Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › A Super-Simple 6DJ8 Headphone Amp
New Posts  All Forums:Forum Nav:

A Super-Simple 6DJ8 Headphone Amp - Page 11

post #151 of 490
Sounds like we'll have a few more joining the party! I still haven't acquired a larger power supply(guests in town til end of next week). I'm thinking of going with 36v(regulating it down to test) since I'm running a different tube and it may want the higher voltage anyways. As long as I don't pass 40v I should be okay. I say that because the 317s are only rated to 40v difference between in and out so with the tube starting cold having a heater voltage of ~1.5v and not wanting to worry about dead regulators I'll play it safe. The fets should be fine up there.
post #152 of 490
Thread Starter 
I picked up a Phillips 6922 from an online merchant, and tried it out; low noise, and pretty decent sound - but it clips badly at 19VDC. Happily, I also acquired a 30VDC 1A power supply that works with little to no noise with this amp, and not only does the amp not clip at this voltage, but it's significantly louder - something like 60db louder. (With the amp's volume control set at 10 o'clock, I have to set Foobar to -12db, and set Windows' master volume to about 75% to get a low enough volume with 40ohm 'phones that I'm not concerned for my hearing. At 19V, I had Foobar set at 0db, and the amp's volume control set to about 2 o'clock...)

It's just as much louder with the RCA 6DJ8, BTW. I'd say if you want to run at low voltage (19-20V), an RCA 6DJ8 is a good choice, and a Phillips JAN 6922 not so much so...

Unfortunately, at 30V the output transistors (well, their heatsinks, anyway) get really, somewhat alarmingly hot... like "hmmn, forced-air cooling might not be entirely out of the question once this thing gets put in an enclosure" hot.

It still sounds ridiculously good, though.
post #153 of 490
Originally Posted by Nemo de Monet View Post
It still sounds ridiculously good, though.
This is always good to read
May I also recommend the GE 6DJ8, I've been listening to a pair for a few weeks now and the mid's are out of this world good.
post #154 of 490
Thread Starter 
I'll keep an eye out for one, but right now I'm trying to get or build a 24V supply that doesn't produce a hum in this amp. The 6922 sounds great at 30V, but the heatsinks were hot enough to burn me (literally - got a little blister on my finger, lol), so I've switched back to the RCA 6DJ8 at 19V. The only real downside to the existing PCB is the limited space for heatsinks; if there were an easy way to bolt the transistors to a really big heatsink, I'd be perfectly happy to stick with 30V, and have an awesome-sounding amp that could double as a coffee-cup warmer...

I've got another PCB on the way, as I've got a second set of transistors thanks to B&D's minimum order policy; I might just run a 6922 in that one at 30V, and attach a pair of VGA heatsink/fan units to the 2SK310s.

I'd be really fairly worried at 36V, to be honest. I'd be afraid it would be a bit like the solid-state amp I was prototyping yesterday: sounded butt-kickingly good for about ninety seconds, then all the magic smoke escaped, doh. (I now believe 2N5486 FETs contain more magic smoke, by volume, than any other electrical component except maybe selenium rectifiers...)
post #155 of 490
Mine gets hot at 33v, but hasn't overheated yet. It also runs at 19v but not with the 6922. My favorite tube so far is the Tung Sol e88cc (I also tried the Philips 6922 but it sounds a little dull, perhaps needs to be burned in?). I also bought two socket savers that I thought would be the best thing to raise the tube so it could be seen better, but they make the amp sound shrill. I will try again after I clean the socket saver contacts.

Here are two pictures of how far I got with the amp, more pictures can be seen at 6dj8 Headphone Amplifier pictures by kansei13 - Photobucket

post #156 of 490
Looks great kansei!

I'm looking in my 1962 RCA interchangability guide and I don't see that RCA made a 6DJ8.

ECC88 crosses to RCA 6J6 directly and is similar to 6BQ7-A.

Just a bit of info:
6J6 - Medium-Mu Twin Triode Heater is 6.3v@300mA(may require resistor changes to keep from exceeding voltage)
6BQ7-A - Medium-Mu Twin Triode, Heater is 6.3v@400mA(may require resistor changes to get the heater to full voltage)

There are lots of these that will work if you want me to go through the list a bit.

Quick question, does anybody know what the tolerances for heater voltage usually is?
post #157 of 490
Thread Starter 
Originally Posted by DKJones96 View Post
I'm looking in my 1962 RCA interchangability guide and I don't see that RCA made a 6DJ8.
They might not have in '62, but they did - in Kentucky - in 1982, when and where mine was made. It seems a lot of the RCA 6DJ8s are actually Mullards, but I seem to have an actual RCA-made one. Go figure.

Quick question, does anybody know what the tolerances are for the heater voltage usually is?
Usually +/-10%, from what I've seen, at least for 6V/12V tubes, though I'm not positive it holds true for tubes with higher-voltage heaters, like the 117L7.
post #158 of 490
This amp is going to be the death of me...

Actually, the tube is. I got a really nice power supply that provides very clean power but apparently not a lot of it. The .6A i need for the tube heater alone is dragging it from 24.1v down below 20. I got it because it's adjustable between +- 22v to 38v but apparently I need to get a straight 24v like the couple i saw there that will give me over 7.5 amps at 24v. Unfortunately, I'm thinking that 24v won't be enough for the tube I've got and it wants 36v.
post #159 of 490
Look around your house and workplace. If you have an inkjet printer you can borrow the power supply to test things out with.

I'm running with a 33v supply from a HP Deskjet, got it for free. I also got two free power supplies for my Starving Student amplifiers. You can also search on eBay, put in "33v power supply", then try 34, 35, 36, etc. There are some supplies listed for very cheap out there. Sometimes you will get lucky on garbage night as well.
post #160 of 490
Nemo de Monet - interesting about the RCAs. All the talk about RCA not making a 6DJ8 - instead creating their own in-house 6FW8 - seems to relate to the 1960s; as here: Tubes Asylum: Re: They MUST Be Phony Mullards ? by Sondek and the next post in that thread. Could the licensing have changed 20 years later?

I've also got some RCA 7DJ8s that are really Mullards and Amperexes; really nice tubes.

I;m intrigued by the idea that at one time, somebody knew all this stuff, and now it's almost impossible to trace back.
post #161 of 490
kansei, thanks for the tip!

I've got clear sound but not at volume and with the 317s heat sinks reading over 180 while not in an enclosure I feel I've hit my input voltage limit without significant modification from the original heater setup.

Here is where I'm at. I swapped out the output caps for 35v 1000uF caps like the original build specifies(tried fixing the following problem but it did nothing). Now, anything above a low listening volume begins clipping. It doesn't matter where you give the extra push(volume knob, player, or adj pots) the sound begins to clip above said volume. Under the thresh hold it sounds great though. So, I swapped tubes, since I have two of these, and it started to not only clip but almost pop in my ear when you got above the good sound and heater voltage was down to 6.08v from the 6.21v I have the first tube at. Needless to say I went back to the first tube again and will be switching out the 15/13/10 resistors for the 13/13/10 to bring the voltage up a bit on the heater.

Right now I'm trying to figure out if the tube or the output stage is the culprit. **ADDED** I've got 6.55v at the heater and it didn't chage anything as far as sound goes(didn't think it would) but did cool the 317s down a bit and make the heater brighter. Something to note. If the player is left at a constant 1/2 volume the volume knob on the amp only adjusts volume up to 1/2 and then stays steady. If the player volume is set to full it will adjust volume and the the last half of the volume adjust seems to just adjust how much it clips.

I could just give up on this tube and build to original specs but where is the fun in that?
post #162 of 490
I don't think your heater voltage is really gonna make a difference, and I haven't used the tube your using(the ones from post #156 I presume) but I don't see why it wouldn't work good in this circuit with 24V. How much current does the adjustable PS put out?
post #163 of 490
Thread Starter 
Have you tried adjusting the trimpot? I really don't completely understand this amp design (doh!), but it seems like the idea is to adjust for "15V" at the 2SK310s, but that's only with 24V - roughly halfway between that and the ~6.3V heater voltage, which seems like a logical place to bias the output (maybe it's just a coincidence?)

If I understand this correctly - which might not be the case, be warned - you might squeeze a little more headroom out of it by adjusting the trimpots until you get, say, ~12V at the 2SK310s, giving your amp a potential +/- 6ish volt output swing. (If your power supply is sagging to 19 or 20V under load, and you have the output FETs trimmed to 15V, I think you're able to swing -9V or so, but only +4 or +5V = clipping.)

I could be really, really wrong, though. Mine has fixed resistors installed, or I'd perform some empirical experiments on my own.
post #164 of 490
******* Edit *********
This is very interesting but you may need a tube tester to get accurate results on anything other than Uh-k. Mine won't measure accurately on anything else.

Here is a very interesting link on how to measure the voltages for these tubes, and what the acceptable voltages are. The table on the top of the page tells you what you should get for different types of measurements. Lower on the page they tell you how to measure on for different things (and on which pins). I'm about to measure my tubes now.

ECC88 List

Like Nemo said, you may be able to tune in your trim pots so these voltages are in the right range (if they weren't to begin with).
post #165 of 490
I think I've found the problem... (on a side note, I'm using a power supply that holds 33v which it turns out is providing more than what the tube needs since it's rated at 18-30v when used as an AF amp)

This isn't a 6N1P, this is in fact the 6N1. There is a HUGE difference in the gain that the two tubes have. The 6DJ8 (or ECC88) has an 'S' factor of 15 mA/V while the 6N1 (or ECC40) only has an 'S' factor of 6.0 mA/V. I know the ECC40 isn't directly pin compatible(8pins vs 9), but it's the closest thing I've found to the 6N1 where info is actually available. It appears to be the ECC40 with a screen to make it pin compatible with the ECC88. I've taken R109 down to 9.5 ohms and R103 to 49.6 and, while it has helped, it isn't spectacular, yet. The pop when you turn the switch on, however, has become annoying(painful). lol

If I can figure out how the solid state half of the amp works I can probably get this tube to work by increasing its sensitivity a bit but I'd rather not.

I've got 33v in and 26.5v at C102.
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › A Super-Simple 6DJ8 Headphone Amp