Jump to content

Let's talk drivers for DHT


captainbeefheart
 Share

Recommended Posts

Let me start off by saying Happy Thanksgiving!! 🦃

 

I would also like to say since joining this fantastic forum that actually has discussion sub-forums for amplifier technologies, and there are intelligent members that are into building and designing equipment I feel right at home. I was very excited to see some even using LTSpice.

 

On to the good stuff. There was a new thread about a 300b amplifier build along with a discussion about the drivers stage. Since these DHT are so linear but require a very large voltage swing to drive to full power the majority of amplifiers you have heard you are actually hearing the driver more than the 300b. Since SET amplifiers are my absolute favorite topology I have spent a lot of time trying many different drivers. This thread is to discuss drivers and front end tubes driving your beloved SET amplifiers. I will be posting many different types of circuits and their performance, if you have a specific tube you are interested in let me know and I will put it into simulation to check the results.

 

On to the fun!!

  • Like 3
Link to comment
Share on other sites

First up is a very popular design, I believe it is popular because of the high distortion, if you are listening to different amplifiers and come upon one that has a different "sound", especially with vocals, jazz, piano, etc.... this circuit will give it a warmth from the addition of the distortion of the 6SN7. The brain notices a difference in sound, a more complex harmonic fullness which sets it apart from the pack. This is good and bad depending on what you listen to and what you like, if you want accuracy this isn't the circuit for you. If you want romantic sound try this circuit.

 

I would limit this driver to an operating point with the 300b that will only need 50-60Vpeak max to drive to full power because over 50v distortion increases to too much for my liking. At 50Vpeak we are already at 7.4% THD!!! It is predominantly second harmonic but higher harmonics are present and will continue to rise with more output.

 

 

jcschem.png

jcsim.png

jcdist.png

  • Like 1
Link to comment
Share on other sites

I think it is poular because it has been on the Internet for 20 + years. I always use a preamp with my tube amps, so I would never use this much gain in the front-end unless I was using a little feedback.

 

However, I would not use this circuit for high gain. I would use another. 

Link to comment
Share on other sites

6 minutes ago, Curious_George said:

I think it is poular because it has been on the Internet for 20 + years. I always use a preamp with my tube amps, so I would never use this much gain in the front-end unless I was using a little feedback.

 

However, I would not use this circuit for high gain. I would use another. 

 

Yes it has been around a long time which is another reason for the high popularity.

 

I agree too much gain unless you want to add negative feedback. 200mV peak was all that was needed to drive it to 50v peak output.

Link to comment
Share on other sites

Since Curious_George made an appearance lets do his SRPP circuit next with the fantastic 6SL7.

 

The 6SL7 is a very linear tube for being a high mu tube as shown by the low distortion. This has less gain so it takes almost 3v peak to drive to 70 volts output which is generally what I like to test for because that is the average 300b bias and hence what is needed for drive signal.

 

This circuit although much less gain is very low distortion, it will put out 70Vpeak with just over 1% THD, impressive compared to the JC circuit indeed. The SRPP is great but suffers poor PSRR so a quiet supply is needed. The distortion does present higher harmonics.

 

Nonetheless a good driver circuit.

 

 

6SL7schem.png

6SL7sim.png

6SL7dist.png

  • Like 1
Link to comment
Share on other sites

Here is the current driver circuit for my 211 amp. I didn't simulate it, I designed and built it. I never put in the time to learn LTSpice, but maybe soon. I've used other "Spice" type programs, but not LT.

 

I'm sure it is fun, but there never seems to be enought time in a day to get all the fun in!

 

 

Analyzer Panel_SRPP_6BX7GT_100pF_560k.jpg

SRPP_6BX7GT_100pF_560k.jpg

6SN7GT SRPP_6BX7GT Driver_100VRMS.jpg

FR 6SN7GT SRPP_6BX7GT Driver_100VRMS.jpg

211_SRPP 6SN7GT_6BX7GT_IT_1 Watt Analyzer Panel.jpg

211_SRPP 6SN7GT_6BX7GT_IT_1 Watt Graph.jpg

  • Like 1
Link to comment
Share on other sites

13 minutes ago, captainbeefheart said:

Since Curious_George made an appearance lets do his SRPP circuit next with the fantastic 6SL7.

 

The 6SL7 is a very linear tube for being a high mu tube as shown by the low distortion. This has less gain so it takes almost 3v peak to drive to 70 volts output which is generally what I like to test for because that is the average 300b bias and hence what is needed for drive signal.

 

This circuit although much less gain is very low distortion, it will put out 70Vpeak with just over 1% THD, impressive compared to the JC circuit indeed. The SRPP is great but suffers poor PSRR so a quiet supply is needed. The distortion does present higher harmonics.

 

Nonetheless a good driver circuit.

 

 

6SL7schem.png

6SL7sim.png

6SL7dist.png

 

If you add the anode resistor to U1, you can tweak the final distortion, but it is negligible. I'm sure you know that. Most people omit that resistor including me.

  • Like 1
Link to comment
Share on other sites

Here is one method I really like. Another single stage design where you can drive from a modern source or preamp, requires 2.5v peak input to reach 70v output. The use of one ECC88/6DJ8/6922 type tubes with both sections in parallel. Plate load is a CCS for high gain and low distortion. Remarkably there is only .6% THD at 70v output, very impressive.

ECC88schem.png

ECC88sim.png

ECC88dist.png

Link to comment
Share on other sites

The next one is the same configuration, plain old common cathode gain stage except just one tube. I chose the high gm pentode, the 6197 because they are only $3 each and plentiful NOS around to purchase. The tube is a true pentode and wired like a triode for this application. In triode mode it is very linear. Again the triode is loaded via a CCS and needs 3v peak input to reach 70v output.

 

There are several high gm pentodes out there dirt cheap that when triode wired make really great DHT drivers. This one has a little over 1% THD at 70v output. Not too shabby for $3

 

 

 

 

6197schem.png

6197sim.png

6197dist.png

Link to comment
Share on other sites

Some history of my 211 driver circuit; I initially built the 211 mono amps back in 2000. At the time, I did not have a lot of disposable money, so I designed a budget driver circuit. All the major components were already adding up and a 45 or 300B to act as drivers were not an option. I wanted to design something original anyway and did not want to just copy a circuit from the Internet. 

 

I have always ahd a fondness for the 6SL7 and it seemed to have enough gain for the purpose of driving the 211 in Class A1 to about 15 watts, which was fine for me at the time. I had plenty of power supply voltage so the 6SL7 (paralleled) was going to have a nice "swing". I direct coupled the 6SL7 to a 6F8G (paralleled also, similar to a 6SN7). This circuit worked great for a long time, until I wanted to see what Class A2 would look like with a different driver. 

 

Since my 211 chassis already had 2 octal socket holes, I pondered different tubes to use so I would not have to add new holes to my chassis. I also wanted to be able to use 6.3VAC filaments since my previous circuit used 6.3VAC too. There was also a current limitation for the 6.3VAC tubes as well. Retro-fitting always involves compromises if you want to preserve something. In this case, my chassis and look of the amp. I did not want it to look like a giant bench experiment. 

 

I also did not want a lot of gain since 2 tubes were going to be cascaded. My tube options were limited, but I finally came up with a circuit that appeared to meet my needs using octal sockets and the right filament voltage/current. 

 

An SRPP 6SN7GT coupled to a 6BX7GT (paralleled) with interstage transformer. This circuit will easily put out 100VRMS with fairly low distortion. 

  • Like 1
Link to comment
Share on other sites

32 minutes ago, Curious_George said:

Here is the current driver circuit for my 211 amp

 

Very nice!! Would you care to share the schematic for our viewing pleasure?

 

32 minutes ago, Curious_George said:

I never put in the time to learn LTSpice, but maybe soon. I've used other "Spice" type programs, but not LT.

 

It is very fun to run simulation for builds. It allows me to try different things very quickly, especially try tubes I do not have but am interested in, if the performance pans out in the simulation I will order some for a build. As for their accuracy I have found so long as you don't go into grid current the distortion etc... is very accurate compared to real world measurements.

Link to comment
Share on other sites

13 minutes ago, Curious_George said:

Some history of my 211 driver circuit; I initially built the 211 mono amps back in 2000. At the time, I did not have a lot of disposable money, so I designed a budget driver circuit. All the major components were already adding up and a 45 or 300B to act as drivers were not an option. I wanted to design something original anyway and did not want to just copy a circuit from the Internet. 

 

I have always ahd a fondness for the 6SL7 and it seemed to have enough gain for the purpose of driving the 211 in Class A1 to about 15 watts, which was fine for me at the time. I had plenty of power supply voltage so the 6SL7 (paralleled) was going to have a nice "swing". I direct coupled the 6SL7 to a 6F8G (paralleled also, similar to a 6SN7). This circuit worked great for a long time, until I wanted to see what Class A2 would look like with a different driver. 

 

Since my 211 chassis already had 2 octal socket holes, I pondered different tubes to use so I would not have to add new holes to my chassis. I also wanted to be able to use 6.3VAC filaments since my previous circuit used 6.3VAC too. There was also a current limitation for the 6.3VAC tubes as well. Retro-fitting always involves compromises if you want to preserve something. In this case, my chassis and look of the amp. I did not want it to look like a giant bench experiment. 

 

I also did not want a lot of gain since 2 tubes were going to be cascaded. My tube options were limited, but I finally came up with a circuit that appeared to meet my needs using octal sockets and the right filament voltage/current. 

 

An SRPP 6SN7GT (paralleled) coupled to a 6BX7GT (paralleled) with interstage transformer. This circuit will easily put out 100VRMS with fairly low distortion. 

 

Nice!! Very similar to what I have in my current amp.

 

Instead of an SRPP I am running a Lundahl push pull to single ended interstage transformer. The push pull input reduces distortion of the driver, has excellent PSRR, and because there is little to no DC current in the interstage and no gap, it has low inter-winding capacitance for good bandwidth.

 

Input is a 6SN7 split load phase inverter feeding a 6N7G push pull to single ended interstage section, works like glue.

Link to comment
Share on other sites

In a 2 stage DHT amp, you can take advantage of the 2 stages by trying to "balance out" the 2 triode stages since they are out of phase. Trying different operating points for each stage can cancel distortion products significantly sometimes. It is a "freebie" and those are rare. 

  • Like 1
Link to comment
Share on other sites

23 minutes ago, Curious_George said:

In a 2 stage DHT amp, you can take advantage of the 2 stages by trying to "balance out" the 2 triode stages since they are out of phase. Trying different operating points for each stage can cancel distortion products significantly sometimes. It is a "freebie" and those are rare. 

 

Yes distortion cancelling between stages is tough to design, mostly found via trial and error but when you do get some cancellation it is a gift well received.

 

This brings up the fact you really need to simulate the entire amplifier to get the whole story with possible distortion cancellation but I felt it would be fun to just discuss drivers themselves and so just focus on their own performance.

 

Your 211 design are pretty much one of my favorites which is the Shishido circuit. Similar to the Ongaku which is also a favorite but the Ongaku uses an cathode follower to drive the output triode.

 

Here is the Shishido amp if you are unfamiliar.

 

 

811A.png

Link to comment
Share on other sites

Yes,

I am familiar with it. The Ongaku was my inspiration when designing my original circuit. Then, as time, money and experience allowed, I designed the 6BX7 driver circuit. 

 

Having 16kohm output transformers with decent bandwidth has been a great asset. They were custom made back in 2000. 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...