still can't eat in the conventional sense but I can sleep, concentrate and I feel a lot more part of the universe than I did this time last week.

It's a shame I had to be at the business end of the poor UK dental health care problem after 44 years of superb (and extensive) dental treatment but It's been an eye opener (an a jaw opener) for me and if that's the way you get treated privately my heart goes out to all the people that can't register under the NHS and are forced into to pulling their own teeth.

We need Dentists in the UK and we need adequate funding from the Government to make "NHS" treatment an option the "dentist" will profit from.............. The NHS character who had a go at pulling my tooth (in the abscence of my dentist) charged £16 for "3" X-Rays and the extraction.............. private treatment is £9 per X-Ray and £35 for an extraction so that's £62 for the same thing........ quite a difference........... If you are unemployed in the UK the NHS treatment is FREE.

Bad state of affairs but thank God I'm pain free at last Nick

Mike.

 

[1UP]

I have an OPA627 in a Headsave Classic headamp and it sounds quite lovely there, so I guess it's all down to synergy per system, as you say.

FYI - I have a MOS100 power amp on order from David which uses the OPA627 in its pcb - I guess it's a completely different application, though, as he said the LM6171 is much too fast for this purpose.

Then again, the LM6171 is used in the buffer amp, which I guess is not so dissimilar from the headamp. I just think the LM6171 sounds great, myself.

Ah, the burden of choice / analysis-paralysis!

 

[PinkFloyd]

Quote:

Originally Posted by 1UP I have an OPA627 in a Headsave Classic headamp and it sounds quite lovely there, so I guess it's all down to synergy per system, as you say. FYI - I have a MOS100 power amp on order from David which uses the OPA627 in its pcb - I guess it's a completely different application, though, as he said the LM6171 is much too fast for this purpose. Then again, the LM6171 is used in the buffer amp, which I guess is not so dissimilar from the headamp. I just think the LM6171 sounds great, myself. Ah, the burden of choice / analysis-paralysis!

Horses for courses.


Pinkie.

 

[PinkFloyd]

It's like winter here today so I spent and enjoyable hour in front of a heater, armed with a soldering iron, fitting the 10M resistor, trimpot and input caps to have another go at the LM6171 without output caps.

Success!

Trimpots secured to the chassis using sticky backed velcro strips

I adjusted both channels to 0.00mV and left the amp on for one hour. The left channel had floated to 0.09mV and the right channel to 0.08mV. I readjusted both channels back to 0.00mV and they have remained at 0.00mV ever since

Had a brief listen and holy torpedo batman it didn't half sound good! After reading Jose's findings when he removed the 330K resistor I went back outside with the amp and desoldered them and back in for another listen.

Yup, the life certainly has returned to the HD-600 now and there definitely is more space between the instruments, extra clarity and dare I say it.... better defined bass. This could be a result of no output caps and removing the 330K resistor but I think removing the output caps has probably contributed the most... those Nitai caps definitely had their own "signature" (as did the ALCAPS) but without output caps the sound is a lot cleaner as it doesn't have to pass through electrolytics before it reaches your ears... makes sense.

I'm leaving this amp with LM6171 in situ (without output caps) as the sound is pretty much bang on and I don't want to spoil it by messing about with other opamps so the lid's going down on it now. I'll build another one in the next couple of months which will be based around the AD-843 or even a composite AD843 / LM6171 in tandem (that would be really cool

) but, for the time being, I don't want to over egg the pudding and spoil the sound of this amp so will keep my soldering iron away from it

All the best.

Mike.

EDIT: a good way of describing the sound without output caps is "extra sparkle"

At last! No output caps and 0.00mV DC offset

 

[BrokenEnglish]

that's quite encouraging, mike! thanks!
what happens dc-wise if you switch off the amp and turn it on again after ~1h of cooling down?

ah... and you didn't try your 10m/100k without output- and input-caps accidentally?

sunny greatings to you.... it's the hottest day over here (~30°C)...
udo

 

[PinkFloyd]

On the subject of the TLE2462 being used in the rail splitter I asked Dr. White and here's his take on it:


[size=medium]This is very easy to arrange, you simply remove the two 10k resistors in the rail splitter (and possibly the associated 100nF cap too) and replace them with the TLE2426. The TLE2426 In terminal goes to the rail splitter positive line ( +V ), the Common terminal to the rail splitter negative line ( -V ), and the Out terminal goes to what was the junction of the two 10k resistors. The LM6171/BD139/BD140 then acts as a high current buffer to the TLE2426[/size]

And just when I thought it was safe to bolt the lid down

 

[PinkFloyd]

Quote:

Originally Posted by BrokenEnglish that's quite encouraging, mike! thanks! what happens dc-wise if you switch off the amp and turn it on again after ~1h of cooling down? ah... and you didn't try your 10m/100k without output- and input-caps accidentally? sunny greatings to you.... it's the hottest day over here (~30°C)... udo

Hi Udo,

It will not work without input caps.... they must be in position. From cold the offset measures 0.09mV L - 0.8mV R after about 20 minutes or so both channels measure 0.00mV so there's not even "one" millivolt to worry about between hot and cold which is excellent!!

Mike.

EDIT: It's 12C (50F) here today and blowing a gale.

 

[BrokenEnglish]

Quote:

Originally Posted by PinkFloyd It will not work without input caps.... they must be in position. From cold the offset measures 0.09mV L - 0.8mV R after about 20 minutes or so both channels measure 0.00mV so there's not even "one" millivolt to worry about between hot and cold which is excellent!!

fine! thanks!
have you also removed r11, mike?