Bob Carver/ Stewart Hegeman

nakdoc

Chief Journeyman
Joined
May 11, 2011
Messages
619
Location
Nashville, TN Music City
Tagline
highly biased
#2
I would guess this article explains the origins of the Silver 7 tube version. Do the tubes really last 15 years? If the DC clamp idea makes output tubes last this long it would be considered a necessity nowadays.
 

Wheel-right

Chief Journeyman
Joined
Jun 24, 2019
Messages
781
Location
SoCal
Tagline
Man, Ya shoulda been here yesterday
#3
I would guess this article explains the origins of the Silver 7 tube version. Do the tubes really last 15 years? If the DC clamp idea makes output tubes last this long it would be considered a necessity nowadays.
No question Stu Hegeman knew his stuff, if only I understood half of what was said in that article.
 

orange

Veteran and General Yakker
Joined
Jul 6, 2010
Messages
17,704
Tagline
Broken beyond repair but highly affable
#4
This sounds like why PhD's and doctors handwriting is progressively worse and why the folks at Rite Aid are confused.
 

J!m

Veteran and General Yakker
Joined
Dec 24, 2019
Messages
9,255
Location
Connecticut
Tagline
BOT
#5
So, Wheel-Right, I shared your link with Mike Urban over at the Hi-Fi museum, and he shared back with me a bunch of stuff he collected from Bob Carver's eBay auctions over the years. Some interesting stuff many probably have seen before, but here it is anyway:



Hi 'blood, I really like your handle! Anyway, the DC restorer circuit works to keep the DC value of the drive signal at exactly the correct value over the entire input signal. In all auto-bias circuits, the Achilles heel is that the grid bias increase substantially when the output stage is driven hard. For example, suppose the grid bias is -12 volts at no signal. When driven close to its maximum output, the bias will change several volts, becoming more negative by roughly four to six volts. This drives the tubes closer to cut-off and causes distortion during high output demands. In this DC restorer circuit, a 12 volt zener is connected to the cathode bias resistor and fed back to the grids such that the grid bias is held at a constant 12 volts, thereby restoring the lost DC component of the drive signal. This is regardless of the cathode current that causes a voltage drop across that cathode resistor! 'viola! No crossover distortion. I'm certain that amplifier designers of the past would have done the same thing, except zener diodes did not exist then (when they were alive and designing auto-bias amplifiers). Great question, Thanks for asking, Bob Carver

Q: Hello Bob, I just finished rebuilding a Citation II as well, and I would also be interested in any document you care to share regarding the updates you made. I’m especially interested in running the KT88’s at a lower current using the “DC restorer” Thanks, Scott Jun-27-11

A: Hi Scott, I would be happy to share the circuit with you. I intend to post it on the Carverfest website in a week or so. In the meantime, you can install the DC restorer by using a pair of 4148 diodes connected anode-to-anode, the cathode of each one connected to the grid of each output tube. Connect the common anode connection to to the lowest (most negative) bias voltage(-55 VDC)through a 91K, 1/2 watt resistor. In other words, one end of the resistor goes to -55 volts, the other goes to the common anode connection of the 4148's. Set the idle current for 26 ma on each output tube. To obtain enough range with the existing controls, you will need to remove R30 and R69 (Sam's schematic), the optional 18K resistor shown on the schematic. Some factory-built units already have it removed, so check your unit to see how it came from the factory so long ago. A vacuum tube 6AL5 tube will work slightly better than the 4148, as it's smooth turnon characteristic does not induce the small non-linearities that the 4148 does. Not a big deal either way. Hope this helps, Bob Carver

The restorer is simple yet ingenious. It uses a 6AL5 dual diode. Each diode section services one bank of power tubes. The 6AL5 cathodes are connected to the control grids of the push-pull power tubes which are held at a nominal fixed bias of -47VDC. The diode plates are at an acquisition threshold of -56VDC. Bob admits that some aspects of the DC restorer operation are somewhat mysterious to him, as they are to me as well. My take is that the circuit aids significantly in recovery from hard cutoff conditions. Under those conditions the KT88 grids act as rectifiers and shift the effective DC level below -56V, to the point of causing the 6AL5 to conduct momentarily and pull the DC bias back to its nominal value. Bob estimates the lifetime of the 6AL5 as about 50 years. "I know that seems wrong, but 6AL5's are ubiquitous in tuners, and 50-year old tuners always have these tubes and they still check as new. There are so many of them in this world that a guy on eBay sells a string of them for three cents (each tube) to be used as Christmas tree lights."

I have the DC restorer (clamp) in all my tube amps now....I only use SS diodes..
Bob has said the 6AL5 tube...Is a little sweeter sounding but the SS diodes work fine..
One thing I do diff from Bob setup... is I put the diodes at the grids of the output tubes ...not at the front of 1k res. that feed the grids....this sounds better to me....

But there more info on the Craver site on the DC restorer (clamp)
on the Audioasylum. site theres h/k Citation 2 schematic that shows the MOD

Q: Hi, as a matter of interest, because the output transformers are so tiny, how is the bass performance of this amp? In my experience, with such small transformer cores (not hi-fi at all, such toy trafos were/are used in guitar amps where there is nothing below 70Hz!)you'd be lucky if the -3db frequency is at 30-40 Hz, even at 1 Watt, let alone at the full power! Cheers, Igor

A: Igor, Great question, and glad you asked. It's a funny thing about output transformers. Their high frequency response is essentially independent of their physical size. The low frequency response can also be independent of its physical size, provided the primary has enough wire turns on it to raise the magnetizing inductance high enough to avoid overloading the tubes at low frequencies. Unfortunately, by winding the primary for good low frequency performance, we lose the high frequency performance. The only way around this problem is to make the transformer's physical size much larger. As you can see from our measurements below, the high frequency response is phenomenal, and even the low frequency response is excellent. How can that be! What's the catch? There is none - almost. This is a modestly powered amplifier after all. ALL designs represent compromises, and with this small transformer its ultimate low frequency output at high power is not as great as it would be if it were simply larger. Despite the small transformers, this unassuming amplifier has phenomenal frequency response; that's why we like it so much. It’s really good! Its measured frequency response is flat over the entire audio bandwidth, with the -3dB down points measuring 16 Hz and 58 kHz. In my book, this is superb performance from an amp this size, or any amplifier for that matter. Having said all this, I have listened to it with and without a subwoofer. Obviously the sub with its unlimited low end gives this amp everything, and eliminates the need for it to ever go below 80-90 Hz. My personal setup incorporates this amp with NO subwoofer, and within its power limits I find it completely satisfying, with all the lows I could rationally want along with a sumptuous high end. There are two primary measures of an amplifier's response. One is called the power bandwidth which measures its frequency response at full output power, the other is simply called its frequency reponse, which measures the output at small levels, typically around one watt. We listen to the frequency reponse in our listening rooms with music, and we measure the power bandwidth on the laboratory bench. Both specifications are important, each describing a fundamental aspect of the amplifier. This amplifier has a flat frequency response extending from 16 Hz to 58kHz, and its power bandwidth is 26 Hz to 43 kHz. I hope my answer didn’t put you to sleep, and thanks for writing and for a great question. Bob Carver


A: Hi zum, The paint, polish, rubber feet and phono jacks would be what I
would call restoration. The new design of the power supply to increase
the energy storage and eliminate low frequency "gulp" distortion and
bounce, the DC restorer, the increased gain structure for the front end,
the new values of negative feedback (which now include current feedback
as well as voltage feedback),the new cathode bias scheme which reduces
distortion (this is in addition to the DC restorer), the new time
constants associated with the grid drive to the output tubes, again to
eliminate low frequency bounce and extend the low frequency response,
the reduction of the power supply impedance - I could go on and on. All
these changes fall into the category of redesign. Each change has a
specific reason, and is designed to help make the sound great sound. It
tickles my fancy that your M-500 is still in your system after all these
years! Thanks for the accolades and even more thanks for writing, Bob
Carver

Q: hi, mr Carver. in your previous offerings of more powerful designs you used 6al5 tube in
DC restorer. now it 12v Zener. any reason to change approach/SS vs vacuum
tube/?any difference in sound?if i understood your 6al5 use correctly, it
actually works as "vacuum" zener-when it conducts it stay at ~20-30
volts range. not as steep as SS zener, but that may be even better. works
similar to compressor use. will appreciate your answer. thank you for your
everlasting engineering desire. with all respect, Arkady.

A: Hi T-53, The 6AL5 is best used for fixed bias amplifiers such as my big
high power amps. On the other hand, the zener diode is is best for an
auto-biased amplifier
because it is the increased voltage drop across
the cathode resistor during hi-drive that generates the need for DC
restoration. As long as the shift is smooth and has no induced
non-linearities, the sound will be the same. I carefully made certain of
that when I designed the DC restorer for this amplifier, as you can see
in the 'scope photos; the truth is in the pudding. Thanks for writing
and especially for asking such an intriguing question, Warmest and best,
Bob Carver

Q: the Achilles heel is that the grid bias increase substantially when the
output stage is driven hard. For example, suppose the grid bias is -12
volts at no signal. When driven close to its maximum output, the bias
will change several volts, becoming more negative by roughly four to six
volts. This drives the tubes closer to cut-off and causes distortion
during high output demands. In this DC restorer circuit, a 12 volt zener
is connected to the cathode bias resistor and fed back to the grids
such that the grid bias is held at a constant 12 volts, thereby
restoring the lost DC component of the drive signal. Wow Bob¡¡ Roberto
again, even with my poor technical knowledge I understud how the DC
restorer work.. great solution thanks for sharing it. Will it work on
single ended to ?? Best regards Roberto

A: Hi Roberto, It would work, but it's not necessary as a single ended output
stage already has its DC center voltage well established, and even if
it does shift off center, crossover distortion cannot occur because it
is intrinsically free of crossover distortion to begin with. The reason
our hi-fidelity business took off in the early fifties was because Frank
McIntosh (of McIntosh fame)invented a way to make class AB circuits not
have any notch distortion, and low crossover distortion followed
rapidly in its footsteps. That allowed a power increase from about 6
watts to 75 watts, a full order of magnitude. And the hi-fi industry
took off! Thanks Roberto! Bob Carver


Q: How can the tubes last for 50 years when most manufacturers recommend changing the tubes in their amps in 3 or so years? Dave

A: Hi Dave, That is a very good question! The answer is the DC restorer.
Here's how. Most amps idle the output tubes at an average of 32 watts or
so. Now the DC restorer allows the tubes to idle at about 9.75 watts.
Since tube longevity is roughly proportional to the plate dissipation
raised to the 2.3 power, we have 32 divided by 9.75 raised to the power
of 2.3 = 15.4. Finally, 15.4 X 3 years = 46 years. I have also noticed
that many console amplifiers have 50 year old tubes right here on eBay,
and those tubes check out great. Great question, hope this helps, Bob
Carver
 

Wheel-right

Chief Journeyman
Joined
Jun 24, 2019
Messages
781
Location
SoCal
Tagline
Man, Ya shoulda been here yesterday
#6
So, Wheel-Right, I shared your link with Mike Urban over at the Hi-Fi museum, and he shared back with me a bunch of stuff he collected from Bob Carver's eBay auctions over the years. Some interesting stuff many probably have seen before, but here it is anyway:



Hi 'blood, I really like your handle! Anyway, the DC restorer circuit works to keep the DC value of the drive signal at exactly the correct value over the entire input signal. In all auto-bias circuits, the Achilles heel is that the grid bias increase substantially when the output stage is driven hard. For example, suppose the grid bias is -12 volts at no signal. When driven close to its maximum output, the bias will change several volts, becoming more negative by roughly four to six volts. This drives the tubes closer to cut-off and causes distortion during high output demands. In this DC restorer circuit, a 12 volt zener is connected to the cathode bias resistor and fed back to the grids such that the grid bias is held at a constant 12 volts, thereby restoring the lost DC component of the drive signal. This is regardless of the cathode current that causes a voltage drop across that cathode resistor! 'viola! No crossover distortion. I'm certain that amplifier designers of the past would have done the same thing, except zener diodes did not exist then (when they were alive and designing auto-bias amplifiers). Great question, Thanks for asking, Bob Carver

Q: Hello Bob, I just finished rebuilding a Citation II as well, and I would also be interested in any document you care to share regarding the updates you made. I’m especially interested in running the KT88’s at a lower current using the “DC restorer” Thanks, Scott Jun-27-11

A: Hi Scott, I would be happy to share the circuit with you. I intend to post it on the Carverfest website in a week or so. In the meantime, you can install the DC restorer by using a pair of 4148 diodes connected anode-to-anode, the cathode of each one connected to the grid of each output tube. Connect the common anode connection to to the lowest (most negative) bias voltage(-55 VDC)through a 91K, 1/2 watt resistor. In other words, one end of the resistor goes to -55 volts, the other goes to the common anode connection of the 4148's. Set the idle current for 26 ma on each output tube. To obtain enough range with the existing controls, you will need to remove R30 and R69 (Sam's schematic), the optional 18K resistor shown on the schematic. Some factory-built units already have it removed, so check your unit to see how it came from the factory so long ago. A vacuum tube 6AL5 tube will work slightly better than the 4148, as it's smooth turnon characteristic does not induce the small non-linearities that the 4148 does. Not a big deal either way. Hope this helps, Bob Carver

The restorer is simple yet ingenious. It uses a 6AL5 dual diode. Each diode section services one bank of power tubes. The 6AL5 cathodes are connected to the control grids of the push-pull power tubes which are held at a nominal fixed bias of -47VDC. The diode plates are at an acquisition threshold of -56VDC. Bob admits that some aspects of the DC restorer operation are somewhat mysterious to him, as they are to me as well. My take is that the circuit aids significantly in recovery from hard cutoff conditions. Under those conditions the KT88 grids act as rectifiers and shift the effective DC level below -56V, to the point of causing the 6AL5 to conduct momentarily and pull the DC bias back to its nominal value. Bob estimates the lifetime of the 6AL5 as about 50 years. "I know that seems wrong, but 6AL5's are ubiquitous in tuners, and 50-year old tuners always have these tubes and they still check as new. There are so many of them in this world that a guy on eBay sells a string of them for three cents (each tube) to be used as Christmas tree lights."

I have the DC restorer (clamp) in all my tube amps now....I only use SS diodes..
Bob has said the 6AL5 tube...Is a little sweeter sounding but the SS diodes work fine..
One thing I do diff from Bob setup... is I put the diodes at the grids of the output tubes ...not at the front of 1k res. that feed the grids....this sounds better to me....

But there more info on the Craver site on the DC restorer (clamp)
on the Audioasylum. site theres h/k Citation 2 schematic that shows the MOD

Q: Hi, as a matter of interest, because the output transformers are so tiny, how is the bass performance of this amp? In my experience, with such small transformer cores (not hi-fi at all, such toy trafos were/are used in guitar amps where there is nothing below 70Hz!)you'd be lucky if the -3db frequency is at 30-40 Hz, even at 1 Watt, let alone at the full power! Cheers, Igor

A: Igor, Great question, and glad you asked. It's a funny thing about output transformers. Their high frequency response is essentially independent of their physical size. The low frequency response can also be independent of its physical size, provided the primary has enough wire turns on it to raise the magnetizing inductance high enough to avoid overloading the tubes at low frequencies. Unfortunately, by winding the primary for good low frequency performance, we lose the high frequency performance. The only way around this problem is to make the transformer's physical size much larger. As you can see from our measurements below, the high frequency response is phenomenal, and even the low frequency response is excellent. How can that be! What's the catch? There is none - almost. This is a modestly powered amplifier after all. ALL designs represent compromises, and with this small transformer its ultimate low frequency output at high power is not as great as it would be if it were simply larger. Despite the small transformers, this unassuming amplifier has phenomenal frequency response; that's why we like it so much. It’s really good! Its measured frequency response is flat over the entire audio bandwidth, with the -3dB down points measuring 16 Hz and 58 kHz. In my book, this is superb performance from an amp this size, or any amplifier for that matter. Having said all this, I have listened to it with and without a subwoofer. Obviously the sub with its unlimited low end gives this amp everything, and eliminates the need for it to ever go below 80-90 Hz. My personal setup incorporates this amp with NO subwoofer, and within its power limits I find it completely satisfying, with all the lows I could rationally want along with a sumptuous high end. There are two primary measures of an amplifier's response. One is called the power bandwidth which measures its frequency response at full output power, the other is simply called its frequency reponse, which measures the output at small levels, typically around one watt. We listen to the frequency reponse in our listening rooms with music, and we measure the power bandwidth on the laboratory bench. Both specifications are important, each describing a fundamental aspect of the amplifier. This amplifier has a flat frequency response extending from 16 Hz to 58kHz, and its power bandwidth is 26 Hz to 43 kHz. I hope my answer didn’t put you to sleep, and thanks for writing and for a great question. Bob Carver


A: Hi zum, The paint, polish, rubber feet and phono jacks would be what I
would call restoration. The new design of the power supply to increase
the energy storage and eliminate low frequency "gulp" distortion and
bounce, the DC restorer, the increased gain structure for the front end,
the new values of negative feedback (which now include current feedback
as well as voltage feedback),the new cathode bias scheme which reduces
distortion (this is in addition to the DC restorer), the new time
constants associated with the grid drive to the output tubes, again to
eliminate low frequency bounce and extend the low frequency response,
the reduction of the power supply impedance - I could go on and on. All
these changes fall into the category of redesign. Each change has a
specific reason, and is designed to help make the sound great sound. It
tickles my fancy that your M-500 is still in your system after all these
years! Thanks for the accolades and even more thanks for writing, Bob
Carver

Q: hi, mr Carver. in your previous offerings of more powerful designs you used 6al5 tube in
DC restorer. now it 12v Zener. any reason to change approach/SS vs vacuum
tube/?any difference in sound?if i understood your 6al5 use correctly, it
actually works as "vacuum" zener-when it conducts it stay at ~20-30
volts range. not as steep as SS zener, but that may be even better. works
similar to compressor use. will appreciate your answer. thank you for your
everlasting engineering desire. with all respect, Arkady.

A: Hi T-53, The 6AL5 is best used for fixed bias amplifiers such as my big
high power amps. On the other hand, the zener diode is is best for an
auto-biased amplifier
because it is the increased voltage drop across
the cathode resistor during hi-drive that generates the need for DC
restoration. As long as the shift is smooth and has no induced
non-linearities, the sound will be the same. I carefully made certain of
that when I designed the DC restorer for this amplifier, as you can see
in the 'scope photos; the truth is in the pudding. Thanks for writing
and especially for asking such an intriguing question, Warmest and best,
Bob Carver

Q: the Achilles heel is that the grid bias increase substantially when the
output stage is driven hard. For example, suppose the grid bias is -12
volts at no signal. When driven close to its maximum output, the bias
will change several volts, becoming more negative by roughly four to six
volts. This drives the tubes closer to cut-off and causes distortion
during high output demands. In this DC restorer circuit, a 12 volt zener
is connected to the cathode bias resistor and fed back to the grids
such that the grid bias is held at a constant 12 volts, thereby
restoring the lost DC component of the drive signal. Wow Bob¡¡ Roberto
again, even with my poor technical knowledge I understud how the DC
restorer work.. great solution thanks for sharing it. Will it work on
single ended to ?? Best regards Roberto

A: Hi Roberto, It would work, but it's not necessary as a single ended output
stage already has its DC center voltage well established, and even if
it does shift off center, crossover distortion cannot occur because it
is intrinsically free of crossover distortion to begin with. The reason
our hi-fidelity business took off in the early fifties was because Frank
McIntosh (of McIntosh fame)invented a way to make class AB circuits not
have any notch distortion, and low crossover distortion followed
rapidly in its footsteps. That allowed a power increase from about 6
watts to 75 watts, a full order of magnitude. And the hi-fi industry
took off! Thanks Roberto! Bob Carver


Q: How can the tubes last for 50 years when most manufacturers recommend changing the tubes in their amps in 3 or so years? Dave

A: Hi Dave, That is a very good question! The answer is the DC restorer.
Here's how. Most amps idle the output tubes at an average of 32 watts or
so. Now the DC restorer allows the tubes to idle at about 9.75 watts.
Since tube longevity is roughly proportional to the plate dissipation
raised to the 2.3 power, we have 32 divided by 9.75 raised to the power
of 2.3 = 15.4. Finally, 15.4 X 3 years = 46 years. I have also noticed
that many console amplifiers have 50 year old tubes right here on eBay,
and those tubes check out great. Great question, hope this helps, Bob
Carver
Thanks J!m, now ya really got my had spinning. My brain works best with three dimensional things but I keep trying to understand electronics.
 

nakdoc

Chief Journeyman
Joined
May 11, 2011
Messages
619
Location
Nashville, TN Music City
Tagline
highly biased
#10
thanks. I love this stuff. The wonderful Walsh tweeter used in many good Infinity speakers (and one dreadful, the 2000AXT) was an obvious copy of Hegeman's, it appears.
 
Top