Elekit TU-8200 DX Headphone/Speaker Amp Review
Jul 28, 2016 at 10:28 PM Post #526 of 1,441
  I recommend to pick option #1 upgarde parts (upgrade the signal path to AMRS carbon resistor). If you have a  bigger budget later, you can replace the Amtrans AMCO to Mundorf EVO Supreme Silver gold.  and the driver tube (12AU7). 

Thank you for the suggestions, that sounds quite reasonable. Cheers.
 
Jul 28, 2016 at 10:30 PM Post #527 of 1,441
  Sorry to say, but you should probably stay away from tubes... Or go with GE 5814A input tubes @$20 a pair, and GE 6L6GC power tubes @$60 a pair, or play it safe and get modern tubes and know your amp will never live up to it's full potential. It's the truth...

Thanks for the thought, I hear where you are coming from. I would tube roll modestly, but it seems clear to me that it can become a crazy addiction rather quickly and lead to one never being content, just one more tube to try ... I know that is the thrill of the chase, but I would not want to get to deeply into rolling, not on principle, just from a practical stand point.
 
Jul 30, 2016 at 8:19 PM Post #528 of 1,441
Total tube newb question; these tubes are for sale close to where I live, are any compatible/recommended? 
http://www.canuckaudiomart.com/details/649290694-assorted-amplifier-tubes/
 
I dont mind grabbing it and keeping one or two to try and selling the rest.
 
Aug 9, 2016 at 12:10 AM Post #529 of 1,441
  Thanks for the thought, I hear where you are coming from. I would tube roll modestly, but it seems clear to me that it can become a crazy addiction rather quickly and lead to one never being content, just one more tube to try ... I know that is the thrill of the chase, but I would not want to get to deeply into rolling, not on principle, just from a practical stand point.

 
You don't have to go crazy if you don't start chasing the best. I would recommend doing the following four steps:
1. make sure you have upgraded rectifier for driver heaters. The stock rectifier can only power 12AU7 tubes and those are literally half the tubes that you can power with upgraded rectifier.
2. E80CC drivers
3. good NOS 6L6GC- if you can afford it go straight for RCA black plates, if not then grab a GE pair. Or if you have a bit of patience you can go for Cossor 807 with round plates and adapters for them (these tubes are awesome in UL mode). Do not look at any other power tubes than what I mentioned here as it will overload the power transformer with E80CC plugged in.
4. And then stop. Improving it further is not trivial and expensive. Better invest into very good headphones and/or speakers and/or source.
 
I have purchased from these sellers with no problems, here are examples of what you can grab on ebay right now:
http://www.ebay.com/itm/2x-E80CC-6085-VALVO-PHILIPS-audio-tubes-NOS-E-80-CC-/281978489076
 
http://www.ebay.com/itm/GE-6L6GC-USA-Tubes-NOS-NIB-OO-Getter-STRONG-103-103-MATCHED-PAIR-1970s-Nice-/371695022192
 
http://www.ebay.com/itm/RCA-6L6GC-Tubes-NOS-NIB-Holy-Grail-BLACK-PLATE-Same-DK-Codes-1971-Mint-PAIR-/371644700151
 
If you don't like ebay you can reach out to Brent at http://audiotubes.com he can set you up both with E80CC and 6L6GC tubes. Either way there are no duties for importing vacuum tubes to Canada.
 
Insist on buying unused tubes- don't pay less for used ones, it is not worth it.
 
Aug 9, 2016 at 12:13 AM Post #530 of 1,441
  Total tube newb question; these tubes are for sale close to where I live, are any compatible/recommended? 
http://www.canuckaudiomart.com/details/649290694-assorted-amplifier-tubes/
 
I dont mind grabbing it and keeping one or two to try and selling the rest.

 
There are thousands upon thousands of tube types. Why would you buy some random assortment just because somebody said it is on sale? Just buy the tubes that you know where to plug them into.
 
Aug 18, 2016 at 12:54 AM Post #531 of 1,441
  Keep the impressions/findings coming! 
beerchug.gif

 
Since you asked for findings here are some from my today's test. I wanted to see how clipped signal looks on this amp. Test environment:
- TU-8200 amp with Cossor 807 in UL mode driving 8Ohm LS50 speakers
- TU-8500 preamp with Mullard 12AT7 equivalent tubes, buffer mode at 1.15x and amplification mode at 2.8x
- both amp and preamp at max volume setting
- Test signal around couple hundred Hz at 0.07V amplitude
 
This is what happens with preamp in buffer mode (top trace is signal at 0.1V/div and bottom trace is speaker at 5V/div):

 
Switching preamp to amplification mode produces this distortion:
 

 
Here is another test at different frequency, buffer mode:
 

 
And distortion when switching to amplification mode:
 

 
So 0.07V signal at 1.15x gain does not get clipped at max volume but at 2.8x gain there is serious clipping going on. Signal from my DAC can peak out at over 2V amplitude (this track is good for driving my DAC past 2V at low frequencies: https://www.youtube.com/watch?v=yRHwWGO_A48) so it gets attenuated by the volume pot being set at pretty low setting. Good news is that for a desktop speaker even with low sensitivity like LS50 sound pressure gets to uncomfortable and unhealthy levels before clipping starts (beyond 90 dB according to my meter).
 
My conclusion is that these tests confirm the reason why this amp sounds so good with both headphones and desktop speakers- no clipping even for near field listening.
 
Aug 23, 2016 at 3:32 PM Post #532 of 1,441
Nice tests!  Very cool, thanks for sharing them.
 
This is good info to know... even though I'm currently only using the TU-8200 as a headphone amplifier, I plan on using it with my 91db at 1 watt efficient monitors in the future, that is as soon as my living situation changes.  Eventually I may go the single driver route and build some nice cabinets with the Fostex Sigma's or something similar.
 
Let us know what else you find/discover/test and thanks again for sharing!
 
beerchug.gif
 
 
Aug 24, 2016 at 2:58 AM Post #533 of 1,441
  Nice tests!  Very cool, thanks for sharing them.
 
This is good info to know... even though I'm currently only using the TU-8200 as a headphone amplifier, I plan on using it with my 91db at 1 watt efficient monitors in the future, that is as soon as my living situation changes.  Eventually I may go the single driver route and build some nice cabinets with the Fostex Sigma's or something similar.
 
Let us know what else you find/discover/test and thanks again for sharing!
 
beerchug.gif
 


Yes, this amp is really handy for desktop use- you get both headphone option and perfectly capable near field setup with desktop speakers.
 
Will do more testing- planning to look at frequency spectrum and distortions of this amp but need to setup some lab equipment first. I know that swapping tubes changes sound but I want to see if this effect can be measured. Would be cool to have a more justified explanation of why those GECs and Cossors sound so different to modern tubes... Speaking of round plate Cossor 807s- I got myself some white ceramic base versions labeled as ATS-25 and VT60A but have not listened to them yet. My brown base Cossors have dual getters and the white ceramic base version has same kind of getter construction but only on one side. The brown base tubes are from December of 1955 and I think the ceramic base version is from 1940's. Interestingly the seller who sold me those white ceramic bases insisted that they were manufactured by GEC, not Cossor (tubes themselves do not have branding).
 
And after I get this lab setup going I will then proceed to swapping out the power transformer. I don't want to do it until I have distortion profile established so that I can see if there is any negative effect from swapping out genuine Japanese transformer for a Chinese-wound one. And if that works then the next step would be to measure higher power tubes like TT21, EL34, EL12, KT66 and compare them to 807, RCA black plates, etc.
 
Aug 28, 2016 at 4:13 AM Post #534 of 1,441
Did a basic distortion test using HP 3561A analyzer. Don't have a good generator yet so using my Gungnir Multibit with a 24/192 sine signal wav file. Gungnir measured around -76.7 dB THD with no harmonics measurable by this analyzer- that's the baseline. Then set volume control to have 300 mV going into my LS50 speakers and proceeded to compare the amp.
 
1. TU-8500 (Mullard 12AT7) with TU-8200 (E80CC with Cossor 807), measured THD at -55.4 dB. Second harmonic at -55.7 dBc and third at -68.4 dBc. So with no other harmonics visible second harmonic dominates which is supposed to be not too bad for human hearing perception.
 
2. connected DAC signal directly to TU-8200, adjusted signal to 300 mV and tested again, this time it measured -63.6 dB THD with second harmonic at -65.3 dBc and third at -71.1 dBc. So TU-8500 adds some distortion and what's interesting is that second harmonic distortion from TU-8500 is much higher than from the third harmonic. I guess that's why I like listening via this preamp (and I measured that separately- my TU-8500 has large second harmonic distortion and no other measurable harmonics).
 
3. swapped in Russian 6L6 that came with the amp. THD is at -67.2 dB. Second harmonic is at -69.8 dBc and third is at -73 dBc. So less distortion than Cossor tubes, and again the difference in 2nd harmonic is more significant than the difference in 3rd.
 
A quick conclusion from these tests is that me liking this amp's sound is proportional to the magnitude of 2nd harmonic distortion.
 
3. So to verify this hypothesis I swapped in RCA 6L6GC black plates, matched signal to 300 mV and got THD of -61.6 dB, second harmonic at -62.3 dBc and third harmonic at -72.6 dBc.
 
4. This looks like a trend, so I stick in GE 6L6GC to confirm it. Again with matched 300 mV output I got THD of -61.3 dB, second harmonic at -62.1 dBc and third harmonic at -72.2 dBc.
 
The difference between 2nd and 3rd harmonics is:
Cossor 807: 5.8 dB
Russian 6L6: 3.2 dB
RCA black plate: 10.3 dB
GE 6L6GC: 10.1 dB
 
This means that with RCA and GE tubes over 90% of distortion is in second harmonic, and this difference goes much higher as power levels go up. I could not reconcile why Cossor is so far behind RCA and GE so I tested at 50% more power (600 mV) and Cossor caught up- it had spread between second and third of 12 dB,  Russian 6L6 had spread of 3.9 dB, RCA had spread of 13.7 dB, and GE had spread of 14.6 dB.
 
I will need to run this test to higher power levels and preferably with more than one pair of newly manufactured tubes, however if this trend holds it shows that Russian tube is similar to NOS tubes on third harmonic distortion but NOS tubes add dramatically more second harmonic distortion, at least in my TU-8200 amp they do. Or maybe Russian tubes need more power to have second harmonic pick up (which would mean they suck for headphones) or I could simply have a freaky Russian tube (which would again speak to the overall quality of NOS tubes vs new manufacture tubes).
 
Final thought- there were no measurable harmonics beyond 3rd with the amp. The combination of very strong 2nd harmonic and lack of measurable harmonics beyond 3rd explains why the amp sounds so good. I'll need to change instrumentation to actually measure where other harmonics are.
 
Aug 28, 2016 at 10:33 PM Post #535 of 1,441
Did more measurements today- checked 10 tubes in the amp, all with the same filament draw.
As you can see from the included graph of 2nd harmonic Russian tubes plus one of my Cossor 807s aren't distorting enough and two of those have some shape to the line. The other 7 tubes are all hanging together and keeping things in a straight band. This was again measured with LS50 speaker as the load. Notice that 1V output is already loud, and at 4V output things get unhealthy for near field listening. So the interesting question to ponder is whether it is a coincidence that only 12.5% of NOS tubes tested are outside of that band while 100 of new manufacture tubes are out...
 
Another interesting aspect is how tight and straight that band is for those 7 tubes considering they were all auto biased by the amp. That is one phenomenal amp!
 
 

 
Sep 9, 2016 at 9:32 PM Post #536 of 1,441
Hello everyone, My name is sunnee and it's been 26 hrs since I rolled my last tube.
 
 
I got my TU-8200DX at the start of August and for some reason I missed this whole thread.  I read Effusion's great review in research of this amp but some how missed the thread.  Well, just finished going through the whole thing and nice to meet everyone.
 
For reference, I bought this amp to drive headphones.  My primary headphone is the MrSpeakers Ether C.  After that is the Sony CD3000.  I did not intend to use it as a speaker amp but after hearing what it can do I have cleaned out my whole system to only the Elekit.  My whole system now consists of a Questyle QP1R, the Elekit TU-8200DX, Ether C and a Jamo C607 3-way tower.
 
Sep 10, 2016 at 11:37 PM Post #537 of 1,441
Welcome to the group Sunnee. 
My 8200 is my primary headphone amp right now, using HD800. I also have a speedballed Crack and I plugged it in for a little bit the other day. After a few songs I just wanted to switch back to the Elekit. I don't get a lot of headphone time, so I feel like I need to maximize it. I think the biggest difference between the amps is how closed-in the Crack sounds, and not necessarily in a more "intimate" way, more of a constricted way. I'd really like to try an extended audition of Bottlehead's Mainline though. 
 
I've been considering a pair of closed headphones too lately. The Ether C really catches my eye, also the new Fostex TH-610. But I've never tried a pair of Fostex cans, and unfortunately the Ether C is just too much for me to invest in.
 
I feel like I must have exceptional self control in this group! I rolled in a pair of KT88 in triode mode and haven't changed in months! Although now I am looking at the PSVANE silver bottle KT88, anyone have any experiences with them?
 
Sep 11, 2016 at 12:15 AM Post #538 of 1,441
 
 
I feel like I must have exceptional self control in this group! I rolled in a pair of KT88 in triode mode and haven't changed in months! Although now I am looking at the PSVANE silver bottle KT88, anyone have any experiences with them?

 
Hello Evan99,  I am using that very tube right now.  The Psvane KT88 T MkII is the silver bottle.  I am also using the Psvane 12AU7 T MkII.  They sound very clean, detail and spacious right off the bat but a needed warmer mids, and bass oomf.  That all changed after about 8 hours of running.  Now they are my favorite combo.  Everything seems perfect.  I don't think I will be rolling for a long time.
 
These are the tubes I have tried.

Top row:  Stock 6L6, Psvane KT88 T-MkII, Tungsol KT150.
Bottom Row:  RCA clear top, Stock 12AU7, Psvane 12AU7 T-MkII, "NOS" Brimar 12AU7.
 
Sep 11, 2016 at 12:18 AM Post #539 of 1,441
Welcome fella's !!!
 
Geez that KT 150 looks like a beast. You sure it's ok to run in the TU-8200?
 

Users who are viewing this thread

Back
Top