| navin |
This is a offshoot from my "EL34 Schematic confusion" thread.
http://www.diyaudio.com/forums/show...10&pagenumber=8
so far we have an ECC88 based crossover and 2 x ECC88-6SN7-EL34 based power amps.
Crossover: 2 x ECC88 (Steve Bench ckt)
Line Stage: 4 x ???
Power amp: 1x ECC 88 AF, 2 x 6SN7 Phase Splitter, 4 x EL34.
octal: ECC88 (4), EL34 (8)
nonal: 6SN7 (4)
However since my power amps will be about 10m away from my preamp (no space near the pre amp) but I can hide the power amps behind the speakers (push - push).
what line drive will do. I was looking at the 6SN7 shown below
http://www.tubebuilder.com/images/s...ps/6sn7_sig.gif |
|
|
| anatech |
Hi navin,
You need to get the lowest impedance output with the lowest distortion possible. Balanced lines wouldn't hurt depending on how much electrical noise there is in the area.
You may be further ahead to drive a transformer coupled balanced line. Everything is a trade off. The answer depends on what your biggest problems are in your house. You may not know this until everything is built and you switch it on.
-Chris |
|
|
| navin |
| quote: | Originally posted by anatech
Hi navin,
You need to get the lowest impedance output with the lowest distortion possible. Balanced lines wouldn't hurt depending on how much electrical noise there is in the area. |
balanced lines seem to be a good idea. any idea of a ckt that can do this though? |
|
|
| SY |
Balanced is good, but the ability to drive capacitance is better. A properly designed cathode follower is what you seek.
Hint: 6SN7 is not a particularly great tube for this application. |
|
|
| navin |
| quote: | Originally posted by SY
Balanced is good, but the ability to drive capacitance is better. A properly designed cathode follower is what you seek.
Hint: 6SN7 is not a particularly great tube for this application. |
ok SY, thanks for that info. 6SN7 is out any idea what ckt i should use for this please! |
|
|
| SY |
| Whoa! I take that back. I was thinking that you have 30' cables, not 30 meter. That's about 100 feet. At 100pF/meter as a typical value for shielded cable, that's 3000 pF. You'd need a much bigger, heftier cathode follower than I think is reasonable For a line that long, I'm with Chris, transformers and 600 ohm balanced line. |
|
|
| pinkmouse |
| Still under the influence of Christmas cheer SY? 10m is about 30'... |
|
|
| SY |
OK, so I was wrong when I thought I was mistaken.
Heh, heh. |
|
|
| navin |
| quote: | Originally posted by SY
My Heretical article (http://www.diyaudio.com/forums/show...&threadid=58757) is a reasonable tutorial on CF design. You don't necessarily need the bells, whistles, transformers, and servos I threw in, but a high gm triode and a current source can do wonders. |
i did consider the Ecc81 and 82 but feared their reputation for stridency.
BTW I know you have answered this before but since m sources are all line level high impedance why the input transformers. OPTs in understand as tube impedances are 1000 time higher that of my speaker but the output impedance of my source is on the order of 10K and so is the input impedance of the ECC88. |
|
|
| anatech |
Hi navin,
One of the things about the preamp to amplifier link are the very low levels that are present. This connection is almost as delicate a signal as your phono to preamp link. I vote for balanced, using transformers. I don't like transformers, but if you are going to run the signal that far, treat it as a studio mic signal.
This is the reason I speak out when I hear people placing the amplifiers near the speakers to reduce the speaker cable run. It's the opposite thing you want to do. You don't have a choice, so do it the best you can. Read SY's article and build something to drive the transformer. Use a transformer at the other end too. CMRR is important here (it's the whole point really).
-Chris |
|
|
| SY |
| Input transformers have better CMR than output transformers. And they provide galvanic isolation, which made a very noticeable difference in my system. |
|
|
| serengetiplains |
| SY, I use a TVC and find it beats my well-designed active preamp hands down on such things as low-level ambient cues etc, though it does soften dynamics somewhat. I assume part of the clarity I hear from the TVC derives from ridding the overall chain of components of one source of noise. |
|
|
| SY |
| You should try an active/tranny input type. It has some real advantages over the TVC, mostly with regards to consistancy of input and output impedances with changes in volume setting. |
|
|
| Steve Eddy |
| quote: | Originally posted by SY
You should try an active/tranny input type. It has some real advantages over the TVC, mostly with regards to consistancy of input and output impedances with changes in volume setting. |
Though some would argue that the increasing input impedance and decreasing output impedance as you step down is a plus. In fact, the primary reason for using a TVC.
But in my experience, I've found transformers sound their best when they're working into their ideal load. With TVCs and autoformers, you effectively have x number of transformers, which means that only a few positions can possibly have their ideal load.
A good example of this is the review of the Sonic Euphoria autoformer volume control in the January issue of Stereophile.
Depending on where you have it set and what the load impedance is, it can have as much as a 13dB resonant peak at about 50kHz which will give you some very nasty group delay in the upper portion of the audio band.
Personally I've always felt that the biggest advantage of the TVC was really just the "T" part. And I've always had excellent results using a fixed, 1:1 input transformer coupled to a passive attenuator of the value of the transformer's ideal load.
se |
|
|
| navin |
| quote: | Originally posted by anatech
Hi navin,
One of the things about the preamp to amplifier link are the very low levels that are present. This connection is almost as delicate a signal as your phono to preamp link. I vote for balanced, using transformers. I don't like transformers, but if you are going to run the signal that far, treat it as a studio mic signal. |
so a nice choice would be to hide the the power amps somewhere close to the crossover/line stage.
Q: How long a cable can a transformerless line stage drive? and what tube is normally used for such line stages. 3m?
my present system is SS (live level preamp suing relay switching and a OP275 driving a power amp using 3 pairs (per channel) of 2 SJ50, 2SK 135 (1200VA trannry). It is clean and clinical and can provide goosbumps at times but I am hoping to take things to the next level by using tubes and going active (no passive crossovers). |
|
|
| moamps |
| quote: | Originally posted by navin
octal: ECC88 (4), EL34 (8)
nonal: 6SN7 (4) |
octal:6SN7, EL34
noval: ECC88
| quote: | | Q: How long a cable can a transformerless line stage drive? |
It depends of the preamp, the cable and the amp characteristics.
However, the ground loop is the biggest problem in a long unbalanced connection, IMO.
| quote: | | and what tube is normally used for such line stages. 3m? |
Transformerless? 6C33. ;)
My vote goes to the balanced output with the output transformer.
Regards,
Milan |
|
|
| pinkmouse |
I know it may be anathema to your ideals, but the Burr Brown balanced line driver is a very good chip, and easy to use!:)
I would go for a transformer on the input of the poweramps though, they do seem to work better at rejecting noise and earth problems. |
|
|
| Upupa Epops |
| Do you mean DRV 134, Al ? " Sound " of this chip isn't good... ;) |
|
|
| navin |
| quote: | Originally posted by moamps
octal:6SN7, EL34
noval: ECC88
It depends of the preamp, the cable and the amp characteristics.
However, the ground loop is the biggest problem in a long unbalanced connection, IMO.
Transformerless? 6C33. ;)
My vote goes to the balanced output with the output transformer.
|
yes the ECC88 is noval and the 6Sn7 is octal. sorry.
how long is long? 1.5m? 3m? 5m?
i dont want to add any more transformers than absoltely nesscary. i'd rather find a location for the power amps that is closer.
I was thinking of a 6SN7 line drive. |
|
|
| anatech |
Hi navin,
The position of the amps is dictated by how much space you have (you said not enough) and the closest power outlet(s). The shortest distance between the pre and power is the best solution IMHO. Given that that doesn't work for your layout I think you are stuck with high quality balancing transformers. At least they aren't noisy. Pay attention to loading so you don't have frequency response problems.
-Chris |
|
|
| anatech |
Hi Al,
I have used SSM-214X for different applications. CMRR can be good, but they are a little noisy. Cheaper than good transformers.
It's a trade off.
-Chris |
|
|
| fdegrove |
Hi,
You could consider a White cathode follower using a stack of 12BH7As or ECC99s.
Zout is around the 40 Ohm mark and it can deliver enough juice to drive a 10 m cable with aplomb.
It does like a stiff, well filtered, preferably regulated PS as PSRR of this topology is pretty much non-existant.
Cheers, ;) |
|
|
| serengetiplains |
| Merry Christmas, Frank! Drop me a line some time! |
|
|
| fdegrove |
Hi,
| quote: | | Merry Christmas, Frank! Drop me a line some time! |
And to yourself. Will ctc you asap.;)
Cheers, :cool: |
|
|
| navin |
| quote: | Originally posted by fdegrove
You could consider a White cathode follower using a stack of 12BH7As or ECC99s.
Zout is around the 40 Ohm mark and it can deliver enough juice to drive a 10 m cable with aplomb.
It does like a stiff, well filtered, preferably regulated PS as PSRR of this topology is pretty much non-existant. |
Frank,
I have been looking for a schematic to your 12B4 line stage but could not find it.
anyway thanks for your suggestion. which tube would you prefer the 12BH7 or ECC99? How long a cable can they safely drive? 5m? 10m?. Where would I find a schematic for this line stage?
My intention is to use a input selector and volume pot before Steve bench's ECC88 based crossover. then use this line drive to drive the signal to 2 power amps. I woul dneed 4 tubes. 2 per channel (one for Low pass and one for High Pass). |
|
|
| fdegrove |
Hi,
The part I'm talking about is the last stage of this preamp:
http://www.diyaudio.com/forums/show...2042&highlight=
The ECC99 is pretty similar but I only did a paperbuild with that tube so no real life experience.
The WCF is capable of driving long cable lengths beyond your requirement without losses other than pure resistive ones but I'd still keep resistance and capacitance as low as possible to be on the safe side of caution.
Especially if you suffer from high EMI in your listening environment it may be wise to use shielded cable which of course will up capacitance/m considerably.
It worked fine for me and anyone else that tried it so far with bandwidth still well beyond audio requirements, + 300KHz was no problem over 15m with 100K Zin and 23pF/m.
Cheers, ;) |
|
|
| navin |
| quote: | Originally posted by fdegrove
Hi,
The part I'm talking about is the last stage of this preamp:
http://www.diyaudio.com/forums/show...2042&highlight=
The ECC99 is pretty similar but I only did a paperbuild with that tube so no real life experience.
The WCF is capable of driving long cable lengths beyond your requirement without losses other than pure resistive ones but I'd still keep resistance and capacitance as low as possible to be on the safe side of caution.
Especially if you suffer from high EMI in your listening environment it may be wise to use shielded cable which of course will up capacitance/m considerably.
It worked fine for me and anyone else that tried it so far with bandwidth still well beyond audio requirements, + 300KHz was no problem over 15m with 100K Zin and 23pF/m.
Cheers, ;) |
Frank, is this the schematic for the preamp?
http://www.diyaudio.com/forums/atta...tamp=1051125046
I have to slip this ckt after Steve's crossover
http://members.aol.com/sbench101/Crossover/xover.gif
is there anything to be concerned about? |
|
|
| fdegrove |
Hi,
| quote: | | Frank, is this the schematic for the preamp? |
That's the one.
There have some minor modifications for using it as a standalone line preamp in a separate thread but nothing major.
The B+ used a dedicated valve regulated supply per channel which should be in the same thread....Somewhere.;)
Cheers, :cool: |
|
|
| navin |
| quote: | Originally posted by fdegrove
That's the one.
There have some minor modifications for using it as a standalone line preamp in a separate thread but nothing major.
The B+ used a dedicated valve regulated supply per channel which should be in the same thread....Somewhere.;)
Cheers, :cool: |
That thread is 30 pages long! and i am on dial up in India. being in India means I am in left field. there are few DIYers here and evne fewer that work on tube amps. None in Bombay (Mumbai).
This forum is my sole source of knowledge.
BTW is pin 2 connected to pin 7 in that schematic.
I am thinking of using a SS power supply. BTW I hope these posts are not too much of a bother to you. |
|
|
| fdegrove |
Hi,
Don't worry about the PS then, about 300VDC well filtered should do the trick.
| quote: | | BTW is pin 2 connected to pin 7 in that schematic. |
NO.....
Input is on the top grid, pin #2, feedback is taken to the lower grid from the plate of the top triode (pin#1 to pin# 7) via a 100nF cap.
Cathode Rs are both 680R/1W, output is taken from the node between plate of the lower triode and cathode of the top one.
Plate stopper of the top triode is 1K/2W but can be made lower, not really important unless you suffer from EMI radiation or a flaky tube.
Rg and Rg' are both 1M 1/8W or higher.
The 1M pot is only there because previous stage (RIAA) needs to see a 1M input so that can be changed if you're not using that part.
The input cap is there to protect the wiper of the pot. If you're not using one, you can discard it but make sure thre's no DC coming from the output of whatever you want to hook up to it.
Finally, this stage is a buffer so take a mild loss of gain (0.97-98) into the equation.
Calculated Zout is roughly 47R into a 100K load, bandwidth is about 400kHz IIRC.
The circuit is the one you first posted.
As said it was orphaned from a phono preamp.
That should about cover it. ;) |
|
|
| navin |
| quote: | Originally posted by fdegrove
The circuit is the one you first posted.
As said it was orphaned from a phono preamp.
That should about cover it. ;) |
that ckt says the tube is a 7025. same as 12BH7? I thought it was similar to the ECC83. can i replace it with the ECC99? The 12BH7 is rather impossible to come by in India. |
|
|
| fdegrove |
Hi,
| quote: | | that ckt says the tube is a 7025. same as 12BH7? |
Hell no!:whazzat:
Like I said : The first one you posted. :)
http://www.diyaudio.com/forums/atta...tamp=1051125046
The other one you dug up with the 2 12BH7As can be used as well as the last section is nearly identical but uses a SRPP of a 12BH7A in front of it to provide some gain. But you don't need that first section I suppose.
Cheers, ;) |
|
|
| fdegrove |
Hi,
| quote: | | can i replace it with the ECC99? The 12BH7 is rather impossible to come by in India. |
Missed that part.............Should work just fine.
I think I must have done the calculus for the ECC99 at some point in time and arrived at the same values.
So that's good news.....;)
Ciao, :cool: |
|
|
| navin |
| quote: | Originally posted by fdegrove
http://www.diyaudio.com/forums/atta...tamp=1051125046
The other one you dug up with the 2 12BH7As can be used as well as the last section is nearly identical but uses a SRPP of a 12BH7A in front of it to provide some gain. But you don't need that first section I suppose.
Cheers, ;) |
thanks Frank,
I cannot find a 7025 in India either. So I'd have to stick to the ECC99. so what i should really use is a ckt that resembles the latter half of the 12BH7 ckt mentioned but using a ECC99 instead.
| quote: | Originally posted by fdegrove
I think I must have done the calculus for the ECC99 at some point in time and arrived at the same values.
So that's good news.....;) |
Now this is good news! where do I chop the ckt. Do connect the output of Steve'c ECC88 based crossover to pin 2 of the second 12BH7 (in my case a ECC99)? What do I connect to Pin 1 (2 x 10uf caps, the 10K resistor, 100K resistor and the 1N4007 diode?
I have included links of both ckts for your ready reference below.
http://members.aol.com/sbench101/Crossover/xover.gif
http://www.diyaudio.com/forums/atta...tamp=1085011046 |
|
|
| fdegrove |
Hi,
| quote: | | Do connect the output of Steve'c ECC88 based crossover to pin 2 of the second 12BH7 (in my case a ECC99)? |
Pin #2 is your input so, yes.
| quote: | | What do I connect to Pin 1 (2 x 10uf caps, the 10K resistor, 100K resistor and the 1N4007 diode? |
Pin #1 is your anode so that takes the PS.
Only conect the 1K resistor to pin #1.
The rest is just local PS decoupling (the two MKP caps) and a bleeder resistor (the 100K) and a voltage dropping resistor (10K) which you probably won't need provided your PS voltage is about 300VDC.
Keep in mind that you'll need a buffer per x-over output/amp input so it may all add up to a substantial investment.
Cheers, ;) |
|
|
| navin |
| quote: | Originally posted by fdegrove
Pin #2 is your input so, yes.
Pin #1 is your anode so that takes the PS.
Only conect the 1K resistor to pin #1.
Keep in mind that you'll need a buffer per x-over output/amp input so it may all add up to a substantial investment.
Cheers, ;) |
I know I would need 4 tubes. The bad nes is that the ECC99 are not available either (in India). I have 2 choices.
1. look for an alternate tube. any opinions/suggestions on this?
2. import 12BH7 or ECC99 from Singapore. |
|
|
| fdegrove |
Hi,
| quote: | | 2. import 12BH7 or ECC99 from Singapore. |
Unfortunately..............
Cheers, ;) |
|
|
|
