Reply
 
Thread Tools Display Modes
  #1   Report Post  
John L Stewart John L Stewart is offline
Senior Member
 
Location: Toronto
Posts: 301
Smile Low Cost SET Amp for Newbies

Here is something to get the conversation going again!

If one throws enough money at a project, then most anything is possible. For this project I was looking to go the other way. The objective was to build a single ended triode (SET) amp of reasonable performance & low cost. It was 1999 & I had $100.00 in mind. That way money is left to spend on the really important part of a sound system, the loudspeaker. If the speaker is a mediocre performer then it won’t matter much what amp is used.

This particular project was designed for newbies, those who had recently graduated to the soldering iron & able to ID the handle end of a screwdriver. All you self professed experts out there should keep that in mind when you make suggestions. I hope the attachments are legible. It is difficult to get everything into the limited space on this forum. More attachments will have to come in a later post.

The original article was published in Glass Audio magazine, Vol 12, Number 4, 2000. Cheers to all, John L Stewart

-----------------------------------------------------------------------

An Affordable SE Triode Amplifier

If the vacuum tube camp hopes to continue for some time in the future it will have to win many more converts. This will not happen if the cost of entry is too high. Most of the amplifiers so far published in the present revival are expensive to build. As well they tend to be heavy & include hard to find & otherwise exotic components. Many make use of lethal voltages.

The amplifier described here is affordable to those who might otherwise bypass the hobby. All of the parts used are commercially available new or NOS at reasonable cost. No attempt beyond good workmanship has been made to make the end product look pretty. However, it does deliver excellent performance when connected to a good loudspeaker system.

For clarity each section of the amplifier is shown in a separate diagram. It occupies a 7 x 13 x 2 chassis. The COM buss is connected to the chassis at only one point. Normally this would be close to the front end.

The Power Amplifier

For the project I had originally set a target of $100.00 as the total cost for the power amplifier & its power supply. It was quickly obvious that most of the money would go to iron cored parts (power & output transformers & choke). That meant there would not be much left for tubes & other parts.

A search thru receiving tube manuals revealed some interesting possibilities. There is a large group of power triodes, which include a triode voltage amplifier in the same envelope. These were originally intended as vertical oscillators & amplifiers in TV. Refer to Table One. Any one of these could be used as the voltage & power amp in a low cost amp project.

For this project I selected the 6EA7/6EM7 driving into a Hammond 125E universal output transformer. Refer to Figure One. The power stage provides performance somewhere between what one would expect from a SE 45 or SE 2A3. Useable output after losses in the output transformer exceeds 3 watts in the midband at distortion levels typical of triodes. The final circuit allows operation both with & without feedback. Performance is summarized in Table Two. Note especially the damping factor is well maintained below 100 HZ. This is important when driving complex reactive loads such as a loudspeaker.

The feedback as shown uses all of the gain of the power amp. To run without feedback, disconnect its lead from terminal one of the 125E & return it to COM. For feedback intermediate between these two conditions refer to the detail of Figure Two. As an example, if the feedback resistor RFB was 220 ohms then feedback would be 11db.

Stray capacity in low cost output transformers can cause problems. In the 125E the secondary is wound around one end of the primary. We need to make a connection such that the power amp is not driving the interwinding capacity between primary & secondary.

Take care with the connection of the leads of the output transformer. They are color coded so that the blue lead should be connected to the power triode plate. The brown lead is connected to the high voltage supply. The red center tap lead is not used in this circuit. Connect an 8 ohm loudspeaker to terminals 2 & 6.

I was a bit surprised to find the overall gain of the amplifier to be somewhat less then I had recalled from earlier work. For comparison I have a 3 watt fully integrated amp with phono preamp I had designed & built in 1968. It is still used almost every day with EICO AM & FM Tuners.

The power amp tube lineup is a 6AU6 driving a 6AQ5 SE output. Gain from the 6AU6 grid to the output 8 ohm tap measured 23.5. In a similar test the subject amp had a gain of 6.7. This low gain could cause problems later. But the 6AU6/6AQ5 amp performance compared surprisingly well with the subject amp.

The connection shown at "D" is used to bias the heaters off ground.

The Power Supply

We will need about 270 volts to make the 6EA7 happy. About 35 of these volts will be lost in the cathode bias resistor. Another 8 volts are lost in the primary of the output transformer. Resulting plate current is 45 to 50 ma & plate dissipation is 10 to 11 watts. Refer to Figure 3.

A Hammond 269GX transformer coupled thru the SS rectifiers to a capacitor input filter will do what we need here. The power & output transformers are mounted so that there core laminations do not line up. This will avoid unwanted magnetic coupling between the pair. All the diodes are 1N4007's.

Transformer coupled SE triode amplifiers are adversely effected by ripple in their high voltage supply. Using the LC filter section as shown results in ripple at the supply output of about 10 mv p-p. That will insure a minimum of residual hum at the amplifier output.

S1 is a 3-pole double throw center off switch. This allows a warmup or standby condition. In the warmup/standby position 4.9 volts is applied to the heaters thru the diode ring D3-D6. As well, the high voltage is turned off.

The 100 ohm resistor connected to the power transformer center tap may seem a bit unusual to some. Without it the high voltage would be about 20 volts too high. It also helps limit the RMS currents in the transformer windings. The RMS current, not the average current is what causes heating in transformer windings. That is why some builders of amplifiers & other electronics have problems when using solid state rectifiers with transformers designed for vacuum rectifiers.

Using a scope I measured peak current of 176 ma in the 100 ohm resistor. Compare that to the average current of 58 ma. RMS current was 69 ma. In this example the RMS current is only 19 % higher then the average current. In some designs the RMS current could be three times the average. That is what cooks transformers.

A 10 ohm resistor is connected in series with the #47 pilot light to keep it from being too bright. In the standby mode the pilot becomes dim. C1 & C2 are 0.01 f & are Underwriters Laboratories recognized for use in across the line applications. Mallory part number is UM102M. The fuse is a 3/4 amp slo-blo.

Enhancements

As the design progressed there appeared to be an opportunity to add more. In order to get some idea of how the finished amplifier would sound I had connected a CD player thru a separate set of tone controls. It was obvious gain of the amplifier was low in comparison with others I had built in the past. If feedback was to be applied, then gain would
have to be applied somewhere else.

Line Stage & Tone Controls

These were not part of the original project. However, there was space on the chassis & enough power was available. I used the 6BQ7/6BK7 family for both since they are readily available at reasonable cost as NOS.

The line stage is needed when feedback is used in the power amplifier. See Figure 4. It can be left out of the non-feedback version. As shown it has a gain of 8. That is enough to make up for the low gain of the power amplifier. For those who may want to drive a transmission line its output impedance measured 1300 ohms. Gain of the line stage is set by cathode resistor marked RX. The feedback resistor should not be changed. Under some conditions it forms a dc path with other components which set the operating point of the line stage.

The tone controls are lifted straight from P. J. Baxandall. Refer to Figure 5. They are simple & have always worked well for me. This configuration was used extensively by EICO & others in the 1950's kit business. Back then they had the convenience of a simple circuit module to perform some of the needed functions. We don't. The two controls, three resistors & three capacitors are best assembled with their 3 lead wires on the workbench before mounting in the chassis. See Figure 6.

Further Notes

Plate dissipation in the power section of the 6EA7/6EM7 exceeds the specifications published. As connected it measured about 10.8 watts in this circuit. However,
the spec's given are in reference to application in a vertical amplifiers were large voltages & currents are encountered. A comparison to some other commonly used vertical amplifiers is in order.

The 6K6GT as an audio amplifier has a combined plate & screen dissipation
rating of 11.3 watts. As a vertical amplifier this is reduced to 7 watts.
Similarly the 6W6GT has ratings of 13.25 & 9.2 watts respectively. There are
several other examples in the tube manuals for those who would like to check.

My signal source for this project was an HP 200CD oscillator. Level & distortion
measurements were made by a Pico Technology ADC-100 analog to digital converter connected to the printer port of a 486DX2/66 running in DOS.

file- An Affordable SE Triode Amplifier
Attached Images
    
Attached Files
File Type: doc 6EM7 Amp Tube Selection.doc (9.0 KB, 223 views)
  #2   Report Post  
John L Stewart John L Stewart is offline
Senior Member
 
Location: Toronto
Posts: 301
Smile

Quote:
Originally Posted by John L Stewart View Post
Here is something to get the conversation going again!

If one throws enough money at a project, then most anything is possible. For this project I was looking to go the other way. The objective was to build a single ended triode (SET) amp of reasonable performance & low cost. It was 1999 & I had $100.00 in mind. That way money is left to spend on the really important part of a sound system, the loudspeaker. If the speaker is a mediocre performer then it won’t matter much what amp is used.

This particular project was designed for newbies, those who had recently graduated to the soldering iron & able to ID the handle end of a screwdriver. All you self professed experts out there should keep that in mind when you make suggestions. I hope the attachments are legible. It is difficult to get everything into the limited space on this forum. More attachments will have to come in a later post.

The original article was published in Glass Audio magazine, Vol 12, Number 4, 2000. Cheers to all, John L Stewart

-----------------------------------------------------------------------

An Affordable SE Triode Amplifier

If the vacuum tube camp hopes to continue for some time in the future it will have to win many more converts. This will not happen if the cost of entry is too high. Most of the amplifiers so far published in the present revival are expensive to build. As well they tend to be heavy & include hard to find & otherwise exotic components. Many make use of lethal voltages.

The amplifier described here is affordable to those who might otherwise bypass the hobby. All of the parts used are commercially available new or NOS at reasonable cost. No attempt beyond good workmanship has been made to make the end product look pretty. However, it does deliver excellent performance when connected to a good loudspeaker system.

For clarity each section of the amplifier is shown in a separate diagram. It occupies a 7 x 13 x 2 chassis. The COM buss is connected to the chassis at only one point. Normally this would be close to the front end.

The Power Amplifier

For the project I had originally set a target of $100.00 as the total cost for the power amplifier & its power supply. It was quickly obvious that most of the money would go to iron cored parts (power & output transformers & choke). That meant there would not be much left for tubes & other parts.

A search thru receiving tube manuals revealed some interesting possibilities. There is a large group of power triodes, which include a triode voltage amplifier in the same envelope. These were originally intended as vertical oscillators & amplifiers in TV. Refer to Table One. Any one of these could be used as the voltage & power amp in a low cost amp project.

For this project I selected the 6EA7/6EM7 driving into a Hammond 125E universal output transformer. Refer to Figure One. The power stage provides performance somewhere between what one would expect from a SE 45 or SE 2A3. Useable output after losses in the output transformer exceeds 3 watts in the midband at distortion levels typical of triodes. The final circuit allows operation both with & without feedback. Performance is summarized in Table Two. Note especially the damping factor is well maintained below 100 HZ. This is important when driving complex reactive loads such as a loudspeaker.

The feedback as shown uses all of the gain of the power amp. To run without feedback, disconnect its lead from terminal one of the 125E & return it to COM. For feedback intermediate between these two conditions refer to the detail of Figure Two. As an example, if the feedback resistor RFB was 220 ohms then feedback would be 11db.

Stray capacity in low cost output transformers can cause problems. In the 125E the secondary is wound around one end of the primary. We need to make a connection such that the power amp is not driving the interwinding capacity between primary & secondary.

Take care with the connection of the leads of the output transformer. They are color coded so that the blue lead should be connected to the power triode plate. The brown lead is connected to the high voltage supply. The red center tap lead is not used in this circuit. Connect an 8 ohm loudspeaker to terminals 2 & 6.

I was a bit surprised to find the overall gain of the amplifier to be somewhat less then I had recalled from earlier work. For comparison I have a 3 watt fully integrated amp with phono preamp I had designed & built in 1968. It is still used almost every day with EICO AM & FM Tuners.

The power amp tube lineup is a 6AU6 driving a 6AQ5 SE output. Gain from the 6AU6 grid to the output 8 ohm tap measured 23.5. In a similar test the subject amp had a gain of 6.7. This low gain could cause problems later. But the 6AU6/6AQ5 amp performance compared surprisingly well with the subject amp.

The connection shown at "D" is used to bias the heaters off ground.

The Power Supply

We will need about 270 volts to make the 6EA7 happy. About 35 of these volts will be lost in the cathode bias resistor. Another 8 volts are lost in the primary of the output transformer. Resulting plate current is 45 to 50 ma & plate dissipation is 10 to 11 watts. Refer to Figure 3.

A Hammond 269GX transformer coupled thru the SS rectifiers to a capacitor input filter will do what we need here. The power & output transformers are mounted so that there core laminations do not line up. This will avoid unwanted magnetic coupling between the pair. All the diodes are 1N4007's.

Transformer coupled SE triode amplifiers are adversely effected by ripple in their high voltage supply. Using the LC filter section as shown results in ripple at the supply output of about 10 mv p-p. That will insure a minimum of residual hum at the amplifier output.

S1 is a 3-pole double throw center off switch. This allows a warmup or standby condition. In the warmup/standby position 4.9 volts is applied to the heaters thru the diode ring D3-D6. As well, the high voltage is turned off.

The 100 ohm resistor connected to the power transformer center tap may seem a bit unusual to some. Without it the high voltage would be about 20 volts too high. It also helps limit the RMS currents in the transformer windings. The RMS current, not the average current is what causes heating in transformer windings. That is why some builders of amplifiers & other electronics have problems when using solid state rectifiers with transformers designed for vacuum rectifiers.

Using a scope I measured peak current of 176 ma in the 100 ohm resistor. Compare that to the average current of 58 ma. RMS current was 69 ma. In this example the RMS current is only 19 % higher then the average current. In some designs the RMS current could be three times the average. That is what cooks transformers.

A 10 ohm resistor is connected in series with the #47 pilot light to keep it from being too bright. In the standby mode the pilot becomes dim. C1 & C2 are 0.01 f & are Underwriters Laboratories recognized for use in across the line applications. Mallory part number is UM102M. The fuse is a 3/4 amp slo-blo.

Enhancements

As the design progressed there appeared to be an opportunity to add more. In order to get some idea of how the finished amplifier would sound I had connected a CD player thru a separate set of tone controls. It was obvious gain of the amplifier was low in comparison with others I had built in the past. If feedback was to be applied, then gain would
have to be applied somewhere else.

Line Stage & Tone Controls

These were not part of the original project. However, there was space on the chassis & enough power was available. I used the 6BQ7/6BK7 family for both since they are readily available at reasonable cost as NOS.

The line stage is needed when feedback is used in the power amplifier. See Figure 4. It can be left out of the non-feedback version. As shown it has a gain of 8. That is enough to make up for the low gain of the power amplifier. For those who may want to drive a transmission line its output impedance measured 1300 ohms. Gain of the line stage is set by cathode resistor marked RX. The feedback resistor should not be changed. Under some conditions it forms a dc path with other components which set the operating point of the line stage.

The tone controls are lifted straight from P. J. Baxandall. Refer to Figure 5. They are simple & have always worked well for me. This configuration was used extensively by EICO & others in the 1950's kit business. Back then they had the convenience of a simple circuit module to perform some of the needed functions. We don't. The two controls, three resistors & three capacitors are best assembled with their 3 lead wires on the workbench before mounting in the chassis. See Figure 6.

Further Notes

Plate dissipation in the power section of the 6EA7/6EM7 exceeds the specifications published. As connected it measured about 10.8 watts in this circuit. However,
the spec's given are in reference to application in a vertical amplifiers were large voltages & currents are encountered. A comparison to some other commonly used vertical amplifiers is in order.

The 6K6GT as an audio amplifier has a combined plate & screen dissipation
rating of 11.3 watts. As a vertical amplifier this is reduced to 7 watts.
Similarly the 6W6GT has ratings of 13.25 & 9.2 watts respectively. There are
several other examples in the tube manuals for those who would like to check.

My signal source for this project was an HP 200CD oscillator. Level & distortion
measurements were made by a Pico Technology ADC-100 analog to digital converter connected to the printer port of a 486DX2/66 running in DOS.

file- An Affordable SE Triode Amplifier
For the faithful & others here are the rest of the images, Etc that go with the previous post.

Cheers to all, John L Stewart
Attached Images
   
Attached Files
File Type: doc 6EM7 Amp Distortion.doc (10.0 KB, 190 views)
Reply
Thread Tools
Display Modes

Posting Rules

Smilies are On
[IMG] code is On
HTML code is Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Note to newbies George M. Middius[_4_] Audio Opinions 1 September 8th 09 02:49 AM
Free Home Recording Information For Newbies! [email protected] Pro Audio 4 May 24th 07 03:25 PM
Good resources for newbies.... Phillip Tech 1 October 16th 04 09:45 PM
Studio wiring advice for newbies by newbies Steve Jorgensen Pro Audio 24 June 29th 04 02:29 AM
Studio wiring advice for newbies by newbies Steve Jorgensen Pro Audio 0 June 25th 04 09:23 AM


All times are GMT +1. The time now is 11:42 AM.

Powered by: vBulletin
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2004-2024 AudioBanter.com.
The comments are property of their posters.
 

About Us

"It's about Audio and hi-fi"