6SL7 instead of 6N1P in this schematic?

[bradberry00]

Hello,

I am planning a build around this schematic:

http://webpages.charter.net/porkchop/tubeamp/se_kt88-full%20schematic.jpg


I would like to use an octal tube in place of the 6n1p. If I replace the 9pin socket with an 8pin, would i have to do any other changes to the components (values, etc) to run a 6sl7?

thanks much,
-david

 

[SY]

6SL7 is a very different tube. You would probably want to use a higher value plate resistor (150-180k would be a good starting point), then adjust the 1k cathode resistors for a current of 1-1.5mA.

 

[bradberry00]

would a value around 3.3k-4.7k be a good starting point for the cathode?

 

[SY]

would a value around 3.3k-4.7k be a good starting point for the cathode?

Let's make a couple of assumptions, then take a look at the datasheet. You want the plate voltage to be at least 150V, since the tube will have to swing all of the drive voltage to the output stage. We'll also assume you've got a 400V supply to work with. Arbitrarily choose 1.5mA as the current (we can change that, but it's a reasonable starting point). To end up with 150V on the plate, we will have to drop 250V. So the plate resistor will be R = V/I = 250/1.5 = 166k. A nearby easy-to-buy value is 180k, but we'll have to reduce the current slightly (I = V/R = 250/180 = 1.4mA).

OK, 150V on the plate, 1.4mA current. Now we can figure out the cathode resistor. Looking at the plate curves on the datasheet, we can see that it will take something less that -1.5V grid-to-cathode voltage to hit that operating point; call it -1.4V. So we can calculate the cathode resistor as R = V/I = 1.4/1.4 = 1k. So there's your rough starting point. As a check, draw a loadline on the plate curves for a 180k resistor passing through the quiescent operating point and see what the total swing is and how symmetrical it is around the operating point.

As a side note, you really, really want to get, read, and understand the book "Valve Amplifiers" by Morgan Jones. There are a lot of free informational sites on the web of varying degrees of reliability, but Morgan's book is reliable, up-to-date, comprehensive, and easy to understand, It will give you a really good grounding in how circuits and tubes are chosen and designed.

 

[Jebem]

Or you could try a 6SN7 instead, if you can accept less gain (about 15 times with this tube).
The 6SN7 is better suited to drive an output tube, as the anode resistor will be of much lower value than what you have to use if you choose a 6SL7.

 

[bradberry00]

I am going to have to get my hands on that book.

So do you think it would be a waste of my time using the 6sl7? I want to use the 6sl7 primarily for pure tinkering reasons, an attempt to learn more by not just copying the schematic. Plus lets be honest octal tubes just look cool.

If I replace the 47k resistors with 180k resistors and leave the 1k Cathode resistors would I get a good sound? or should I just stick with the 6n1p?

is there a sound difference between these two tubes? good or bad?

thanks much for the help thus far!
-david

 

[bradberry00]

Or you could try a 6SN7 instead, if you can accept less gain (about 15 times with this tube).
The 6SN7 is better suited to drive an output tube, as the anode resistor will be of much lower value than what you have to use if you choose a 6SL7.

would this be a mostly direct swap? what components would need changing? Good sound?

 

[SY]

Tinker away, you'll learn something and probably cook a few cheap parts along the way.

"Good sound" is a very relative term! It's like asking if a girl is hot- it depends on what your tastes are and what you expect.

 

[bradberry00]

haha, i would like to avoid cooking too many things. I am trying to find datasheets to compare the 6sn7 and 6n1p...from what i read the 6n1p was designed to replace the 6sn7. Does this mean is is a smaller version of a similar tube with similar specs??

alternately, is there any 2 kt88 single-ended schematics that run a 6sn7 or 6sl7 that i could build from?

 

[Jebem]

The 6SN7 octal tube is closer to the russian 6N1P than the 6SL7.
The 6N1P has got a amplifier factor of 33, against 20 for the 6SN7.

You could start using the 6SN7 with the indicated schematic anode resistor (47Kohm) but I would use a 3 Watt value, just in case.
You will get the same 4.2mA of anode current as proposed on the original circuit and about 200Volt on the tube anode.

For the cathode resistor, change to a 1.8Kohm (1/2 Watt) to have about 7,5Volt of bias, that corresponds to the 200Volt / 4.2 mA operating point according to the 6SN7 anode curves.

However, make sure you wire the tube with the correct pinout (check the net for the 6N1P and 6SN7 pinouts).

Then, I would recommend some reading, as SY has suggested before, to find out how to design a class A single end circuit using tube plate curves diagrams to start with.

You will get a smaller gain from the 6SN7 (about 17 with the above resistor values), when compared to the original circuit with the 6N1P ( about 30), so this can be a problem depending on the source signal You will use to drive the amp. Check this to calculate the gain Calculating Valve Voltage Gain

Assuming we will need about 15Vpp to drive the KT88 in single end, we need a minimum input signal at the 6SN7 of 1V. So a CD player (output of 2Volt standard) will drive it without problems, as one pre-amplifier will do.

Now, an older system giving just like 150mV will not drive the KT88 to full power, so here the original 6N1P is superior and a better solution.

Or You can try the 6SL7 as you mention initially, but it will give you an amplifier factor of 70 (!) that will be excessive if you use a CD player or a preamp to drive it, as it may distort due to excessive signal (think that the 6SL7 bias is about 1.4V, so anything above that will put the tube well out of linear region.

Have fun!

 

[bradberry00]

Cool! This will primarily be a phono amp with a grado ph-1 phono pre in front....and possibly an ipod from time to time (will the internal amp of the ipod be enough?)

one more question, and this may be crazy talk...to double the gain, could i run a 6sn7 for each channel wired in parallel...my brain is just working, might be more work than its work.

I may just build it with the 6n1p first and build a second later.


thanks so much for you help so far!
-david

 

[Jebem]

Concerning the sound quality... The 6SN7 is one of the most linear tubes ever made.

The sound is very good to my ears, and it can beat the 6N1P when properly implemented.
You may like to try some NOS tubes if you fancy paying for them, instead of the chinese and russian specimens.

That said, the 6N1P is also an excellent tube, more modern, showing very low internal resistance and medium amplifier factor, very good for a driver stage, and usually very well manufactured in the russian motherland..

 

[bradberry00]

while I am on a question asking roll...I am pricing parts for the build and seeing what I have in my parts pile (from building guitar amps) and I found two extra chokes. Looking at the schematic I posted, it calls for a 10H 200mA choke. I would like to use what I have on hand being a pair of 5H 120mA 115 Ohm chokes. would a pair of these run in series equalling 10H work even though the current rating would be lower than called for, is 120mA enough?

thanks
-david

 

[Jebem]

Well, looking into your schematic table for the cathode voltage, the worst case scenario using 470 ohm resistor (560 + 3K in parallel) will give about 77mA per output tube, and about 66mA per tube if running the cathodes at 560ohm.
So, the total output tubes current will be a minimum of 132mA or 154mA depending on the bias you set.
The input tubes will consume a minimum of 8.8mA.

So the total current will be about 162mA minimum on the worst case scenario.

I would use a minimum choke of 250mA to have some headroom on the power supply. A stressed power supply will kill the amplifier ability for great sound.

Also make sure the chokes you select are appropriate for CLC filters (Capacitor input, as stated in the schematics), as it are different (at the iron construction and used air gaps) from the ones we use for LC filters (Choke input). This is just a optimization, but it counts for the final sound.

As a temporary fix, You may use your two 5H chokes, one per channel, but then You need two output capacitors (the one referenced as 100uF in the schema).
This will be a better solution, as You will have independent filtering per channel, despite only using 5H / 100uF filters, because the load current per channel will also be half of the calculated value.

One side note, make sure the power supply caps can stand a minimum of 500Volt, because the maximum unloaded voltage can be as high as 490Volt (350VAC times 1.414).

Cheers.

 

[dickot]

Great thread...8^)

 

[bradberry00]

Nice...now I just need to finish up a project I am currently working on and try to put together all this information!

 

[Jebem]

one more question, and this may be crazy talk...to double the gain, could i run a 6sn7 for each channel wired in parallel...my brain is just working, might be more work than its work.
-david

Unfortunately to have more gain You actually need two stages one after another.

Paralleling the tubes will give You a lower internal resistance and double the anode current of the stage; also theoretically the noise can be lower too.

Cheers,
Jose

 

[bradberry00]

I am nearly ready to order the parts for this build....what should the voltage rating be on the power supply caps? will 500volt be enough for the 100uf cap? I am planning to run this power transformer http://www.classictone.net/40-18029.pdf

thanks,
-david

 

[fp2000]

I'd like to know if there has been some progress and if you already have heard how it sounds so you can provide feedback. I am planning on doing something similar.
For the 400V rating cap I think 500v will work just fine, but if you want extra protection you can always run two 200uF in series with a rating of 350v or maybe higher if you have any available. two 350v will take 700v.
let us know on the progress