[francisdemarte]

Quote:

Could be a diode in the wrong direction?

 
I am inclined to agree. Double check the polarities of all your components, something is on backword or there maybe a short/bad soldier joint someplace.

 

[francisdemarte]

 
 
Here's a tip from Stephen one of the ebay sellers of the Lovely Cube board.
Which reminds me. Make sure your input is AC!
 

In these 2 days I saw someone has problem in soldering kit.  I guess the problem may be the wrong wiring of AC input as 15V-15V-0.  It should be 15V-0-15V.  Since I am not the member of headfi forum, would you mind to post this possibility of problem to see whether it helps.
 
Thank you again for your contribution in this forum.
 
Best Regards
 
Stephen
 

 

 

[sobol]

Lm317 was dead. Now all works. I had some problems with a huge offset (about 200mV) on the output, but I've reduced it with a trimpot to about 10mV. Now the most annoying thing is hum. It's quite loud and in my opinion its caused by a pot and resistors (poor chinese quality like all parts - a kit from ebay). In future I'll change nearly whole parts for some good quality parts.
However, the sound is awesome, even with this parts, and its totally worth to buy it.

 

[liamstrain]

Make sure your pot and such are all grounded well - even if you have to run another wire from the casing (if metal). 

 

[sobol]

Everything is grounded well and hum is still there. Nevermind, I'll try to do something with it.
 

 

[BlaBlaBla]

200mV DC offset? There is not chance that properly soldered parts would give you such big offset. Even if transistors are completely different. You did something wrong and you must check connections again. Parts can't be so bad.

The highest offset that I saw once, was with combination of different BC and 2N transistors together - and it was 40 mV!

 

 

[sobol]

I guess different voltages on R3/R4 and R5/R6 are a prove that something is wrong in the circuit. Am I right? These voltages were measured with desoldered lm317 and lm337. Voltages are different also on the inputs of lm's. I'm waiting for rectifier diodes and filter capacitors to change them - though multimeter says that diodes are ok, I'll change them, just for sure.

 

[prairiemystic]

What is typical output stage bias current? I got the Yuanjing Lehmann clone (from eBay) and just started digging into it.
Rails are +/-16.7VDC and 53mA bias current; heatsinks are running 70°C which seems a bit too toasty. I could lower the rails to +/-15VDC or ?

 

[prairiemystic]

Lowering rails from 16.7 to 15VDC, a few mA less bias, Vregs a little warmer but still a finger burner.

 

[headfinn]

Hello Everyone,
 
I just had to create a username to thank everyone (especially BlaBlaBla) who has shared their knowledge in electronics and contributed to this great thread. About a year ago I implemented almost all of the mods mentioned in this thread successfully and it was really interesting project that taught me a LOT about electronics. Prior to this I had practically no electronics experiense beyond school physics (and that was a while ago .
 
To the discussion about the DC-offset I would like to add that I found out the hard way that you should make sure that the BC550 and BC560 transistors are from the same manufacturer and/or have same but opposite "Base−Emitter On Voltage" (for example ONSEMI -0,62V and +0,62V). To my understanding because of the push-pull operating principle, difference here contributes directly to the DC-offset.
 
Also, one interesting thing I would like to mention is that when doing some RMAA measurements with EMU 0404 USB, I found out that all of the above zero gain modes gave worse noise-level results than the zero-gain setting. My own assesment is that by enabling gain you also insert ground connection to the OPA feedback loop and it picks up some noise from there. I think the original BCL could be better in this regard because it uses high quality custom toroidal transformer which has a ground connector and I guess it has some sort of a grounded electrostatic shield between primary and secondary to suppress noise.
 
And thanks again for everyone.

 

[vinodt1347]

Hi,
Please check the link below
http://www.ebay.co.uk/itm/200723854874?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1438.l2649
 
This seems very similar. Thoughts on buying this with a view to change the components. LC is 3 times the cost.

 

[Nixon]

Similar one here as well http://www.ebay.co.uk/itm/YJ-lehmann-headphone-amplifier-BD139-BD140-LM833-/190651055620?pt=UK_AudioTVElectronics_HomeAudioHiFi_Amplifiers&hash=item2c63b03e04

 

[Diseree]

Building quality of Lovery Cube is great, but I can hear slight hum noise anyway when I use low-Z and high sensitivity headphones.
It may be designed for middle-Z or hi-Z headphones.

 

[jimbomak]

Have built a clone recently and am enjoying the sound although my Graham Slee Novo beats it. Will have to change some of the components to see if it will sound better.

 

[nippon]

hi,
i have build a diy kit from ebay and get a bad DC Offset 200mV on the Headphone-jack.
Most of the parts from the Kit i have changed against matched Transistors and so one (caps, elkos...).
But here is my Question:

Behind the AC15V-0V-AC15V i have measured the V and my multimeter said on one side of the Diodes there are +21V and on the other side -17V
The DC Output to Opamp is +14,4V and - 14,5V
and the DC out is about + 14,9V and - 14,7V
The LM337 Output is 14,7V
And the LM317 Output is 14,9V

Maybe the LM337 and LM317 are producing the 200mV DC offset?
Or could it be the "energy section" with the diodes and the +21V and -17V readings?

Greetings from berlin

PS: i'm using a quality 1A 2x15V ring trafo!
 
http://bilder.hifi-forum.de/medium/26674/img-20130131-00549_264128.jpg
 
Edit: Offset was only bad in left channel because of bad BC550C/560C pair. Now everything is fine execpt the hum. Will work on that issue.