The "Lovely Cube" Headphone Amp (Lehmann Black Cube Linear Clone)
Jan 30, 2011 at 4:29 PM Post #346 of 1,624


Quote:

ringer
 
Not 10000uF 63V!
6800uF 63V is maximum. 10000uF 40V also fits.
 
But 10000uF 63V does NOT!
 
I did not tell you exact dimensions that fit in a case.... 30x40mm is 4700uF 63V, 10000uF 63V is 35x50mm....
 
http://www.partsconnexion.com/capacitor_ele_mundorf_snapin.html
 
Could you recall it, or change it for less? 4700uF is enough even for the most demanding headphones. Usually people say that 1A goes for every 2000uF max. 4700uF is enough for  2.2A on each channel. If you load output with 2A on each channel, you will probably burn LM317/337 before Mlytic falls down. They have sick discharging characteristic.


 
This is strange. In this catalog http://www.mundorf.com/downloads/info-news/english/MUNDORF_Fidelity_Components_catalog.pdf they say that dimensions is 30x40mm.
 
Jan 30, 2011 at 6:06 PM Post #347 of 1,624
There is small problem. Mundorf M-Lytic has, actually, a few M-Lytics. "PI", "SI" and rest.
 
http://www.google.hr/imgres?imgurl=http://img.audiomania.ru/pics/goods/big/mundorf_m-lytic_pi_63_vdc_10000_uf1.jpg&imgrefurl=http://audio.i-mho.ru/category_65000000_page_11/&usg=__7uN-1p4Ts5rDSEQsUdVt2KyxiS8=&h=600&w=600&sz=173&hl=hr&start=0&sig2=qvnFXal6Bs6MAF_YrCE5DQ&zoom=1&tbnid=MAOu5veHvm7WDM:&tbnh=149&tbnw=134&ei=5-lFTaXUOIKxtAaYrdnuDQ&prev=/images%3Fq%3DM-lytic%2BPI%26um%3D1%26hl%3Dhr%26sa%3DG%26biw%3D1280%26bih%3D770%26tbs%3Disch:1&um=1&itbs=1&iact=hc&vpx=125&vpy=69&dur=1556&hovh=225&hovw=225&tx=97&ty=115&oei=5-lFTaXUOIKxtAaYrdnuDQ&esq=1&page=1&ndsp=24&ved=1t:429,r:0,s:0
This one is "PI". 125°C.
 
 
http://www.hificollective.co.uk/components/mcap_mlytic_si.html
This one is "SI". 85°C.
 
Mine has "PI" mark on iself and is rated for 120°C. It is bigger than "SI" with equal capacitance and voltage. (Mine is 30x40mm) Problem is that, stores (usually) do not take care, because price is almost equal. But dimensions are not. So maybe you will get PI (10000uF, 63V is a real beast, and you will have to drill the enclosure to put it in) or you will have new version SI which is 1/3 smaler. And it will fit in, without problem.
 
I bought Mundorf M-lytic in Germany (Banzai Music). They have good and fresh stuff (capacitors) but sometimes are very unprofessional and imprecise. Last month I had ordered BDs 139-16/140-16 for push-pull amp, and they sended me 139-16 and 140-10. They cost is same but I can not put them together in a balanced push-pull. (because different gain)
 
Lesson learned - do not buy from Banzai Music if New Year is close, or in summer holidays. Student volunteers mess everything.
 
Jan 31, 2011 at 2:08 AM Post #348 of 1,624


Quote:
 
What do you think about bipolar Mundorf E-cap 470uF?

 
Hi ringer,
 
Unfortunately I’ve never had the chance to try any of the Mundorf products; someone gave me some of those V-Cap OIMP series (Oil Impregnated Metallized Polypropylene) film capacitors, 4.7uF 250V when I was travelling through LA once.
 
When I got them home, back in NZ, I stuck them in my Clone Beyerdynamic A1 and sat back and listened for a week. About two weeks later I put the EPCOS Polypropylenes, bypassed with original PHILIPS KP460 22nF Polypropylenes, back in and wondered what I had been listening too in the V-Cap. I came away really confused about what I should be getting from those V-Caps: IMO they seemed to cast their own veil over the music...
 
I suspect the V-Caps work really well in Valve gear, especially SEPP Class A Triode and Tetrode based designs, awash with even order harmonics that do sound pleasing to the human ear – so I’m told. The problem I found with valve gear I had was that “everything” sounds sweet and warm…you get sick of it after a while…
 
IMO, caps like the V-Caps assert their own timbre and character on the music, especially when you use them as AC coupling capacitors as I was.
Also, I would be curious to know what the ESR, ESL and other parasitic characteristics are of the V-Cap, including the transfer function for the dielectric – I bet it’s non-linear, especially across the 0 axis.
I found out latter that those V-Caps were about USD$90 each! I was really shocked at the price. IMO I can’t say that they were value for money or I would even consider them again….
 
I gave them away to another friend and I heard he used them for a while in his own amp for a month and then pulled them out and switched back to using the PP Orange Drops bypassing and EPCOS MKP cap; he too was using a solid state amp design. He made similar comments to what I found...
 
Jan 31, 2011 at 2:28 AM Post #349 of 1,624

 
Quote:
Is not 820uF too much?
In the specification for LM317/337, they suggest that output cap is between 100uF and 1000uF. Two 820uF (per channel) are 1600uF.
What about low impedance headphones? Do they need big caps on the output?

Hi BlaBlaBla,
 
Don't get too concerned about overloading the LM317T/LM337T regulators in the amp, provided they are heatsinked adequately, they will go into current limiting mode, on switch-on, (Iinrush to the caps will be limited to about 1,5A to 3,4A (depending on device production spread) for the NSC LM317T, until the capacitors charge up, kind of a "soft start" function.
 
In my amps I installed schottky diodes such as the 1N5819 across the +15V rail to ground and the -15V rail to ground, reverse biased; this measure will ensure that the supply rails are never reverse biased during power-up or if one of those caps fails and shorts; it protects the OPAMP primarily,
 
The NSC LM317T regulator and it's negative counterpart, are pretty well protected thermally and seem to be able to take a real beating - I like them a lot, the only issue is that you need to make sure that you have a minimum load applied of at least 10mA (as per the data sheet) else they will goof you up in your circuit and can be a little unstable...other than that they are great. Also they like being decoupled well; see the Datasheet.
 
On the Lovely Cube, I found bypassing the Adj pin with a 100uF 25V NICHICON PM Electrolytic in parallel with an EPCOS 470nF MKT (such as EPCOS B32529C474J) Stacked Film caps, they sound better IMO than just having the 150nF WIMAs - as per the original Lehmann design - you'll need to install a diode from the Adj Pin to the Vout Pin to avoid blowing them up when you turn off the amp -- see the National Data Sheet there's a lot of good info in there.
This arrangement looks ugly with bits soldered underneath the PCB etc, but it sound's great IMO...
 
 
Jan 31, 2011 at 3:33 AM Post #350 of 1,624

 
Quote:
Well I don't care if these parts are counterfeit products as long as the amp sounds good. I think the slogan "you get what you pay for" fits good in here. You should always consider that the seller wants to make profit, and you should also calculate that you could get fake parts especially if you're buying from China. But I really wonder why the Matrix M-Stage is considered to sound worse in comparison to LC. Does it mean that the M-Stage uses also counterfeit parts ?
Maybe these transistors are just fake labeled if there on par with the STMicro BD139, BD140? Can't imagine that chinese fabs can produce transistors at STMicro quality which is definitaly not the worst in the industry.
All I can say is that as long as the quality of the parts don't deceed a minimum level, sound good and live long I can live with ungenuine transistors. And if I want to upgrade I can still change them^^
Does anybody tried the Infineon BD139, BD140 ? 


Hi Knusperfisch,

 

Fakes are important IMO: I once spent two weeks trying to figure out why a headphone amp that I was designing was unstable and then when I connected low impedance headphones (GRADO SR60s) would blow the output stage and then assert a DC offset on the output terminal and cooked both drivers – this really pissed me off. I spent ages screwing around with these fake devices until someone at work suggested that may be they were fake – that’s when the penny dropped…

 

The amp used the fast and good HITACHI 2SD669A-D and 2SB649A-D, ft=140MHz, low Cob=14pF, and good Beta (Hfe) out to 150mA or so.

During my hair pulling I even substituted inferior devices with the same Ic and Vce and the amp appeared to work fine, when I put the fake 2SD669A and 2SB649As in the circuit, it went crazy oscillating, bouncing around and then BANG when I loaded the output stage - no output stage and bad DC offset on the output.

 

Recently someone I know quite well, was rubbishing how bad the BD139 and BD140 transistors were, he bought them off eBay – I gave him some of the PHILIPS devices and he couldn’t believe the difference he was hearing.

 

So:

1.     Fakes will always goof you up – you’ll spend ages pulling out your hair trying to figure out what’s wrong with your circuit; or it will blow up and burnout your precious headphones!

2.     Fakes mean that you might unfairly criticize the performance of a part, as my friend did about the BD139 and BD140s.

3.     Listening to Fakes might mean your missing out on true performance…

 

Attached is a photo of the fake HITACHI’s I bought from some Chinese dude on eBay – I kept the 2SD669As but smashed the fake 2SB649As with a hammer...it felt good



 

How did I know they were fakes? I got a friend in Tokyo to contact Renesas Electronics; When I emailed them I got ignored – probably because I didn’t write in Japanese!

They came back and said: The markings are wrong, laser etching wasn’t used on this part, the silicon die is approximately 60% the correct size and mounted in the wrong location, and the HITACHI round logo is the wrong dimensions and shape, the packaging epoxy was wrong and the 2SD669A's were made with a green epoxy – gosh! All this from just looking at the part in a photo - it would have been good if they had been interested to test these fakes...

 
Jan 31, 2011 at 6:11 AM Post #351 of 1,624
So, if Lovely Cube is full of fake parts, then other Chinese amps and DACs (eg. Audio-GD, Matrix, Hlly, Musiland etc.) contain them too. Otherwise, the sound quality between LC and them would be very different.
 
I almost ready to buy components for the first step:
usb -> Audio-GD NFB-12 ($200 promo) -> HD650 
 
and then, second step, to buy Lovely Cube:
usb -> Audio-GD NFB-12 ($200) -> LC (~$200) -> HD650
 
After the second step Audio-GD NFB-12 will be used only as a DAC with dual WM8741 inside.
 
Please criticize my choice )
 
Jan 31, 2011 at 8:15 AM Post #352 of 1,624
I have only low impedance headphones. :) (Grado, ATH)  But...
 
When I assemble (solder) something there is always FIRST test. In that test I use 5 USD headphones (IEMs usually) turn my assembled amp on and turn the pot to 12 o'clock. In that state I leave amplifier for 4 hours. After 4 hours, I check DC offset in loaded and unloaded state. I check temperature (with my finger - if I can hold for more than 10s it is below 60°C). After that I switch amplifier off and leave it for about half hour. Then I switch it on and check DC offset again. I also check noises through headphones (IEMs are good for that-very sensitive). It is good when I am able to hear noise that comes from transformer. That noise is equal trough entire range of amplification and it is not something to worry about. Besides, I like it. It reminds me on human nature. (Technical imperfection is equal to nature perfection.)
 
If it blows IEM headphones that cost me 5 USD I do not care.
 
After that test amplifier is ready to take ATH headphones. Then there is first listening that takes 1 hour. After 1 hour I check DC offset again. Every time when I replace capacitor or Opamp I check DC offset. That is the most important thing. DC signal (offset) will kill your headphones, not pure AC noise or AC oscillation. Or transformer hum. Because I tested it (15 hours) and DC offset is always below 3 mV, I have full trust in DIY LC.
 
Be aware of DC offset in LC, because it will depend on your Opamp. (eg. in my setup OPA627 has 2.6mV on left and 0.3mV on right; OPA2107 (biased 3mA) has 2.3mV on each channel, OPA2134 has 4.2mV on left, 0.1mV on right) measured after 2 hours (warm amp).
 
DC offset changes within temperature of your amp. When it is cold it shows (in my setup) 5mV on left and 4mV on right channel. After warming up offfset settles below 3mV.
 
If something goes wrong accidentally (because unknown reason), and headphones blow off, well I can do nothing. :))) In the same way, I can walk down the street and be killed because of careless truck driver ("because I was at the wrong time in wrong place"). :)))
 
 
I write long posts, because I want to improve my writing skills (english language)...
 
Jan 31, 2011 at 8:27 AM Post #353 of 1,624
We share in your predicament of buying fake hitachi items off ebay. Certainly it's been a nightmare. Should I be complaining i paid money for my cds and records while none of them sang live in front of me? To illustrate your experience with us here is far from edifying for the thread, especially no one is hinting for a cause of being cheated with the lovely cube. For some of us here who has placed good trust on the seller over the lovely cube. I doubt there is any complaints given the cost performance ratio. Surely its not engineered to be a BCL replacement or equivalent, it does however allows opamp switching with minimal fuss something the BCL can't. Common sense will dictate the similar overall design accounts for the common traits between the two amps at a rather large price gap. $200+ probably won't buy you an empty lehmann BCL case, let alone the workmanship and the "real" components involved.
 
Jan 31, 2011 at 5:11 PM Post #356 of 1,624
Cool let us know how it'll perform! I'm really curious about it^^
 
 
Quote:
Hi Knusperfisch,

 

Fakes are important IMO: I once spent two weeks trying to figure out why a headphone amp that I was designing was unstable and then when I connected low impedance headphones (GRADO SR60s) would blow the output stage and then assert a DC offset on the output terminal and cooked both drivers – this really pissed me off. I spent ages screwing around with these fake devices until someone at work suggested that may be they were fake – that’s when the penny dropped…

 

The amp used the fast and good HITACHI 2SD669A-D and 2SB649A-D, ft=140MHz, low Cob=14pF, and good Beta (Hfe) out to 150mA or so.

During my hair pulling I even substituted inferior devices with the same Ic and Vce and the amp appeared to work fine, when I put the fake 2SD669A and 2SB649As in the circuit, it went crazy oscillating, bouncing around and then BANG when I loaded the output stage - no output stage and bad DC offset on the output.

 

Recently someone I know quite well, was rubbishing how bad the BD139 and BD140 transistors were, he bought them off eBay – I gave him some of the PHILIPS devices and he couldn’t believe the difference he was hearing.

 

So:

1.     Fakes will always goof you up – you’ll spend ages pulling out your hair trying to figure out what’s wrong with your circuit; or it will blow up and burnout your precious headphones!

2.     Fakes mean that you might unfairly criticize the performance of a part, as my friend did about the BD139 and BD140s.

3.     Listening to Fakes might mean your missing out on true performance…

 

Attached is a photo of the fake HITACHI’s I bought from some Chinese dude on eBay – I kept the 2SD669As but smashed the fake 2SB649As with a hammer...it felt good

Well thats surely a bummer.
Therefore I would never buy such crucial parts from a chinese guy at ebay. Because you will certainly always get fake parts no matter how genuine these may look like. I think that this should be clear to everybody and thats truly a fault by the buyer to some extend. You just can't expect that you will get real quality parts at low price thats naive. (at least not in china)  Thats just the risk you should be aware of if you're buying things over there.
But on the other hand fake parts allow poor guys like me to get a headamp which I can afford. I'm just not able to spend vast amounts of money for a headamp as a student.
Also you throw every fake part in the same pot. There should be mentioned that fake parts also differ from quality. Those in the LC or M-Stage seem to be relatively good if you're considering that Headfonia mentioned the M-Stage playing at a satisfying level even in comparison to the HA-160. Also I've never heard that one of these amps caught fire or so. I'm not defending fake parts it's just my honest opinion.
Well... that the M-Stage killed someones headphones is a real pity.
Fortunately the LC is getting tested before shipping^^
 
Jan 31, 2011 at 6:28 PM Post #357 of 1,624

 
Quote:
Cool let us know how it'll perform! I'm really curious about it^^
 
Well that's surely a bummer.
Therefore I would never buy such crucial parts from a chinese guy at ebay. Because you will certainly always get fake parts no matter how genuine these may look like. I think that this should be clear to everybody and thats truly a fault by the buyer to some extend. You just can't expect that you will get real quality parts at low price thats naive. (at least not in china)  Thats just the risk you should be aware of if you're buying things over there.
But on the other hand fake parts allow poor guys like me to get a headamp which I can afford. I'm just not able to spend vast amounts of money for a headamp as a student.
Also you throw every fake part in the same pot. There should be mentioned that fake parts also differ from quality. Those in the LC or M-Stage seem to be relatively good if you're considering that Headfonia mentioned the M-Stage playing at a satisfying level even in comparison to the HA-160. Also I've never heard that one of these amps caught fire or so. I'm not defending fake parts it's just my honest opinion.
Well... that the M-Stage killed someones headphones is a real pity.
Fortunately the LC is getting tested before shipping^^
 

Whoa! Sorry didn't mean to touch a raw nerve here...But, after a couple of friends re-worked their Lovely Cubes, one of them encouraged me to post this stuff up here..
After all we buy these Kits because they are a lot cheaper than the USD$800 for the real McCoy...but to get fake parts on top of that - I guess you take it or leave it; I figured you could do better, as when I had the chance to listen to these two side by side, the gap was bigger than the Grand Cannon IMO...The genuine Lehmann kicked arse - no wonder Lehmann wasn't on EBay getting the Lovely Cube pulled under a breach of US copyright - which incidentally is probably one of the most draconian pieces of legislation ever devised by industry and passed off as a Government initiative IMO!
 
In the mods we came up with, we started trying to make the Lovely Cube sound like the original, when we figured we'd got there we went out and tried to make it sound better...this is where this post will go...
 
IMO the Lovely Cube PCB is about the best thing, buying it on it's own and putting your own parts into the PCB saved a friend around 50% on buying the finished thing from those guys on EBay selling the completed PBA. (PBA= PCB with parts installed - industry term).
 
Fakes to me are like playing Russian Roulette, yet get lucky sometimes, sometimes you don't - that was my case...I don't see any point in justifying why fakes should be supported IMO, but neither am I telling you what to eat for lunch either...
 
For me, it took me all of 20mins to cut out the fake BD139s and BD140s from the Lovely Cube PCB, unscrew the BJTs from the board clean out the holes with a soldering iron and some solder wick, then install some new Fairchild devices! Switched it back on and was really taken back by the sound. Replacing the BC550 and BC560 with the higher gain BC560C and BC550C really picked things up - it took away a harshness that I would get with some music listening with my T1s.
 
The rest of the MODs were fine tuning...
 
Jan 31, 2011 at 7:02 PM Post #358 of 1,624

 
Quote:
I want to change 4700uF caps (that has dimansions 30x30 mm) on bigger caps 30x40 mm. Is it fit in Lovely Cube body?
 
(Sory for my English. I have a little talking experience.)


You need to be careful about the size of capacitance of the smoothing capacitor you use; this is complicated, as it depends on a number of factors as to whether or not you blow up or stress your rectifier diodes at switch-on, and / or burn out the series resistors inline with the smoothing capacitors...
 
If you use, say 30VA 15-0-15, Toridial, (Lehmann used a 28VA with quite good regulation compared to what you can buy off the shelf), the 3.3 Ohm Carbon resistors will limit the inrush current to the 4700uF filter cap, as well as providing a basic first-order LPF. However, too bigger capacitor, you run the risk of stressing those resistors or even blowing them up.
 
To work this out, you need to know the internal resistance of the transformer's secondary, you can figure this out from the no load and full load output voltage, the difference between these two values times the full-load current will give you the internal resistance of the windings (Ohms law)
 
Then the next thing to factor is the ESR of the smoothing capacitors, this is the only thing that will limit the inrush current into the capacitor when it is fully discharged - this is given in the manufacturer's data sheet, use the lowest value for calculation purposes, as the production tolerances of Electrolytics are pretty big, typically as much as +80%!
 
Then through a series circuit, you can estimate the inrush current seen by the capacitors and the diodes, you can neglect the intrinsic resistance of the diodes, as this will be small.
From the current you calculate, you can then figure out whether the peak inrush current is within the repetitive ratings for the diodes (see the datasheet - there's lot's of useful information, but it's just in the form of numbers), within the I^2t characteristics of any fusing, especially on the secondary side, and within the peak power rating/ Imax of the 3.3 ohm resistors.
 
If you're hell bent on using a certain capacitor, like my friend had this 47.000uF 50V Electrolytics, that he mounted the rectifier and caps on the chassis, with Tag Strips. Being a NAIM Audio owner, and a believer in over-sized transformers, he went out an purchased 300VA 15V-0-15V toroid.
On power up he blew-up the T1A fuses, no surprises I guess, then we he wired out the fuses; switched on again he blew up the 3.3 Ohm resistors that he'd wired from the rectifier diodes, no surprises here I guess when you think about it. So after he replaced the resistors with wire links and switched on, all the rectifier diodes exploded into little bits - that was way cool!
 
He ended up getting much bigger Schottky diodes, which was fine, after wiring it all up, he blew the mains fuse on the transformer. It's not a good idea to go overrating the mains fuse on the primary otherwise you risk all sorts of trouble like blowing the internal thermal fuse of the transformer; it's only good for scrap then, as this fuse is usually buried in the middle of the windings.
So we built a soft-start circuit with a 20 Ohm 25W power resistor and a 250ms switch on delay circuit that shorts the 20 ohm fuse in series with the Active (Live) connection to the transformer - it all works fine now....it was a good learning experience if not a little expensive!
 
Jan 31, 2011 at 7:09 PM Post #359 of 1,624
Well alright you convinced me..
smile.gif

50%? Dang...
 
Well see if I'll tweak my device. I never soldered anything so I guess I'm a total klutz at it.
 
Jan 31, 2011 at 7:19 PM Post #360 of 1,624
It's good to have a go...it's only a $90.00 board compared to trying to mod a $800 genuine article. Someone asked me to MOD their Lehmann after they heard the moded Lovely Cube belonging to a mutual friend. I was seriously sweating when I was desoldering the parts on the board!! In NZ money, that things cost about twice what it's worth, in relative terms, in the US!!
 
Only advice is to cut the legs of the transistors with a good pair of side-cutters, don't try and desolder them unless you can get hold of a machine; else you risk damaging the holes and the tracks. Use Solder Wick to soak up the solder out of the holes...don't force anything...
 
Good Luck!
 

Users who are viewing this thread

Back
Top