• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

27 & 6SN7 Preamp - Crossover

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
This idea has been bouncing around in the back of my head for some time now.

Linkwitz alignment 12db @ 100hz

All the math has my head swimming!

Please check/suggest, etc.

Only issue now is how to elevate the 27's heaters since the followers cathode is at 120V ??

BTW, Input impedance to the amps is 100K and the 8hz HP point is intentional.
 

Attachments

  • 27 preamp.jpg
    27 preamp.jpg
    41.8 KB · Views: 1,332
Administrator
Joined 2004
Paid Member
You'll need separate heater supplies and you should not elevate the filaments too much beyond what is on the cathode - also make sure that the time constant on the filament bias voltage is similar to the supply and heater time constant so that the voltage difference between them is not large. I have had 27 cathode to filament insulation failure at a differential of about 12 - 15V IIRC. One of the recommendations of the time was to use a 2.5Vct filament winding and connect the ct directly to the cathode. This might work with some resistance to isolate the shunt capacitance between the filament transformer and the cathode circuit.

You might want to consider replacing the second 27 with something else, and note that the 56 is not better in this regard. You might try a 6C5 or 6J5 type as a follower.

I use a variety of early dht and idht in my designs and they have some less than endearing shortcomings, but can sound great.
 
Last edited:
Kevin,

The "older" numbered tubes do have some "shortcomings". This is my first real "design" effort using tubes like this and heater/cathode voltage tolerances this small were not on my mind.

I have decided that rather than run the risk of a catastropic failure of my (4) NOS 27's (mesh plate) I will re-design using 6SN7's.

An all Octal Preamp/Crossover sounds like a nice idea to me.

BTW, your avatar is great! See the attached 829B PP "Triode" Williamson I am building in dribs and drabs.
 

Attachments

  • 829B PP Triode  Input-Splitter-Driver.jpg
    829B PP Triode Input-Splitter-Driver.jpg
    140.2 KB · Views: 1,152
  • 829B PP Triode  Output & Power Supply.jpg
    829B PP Triode Output & Power Supply.jpg
    84.3 KB · Views: 1,078
Administrator
Joined 2004
Paid Member
Kevin,

The "older" numbered tubes do have some "shortcomings". This is my first real "design" effort using tubes like this and heater/cathode voltage tolerances this small were not on my mind.

I have decided that rather than run the risk of a catastropic failure of my (4) NOS 27's (mesh plate) I will re-design using 6SN7's.

An all Octal Preamp/Crossover sounds like a nice idea to me.

BTW, your avatar is great! See the attached 829B PP "Triode" Williamson I am building in dribs and drabs.

I've always thought your avatar was pretty cool too..

The 27 mesh plates are very nice sounding, perhaps you could keep the first stage a 27 and make the rest 6SN7.. The 829 looks like a cool project.:D

One of the other issues with these early tubes is that they are incredibly microphonic by modern standards. I haven't had much problem with that, but have had significant issues with electrostatic pick up so I now run 26/27 and their kin in WE style shield cans - kills that problem stone dead, but unfortunately completely hides the tube. It's obvious to me now why WE shielded similar tubes in their theater amps.
 
I've always thought your avatar was pretty cool too..
One of the other issues with these early tubes is that they are incredibly microphonic by modern standards. I haven't had much problem with that, but have had significant issues with electrostatic pick up so I now run 26/27 and their kin in WE style shield cans - kills that problem stone dead, but unfortunately completely hides the tube. It's obvious to me now why WE shielded similar tubes in their theater amps.

Kevin,

Your comment about "microphonics" leads me to something I wanted to try. I think it would be "cool" to place some sort of "glass bell" over the tubes. I am thinking that I could use some sort of glass "hurricane lamp" chimney. If there were large holes in the top plate of the chassis surrounding the socket the glass might "shield" the tube from vibration but still allow cooling and visibility??
 

Attachments

  • glch2-12x7-12.bmp
    59.2 KB · Views: 1,100
  • 462_large_image.jpg
    462_large_image.jpg
    7.5 KB · Views: 663
UPDATE Starting to "Breadboard"

Kevin,

I took your advice and re-designed to include 27 direct coupled to a 6SN7 follower, 150V worked out nice. (141V on the 27's Plates)

Attached is one channel schematic and PS schematic.

Anyone please CHECK and make any suggestions. I am going to start making one of my "Wood and PC board hybrid chassis". I think I will do separate PS since I still have about 20MA's or so to "spare" for an RIAA preamp. (6SL7???)

Anyone have suggestions as to a SIMPLE RIAA preamp? Using an Octal tube?
 

Attachments

  • Preamp-Xover PS.JPG
    Preamp-Xover PS.JPG
    50.7 KB · Views: 369
  • 27-6SN7 pre-xover.JPG
    27-6SN7 pre-xover.JPG
    66 KB · Views: 342
Coldcathode,

The idea on post #1 is also an idea on my mind for long time. Did you finish and have it running yet??? I really would like to hear the result from you.
I have a pair of 18" Goldwood sitting for the open baffle project and would love to have a active crossover to fit into.

Albert:)
 
albertli,

I am starting this now. It might be a week or so until I can have it running to some extent as I need to source the resistors & caps.
I spent a lot of time on the math, etc so I should have little trouble getting it to run, SOUND will be the fun part.

I would be interested in knowing about the goldwood Open Baffle project?

Speakers are my next foray after completing the preamp and xover.
 
Coldcathode

I see that you are using an indirectly heated tube for rectifier, so the voltages come up slowly and you won't have problems with stripping cathodes, etc. Still you may have a look at this article from glass-ware, specially item 6 (about the diode). glass-ware

If the circuit biases as you show, it may be possible to eliminate the 22nF and 2x 1M resistor at the input of the LP section. Another issue regards the 2x 710k resistors used to bias up the lowest triode. For AC they are actually in parallel, so you have a 305k load there. Therefore I would say you need 2x 1420k resistors... but, I would really like to have someone confirm this!

Erik
 
I modded a 6SL7 Phono (Bob Danielak) schematic I found to add a phono section.

I changed some values to better coincide with RIAA time constants (I think I did better not 100% sure) I also am using the output coupling cap and the parallel R of the volume pot and output R to form the 20hz pole.

Does this make sense?
 

Attachments

  • dc_phono.gif
    dc_phono.gif
    7.2 KB · Views: 443
  • 27-6SN7 pre-xover.JPG
    27-6SN7 pre-xover.JPG
    165.5 KB · Views: 673
  • Phono octal.jpg
    Phono octal.jpg
    77.8 KB · Views: 453
Coldcathode

I see that you are using an indirectly heated tube for rectifier, so the voltages come up slowly and you won't have problems with stripping cathodes, etc. Still you may have a look at this article from glass-ware, specially item 6 (about the diode). glass-ware

If the circuit biases as you show, it may be possible to eliminate the 22nF and 2x 1M resistor at the input of the LP section. Another issue regards the 2x 710k resistors used to bias up the lowest triode. For AC they are actually in parallel, so you have a 305k load there. Therefore I would say you need 2x 1420k resistors... but, I would really like to have someone confirm this!

Erik

Erik,

Point 1 - could possible do that but I would have to raise the cathode on the LP follower another 5V's that would put me un-balanced gain wise LP and HP (which I suppose is OK)

Point 2 - Holy Cow I think you are correct! I will go back over the calcs, probably just change the cap size rather than go that big on the R's
 
If indeed they are parallel then I do need to increase the load. But, I am not sure they operate that way with AC?

I could do as you suggested on point 1.
Then add one more section to the HP side.

I will look over the 6SN7 curves to see if I can run it at 140V on the plates with the cathode at 160 or so.

Thanks!
 
Erik,
You are DE BEST!

Easy fix to run the top SN7 at 140VP 2.6ma 160 on the cathode so -5V bias relative to the cathode output of the previous stage.

Thought about the AC and I am pretty sure you are correct, it has two paths throught the 710K so they are in effect parallel. Increased to 1.43M (715 is close enough to 710 for Government work!)

attached is revised schematic,
 

Attachments

  • 27-6SN7 pre-xover.JPG
    27-6SN7 pre-xover.JPG
    154.9 KB · Views: 432
Kevin,

Response was intended to be Linkwitz-Riley. This avoids the "hump" in the summed response @ crossover frequency of 100hz.

I am also revising my Phono section because I needed to "digest" all the information I have gathered relative to AC and R/C reactance. This is a STEEP learning curve once you get beyond "copying" someone elses circuits.
 
Revised the two schematics, including a diode to protect the first 6SN7 from cathode stripping at turn on, but I can't come up with a way to protect the HP follower?

Also, made major changes to design of the Phono section but just now noticed that when in Phono mode the signal will be inverted relative to "normal" because of the odd number of gain stages. Any suggestions?

The only other idea I had was to bypass the Rk on the 2md phono gain stage (thus doubling gain) and recalc the RC values for that pole, then I could bring it's signal in at the junction of the 27 with the SN7 follower?
 

Attachments

  • New Phono.jpg
    New Phono.jpg
    116.1 KB · Views: 368
  • 27-6SN7 pre-xover.JPG
    27-6SN7 pre-xover.JPG
    155.9 KB · Views: 163
Administrator
Joined 2004
Paid Member
Revised the two schematics, including a diode to protect the first 6SN7 from cathode stripping at turn on, but I can't come up with a way to protect the HP follower?

Also, made major changes to design of the Phono section but just now noticed that when in Phono mode the signal will be inverted relative to "normal" because of the odd number of gain stages. Any suggestions?

The only other idea I had was to bypass the Rk on the 2md phono gain stage (thus doubling gain) and recalc the RC values for that pole, then I could bring it's signal in at the junction of the 27 with the SN7 follower?

A couple of comments, wrt to cathode stripping I've never seen it in a 6SN7 under any circumstances. (And as proponent of the type I have used a lot of them in line stages and power amplifiers.) I really don't think the diode is necessary, although the small amount of capacitance it introduces isn't really an issue either. And holding the grid 0.7V positive relative to the cathode means it isn't really preventing unwanted cathode current from flowing anyway. You really don't need to protect anything (including the HF follower) and frankly if you are concerned I would just make sure that the B+ time constant is just long enough to address this concern. This is my preferred approach and can result in relatively benign warm up transients at the output which concerns me more.

The third stage in your phono stage is a non-inverting cathode follower so you only have the two prior gain stage inversions, resulting in an output that is in phase with the input, although somewhat delayed.
 
Last edited:
The third stage in your phono stage is a non-inverting cathode follower so you only have the two prior gain stage inversions, resulting in an output that is in phase with the input, although somewhat delayed.

Kevin,

I have used similar couplings without the diode, I might just leave it out.

I know the phono stage itself is non inverting, but running it into the 27 inversion stage then inverts it the rest of the way to the amps. Since the amps I plan to use will be non-inverting and I would rather not have to invert the speaker leads I think I will increase the gain on the second stage of the phono section and bring the signal in to the grid of the follower. I can use a switch arrangement that "shorts" the input to the 27 and switches in the Phono lineout. I can then just add another 2 section pot to adjust the level of the Phono.

I do not have much experience with Phono Cartridges but my guess is that I can expect about 4-5mV? So overall gain of 500+ should be ok?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.