luvdunhill Posted October 22, 2010 Report Share Posted October 22, 2010 Doug the amp was a hit. I left it for extended listening. Everyone prefered my orthos to Ari's Grados though Quote Link to comment Share on other sites More sharing options...
TomB Posted October 22, 2010 Report Share Posted October 22, 2010 When I pulled my tubes out to get ready to put the new led's in I noticed that one of the tube pins decided to stay in the socket Any 6J6 tube should work, right? Nothing special to look for when buying a replacement?Yep - or 6J6-A, 6J6-W, or 6J6-WA. There's also 5964 and 5844, but so far with the WE connection and higher bias, I prefer the 6J6 versions. Looking forward to trying the new setup Doug just posted this weekend ... Nice work on the amp, Nate! Marc, that pic of yours looks great, too! Quote Link to comment Share on other sites More sharing options...
Nebby Posted October 22, 2010 Report Share Posted October 22, 2010 Thanks for the info TomB! For what it's worth, the single channel that's working sounds good with the two red led's in series Quote Link to comment Share on other sites More sharing options...
luvdunhill Posted October 22, 2010 Report Share Posted October 22, 2010 Doug/Colin: Search for "No Light District" at DiyA (sorry on phone or I'd link it) Quote Link to comment Share on other sites More sharing options...
Nebby Posted October 22, 2010 Report Share Posted October 22, 2010 Doug/Colin: Search for "No Light District" at DiyA (sorry on phone or I'd link it) "No Light District" P-P amplifier - diyAudio ? Quote Link to comment Share on other sites More sharing options...
luvdunhill Posted October 22, 2010 Report Share Posted October 22, 2010 "No Light District" P-P amplifier - diyAudio ? yes, Doug has probably seen this, but it's an example of some more complicated biasing schemes. Quote Link to comment Share on other sites More sharing options...
Fitz Posted October 22, 2010 Report Share Posted October 22, 2010 Well the power tranny has stopped buzzing like it was and the noise is very faint now even when up close, so maybe it was something with the power after all. I found a pair of transformers I can use as an isolation transformer should it start doing it again though. Quote Link to comment Share on other sites More sharing options...
Fitz Posted October 24, 2010 Report Share Posted October 24, 2010 Okay, so after listening for a while tonight I noticed the transformer started buzzing again. The makeshift isolation transformer made no difference, and if I press on one corner of the transformer, it goes dead silent, whereas if I press on a different corner it gets much louder. Seems like it only starts doing it after it's warmed up a bit, which is why it seemed to go away when I was only doing short listening sessions. The new bias scheme is pretty cool though. I'm being cruel and using the K1000 on it, which sounded kinda hazy and dull with the original LED biasing, but sounds quite lovely now. I'm using some 1.9V green LEDs I had laying around for a bias of 2.6V. Quote Link to comment Share on other sites More sharing options...
dsavitsk Posted October 24, 2010 Author Report Share Posted October 24, 2010 Okay, so after listening for a while tonight I noticed the transformer started buzzing again. The makeshift isolation transformer made no difference, and if I press on one corner of the transformer, it goes dead silent, whereas if I press on a different corner it gets much louder. Seems like it only starts doing it after it's warmed up a bit, which is why it seemed to go away when I was only doing short listening sessions. Sounds like a loose lam. Easiest fix might be to find what is rattling, press it into place, and put some epoxy on it. Or, I can swap you for another one. The new bias scheme is pretty cool though. I'm being cruel and using the K1000 on it, which sounded kinda hazy and dull with the original LED biasing, but sounds quite lovely now. I'm using some 1.9V green LEDs I had laying around for a bias of 2.6V. Quote Link to comment Share on other sites More sharing options...
Fitz Posted October 24, 2010 Report Share Posted October 24, 2010 Sounds like a loose lam. Easiest fix might be to find what is rattling, press it into place, and put some epoxy on it. Or, I can swap you for another one. Hmm.. I'll have to give that a try then. Quote Link to comment Share on other sites More sharing options...
Pars Posted October 25, 2010 Report Share Posted October 25, 2010 (edited) Implemented the new bias scheme and ran some tests yesterday. I used some BC556C's that I had on hand, as well as some green LEDs which measured ~1.82Vdc. Voltage at the emitter is ~2.5Vdc. I was finding that the left channel was very sensitive to the particular LED in place, and not sure why? The first one that was in was showing a much higher threshold level in RMAA than the right channel. I tried swapping the tubes L-R, but no change. Changing out that LED resulted in the channels being much more closely matched. THD is still slightly higher than with the previous bias scheme (Blue LEDs) as well as most of the previous schemes not involving higher CCS current. I haven't tried listening to it yet. EDIT: one question: why would you not want to jumper the transformers to the anode instead of ground? Seems like this would be similar to the original instructions using LED bias? EDIT2: I did not do the extra credit of the cap Edited October 25, 2010 by Pars Quote Link to comment Share on other sites More sharing options...
Nebby Posted October 27, 2010 Report Share Posted October 27, 2010 I really like the new bias method. Quote Link to comment Share on other sites More sharing options...
TomB Posted October 27, 2010 Report Share Posted October 27, 2010 I really like the new bias method. Nebby - just curious - are you talking about Doug's newest scheme with the transistor bias or the WE connection with the higher CCS current? Quote Link to comment Share on other sites More sharing options...
Fitz Posted October 27, 2010 Report Share Posted October 27, 2010 I'd assume he's talking about the newest one, since he knew about it before it was even posted. Quote Link to comment Share on other sites More sharing options...
Nebby Posted October 27, 2010 Report Share Posted October 27, 2010 Nebby - just curious - are you talking about Doug's newest scheme with the transistor bias or the WE connection with the higher CCS current? The new transistor bias scheme, to be clear. Prior to that I've been running the WE connection with the higher CCS connection; I like this new scheme better on both my HD800 and LCD2, haven't given the HF2 a try yet. Quote Link to comment Share on other sites More sharing options...
Pars Posted October 31, 2010 Report Share Posted October 31, 2010 Had a chance to do some more critical listening with my K601s to this the other night. Bias scheme is the new transistor bias. I decreased the CCS to 65 ohms; measured drop is 0.94V, so 14.5mA. I like this biasing the best of all that I have tried; thought the sound was quite natural, drove the phones well. Possibly a touch soft in the bass compared to my dynalo, still sounds slightly darker also, but really pretty close overall. A keeper. Quote Link to comment Share on other sites More sharing options...
dsavitsk Posted November 6, 2010 Author Report Share Posted November 6, 2010 After much sitting around, I think we are about ready to move this project to the next stage. We have implemented several changes to the PCB, but if anyone has any thoughts on additional changes that should be made, please let us know. The changes so far are as follows: - implemented the BJT biasing scheme - added larger CCS heatsinks - moved the heater regulator sink further from the side - added various venting holes - switched to a fixed, LDO, regulator to reduce heat (we'll lower the heater winding voltage, too) - increased board length (!) to 14" to move the OPTs further from the PT to further reduce hum - also increased board width slightly to utilize split construction cases (Context, Heeger, etc.) - switched package for low voltage PS cap to eliminate chance of putting the wrong part in the wrong spot - all jumpers are gone, save for the 120-220 option ... though, we are thinking of making the first batch 120 only due to RoHS issues. We''ll see. - power jack and input jacks have a little more space between them - all panel parts should be spaced correctly for panel mounting - changed part names so no more L/R designation. - added spots for "box" type parafeed caps Quote Link to comment Share on other sites More sharing options...
Pars Posted November 6, 2010 Report Share Posted November 6, 2010 Sounds good Doug. Did you flip the heater rectifiers so they are all lined up the same way (eliminates any chance of tabs shorting to one another)? Post a pick of the board layout if you want any other comments. Quote Link to comment Share on other sites More sharing options...
dsavitsk Posted November 6, 2010 Author Report Share Posted November 6, 2010 Sounds good Doug. Did you flip the heater rectifiers so they are all lined up the same way (eliminates any chance of tabs shorting to one another)? Post a pick of the board layout if you want any other comments. I think I'll switch to a DO-201 package if that's an issue -- 1N5820 layout pic later today. Quote Link to comment Share on other sites More sharing options...
Beefy Posted November 6, 2010 Report Share Posted November 6, 2010 Sounds awesome. Still keen to pick up a production version of this amp if the price is right Quote Link to comment Share on other sites More sharing options...
dsavitsk Posted November 6, 2010 Author Report Share Posted November 6, 2010 Here's the layout -- comments and suggestions appreciated http://www.ecpaudio.com/pdf/Les_6J6.pdf Quote Link to comment Share on other sites More sharing options...
n_maher Posted November 7, 2010 Report Share Posted November 7, 2010 Doug, I like the sound of everything that you changed. The only item on my wish list would be a 2 pcb, 2 chassis version. Quote Link to comment Share on other sites More sharing options...
Pars Posted November 11, 2010 Report Share Posted November 11, 2010 If I wanted to add a power indicator (LED), would it be best to power this from the heater supply, or from B+? I tried it with jumper clips from the heater supply and it worked fine. Kind of hard to tell when this is on or not, at least from a bit away from it. Quote Link to comment Share on other sites More sharing options...
dsavitsk Posted November 11, 2010 Author Report Share Posted November 11, 2010 It doesn't really matter. The PS has a bleeder resistor that you could tie an LED to the end of, but either is fine. But, the amp does already have 6 LEDs and two tubes, all of which glow when it is on ... Quote Link to comment Share on other sites More sharing options...
CYoung234 Posted December 11, 2010 Report Share Posted December 11, 2010 Not much traffic in this thread lately. I wanted to post to let everyone know that I have been listening to this amp for the last several weeks, and am quite happy with the way it sounds. It is a good matchup for my K501's, and really helps solidify the soundstage and low end on these cans. My 501's are the so-called bass heavy version, and with this amp and my source I am having a hard time figuring out why you would want more bass, at least for classical music. Or I was, until I got my pair of Fischer FA-003's back from Chris yesterday. I listened to some rock on them last night (Peter Gabriel Plays Live), and they sound great with the amp as well, and have more punch in the low end, so I guess for some genres more is better. But, at any rate, thanks for a very nice low cost tube amp design - the latest changes made really help the design, at least on my setup. Oh, by the way, Chris did add a blue power LED, and it is useful, because with the top cover on you really cannot see the LEDs inside too well. The tubes do glow, but the LED helps, especially if the room is brightly lit. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.