Ampeg 50th Heritage SVT 1969/1975 channels question

[Wasnex]

Not really
I do remember that I had to replace R49 (4.7k) on my my ~78 SVT long time ago (cause it was burned/fried).
Instead of 4.7k I did replace something ~8k for R49 without any noticeable difference soundwise to both channel #1 and channel #2, cause the preamp supply voltage didn't change much.

I don't have any idea how much of difference it would make to tone and feel, I wouldn't expect much....but the 69 PA schematic show 430V at node B versus 462V on the mid 70s schematic.

I have a hard time believing you doubled the resistance and there was no impact on the voltage. What accounts for the voltage regulation?

In your favor, the 69 preamp schematic (Rev B) shows 300V on the plate of the 6C4. The mid 70s schematic Rev D (from the 77 service manual) also shows 300V. But I think the voltage was simply not updated.

Both preamp schematics show an 8.2k resistor feeding the 6C4.

 

[ThisBass]

I don't have any idea how much of difference it would make to tone and feel, I wouldn't expect much....but the 69 PA schematic show 430V at node B versus 462V on the mid 70s schematic.

The idea behind the tests I did in the past was to figure how much the sound of the preamp does change soundwise at full output power of the amp when all of the supplies inside the amp do noticeable drop downward. And that's quite a lot of voltage drop when the power amp stage does push out its full power 300 Watts.

I have a hard time believing you doubled the resistance and there was no impact on the voltage. What accounts for the voltage regulation?

Me myself I tried hard to believe what I did notice.

In your favor, the 69 preamp schematic (Rev B) shows 300V on the plate of the 6C4. The mid 70s schematic Rev D (from the 77 service manual) also shows 300V. But I think the voltage was simply not updated.

Both preamp schematics show an 8.2k resistor feeding the 6C4.

A buffer stage (6c4) is anyway very robust versus different supply voltage. A buffer stage does not have such a thing like an idle mode biasing, such as is for a gain stage.

 

[Wasnex]

A buffer stage (6c4) is anyway very robust versus different supply voltage. A buffer stage does not have such a thing like an idle mode biasing, such as is for a gain stage.

I only mentioned the 6C4 (V5) because it's the first tube on the preamp voltage supply line. However, we should consider the holistic affect on the entire preamp.

Significance: If the voltage is lower on the plate of V5, the voltage will also be lower for V1, V2, V3, and V4.

In all honesty, I wouldn't be surprised if I could not hear any difference playing through an 810E. On the other hand, I would be surprised if I did not hear or feel at least a slight difference with a more transparent cab. However, I wouldn't expect a huge difference. I would still expect the amp to sound like an SVT regardless of whether R49 is 4.7k or 8.3k.

The value of R49 would also effect plate voltage of the phase inverter, which is fed from node C. The mid-70's schematic shows 207V and 280V on the plates of V1A and V1B respectively. The 69 schematic shows 175V and 235V

 

[beans-on-toast]

In the vintage designs they sometimes exceeded the maximum ratings of that recommended in the tube data sheets. These recommendations were for a long tube service life. The designers knew that they could push the tubes harder to get the headroom that they wanted. So it's a good idea to check the data sheets when looking at voltages to see how the readings compare with the specified maximums.

These days, they are more concerned with tube life. In some cases, voltages are lowered to contend with changing line voltages. The North American and European line voltages that keep creeping up. Where I live, I have a rock solid 117 VAC. Other areas are above 125 VAC. I don't know about Europe but I've been told of 220, 230, 240 VAC by people that tour. This variation results in a range of voltages inside the amps. Best to set the line voltage with a variac when taking readings to be consistent. Some amps have multiple taps such as 220 and 240 to deal with Europe. Others have one tap for each of Japan (100VAC), Europe, and North America, so internal voltage is a compromise. Of course amps with switching power supplies don't have this issue. The modern SVT's, including the Heritage, have jumpers for 100V, 120V, 230V, and 240VAC.

 

[ThisBass]

There is also R50 to be considered that reads 15k and that feeds from node B to the phase inverter node C
And as well there is R47 that reads 8.2k and that feeds from node B to +300V supply for the preamp.
According to the schematic the plate voltage for all the preamp buffer stages does read +300V.
Manufacturer maximum ratings for plate voltage do read (quite) often 300V.
In real practice this means that plate voltage for these tubes should not exceed 300V too much. I don't care by a couple of volts higher than this (absolute) maximum rating, yet there should not be too much of voltage supplied to the plates.

IIRC then I did notice about 350..360V for the plates that I regarded as a little bit too critical for the plates at the buffer stages.
In the year of 1987 the EU countries started to successively raise the supply line voltage from 220 VAC to 230VAC. The maximum of supply was 240VAC.
Since 2009 the line voltage specification does read 230VAC +/- 10%
In practice this means that the line voltage at the wall socket may even reach ~250VAC.

Here in my home the voltage at the wall socket is equal to 235VAC.

The SVT was designed for 110VAC/220VAC
The voltage readings in the schematic do apply to both AC supplies, either 110VAC or 220VAC can be used with proper wiring of the line transformer.

Once the SVT was connected to a line voltage that delivers higher voltage than that, all of the voltages inside the amplifier will raise accordingly!
I don't remember exactly but I think I did not had the output tubes installed when I did the mod for R49, the absence of power tubes may have caused some additional raise of supply voltage at the preamp.
Either way, I'd hate it to see wrecked preamp tubes when the amp was connected to ~240VAC wall socket (or even higher voltage) when the power tubes are not installed (or not working for ever what reason...).

If the amplifier was completely up to specs (according to 220VAC line voltage) yet connected to 250VAC (that's what randomly might happen in the meanwhile here in EU countries) then the +300V at the preamp would raise to ~340 Volts.

At the other hand side, for a dropped line voltage at the wall that was down at 207 VAC (230V - 10%) the +300V at preamp was dropped accordingly down to +270V

Personally I thing the SVT preamp works pretty fine once its rated +300V preamp supply voltage was in the range of 270..330 Volts.

The mod I once did was done with some respect to the higher AC supply that was introduced in the year of 1987 here in the EU.
At 235VAC on the wall socket the supply voltage at the preamp may now (with the mod I did to R49) measure a couple of volts lower than 300V, however I don't care about it too much.

My SVT did power very often my 212 "high definition" cab. Once in the past it happened that one of the six power tubes failed and quit working.
At least on stage and rehearsal I did not notice, neither soundwise nor powerwise, that one of the power tubes was gone.
Once there was an issue with voltage supply of the phase inverter. The voltage was way down its schematic wise rated magnitude. I was very astonished that I did not even notice this obvious issue with my high definition 212 cab.
Sometimes issues look soundwise way bigger "on paper" than they are in real practice! Yet sometimes there are issues that are even hardly recognizable on an oscilloscope, yet clearly audible soundwise. It all depends!

All of the gain stages inside the SVT don't have cathode R bypassing C. I think that's one of the reason why this amplifier is very robust vers tube rollings and also very robust vers some deviation at the voltage supplies.
Other amplifiers such as Bassman 100/135 may be more sensitive to tube rollings and furthermore more sensitive to any deviations regarding voltage supplies

There is another side effect that comes along with higher AC voltage at the wall. For 220VAC supply the full output power does read 330 Watt rms.
At 235VAC at the wall my SVT does push 350..360 Watts rms full output power rather than 330 Watts.
And I have to tell, I didn't even notice in real practice that my SVT did push some additional output power, since ~1987, the time when AC line voltage was raised.

 

[Wasnex]

AFAIK there was at least one other change that raised the voltage sent to the preamp (node B). The earliest SVTs did not have node H and the plates of V2A and V3A were originally fed from node B. In 1970 they added node H, which became the new plate supply for V2A and V3A.

My guess is the impact of the voltage at node B would be rather small at idle, and a bit more when the amp is pushed.

My 69 did not have node H when I bought it. I don't remember a significant change in the tone or feel of the amp after I did the update, but I had no way to do an AB test. However, my understanding is the update was critical. Why? Keep in mind, the bias for the output tubes is set/controlled by the current passing through the second driver stages V2B and V3B. With the original setup, V2A and V3A can drive V2B and V3B into blocking distortion. When this occurs the output tubes loose bias and conduct full on until the V2B and V3B recover.

This update would only be a concern if you have one of the earliest SVTs.

 

[ThisBass]

"Full conducting" of output tubes might be critical if there was no 47k resistors present to each signal path that feeds bias/control-voltage from driver tube to pentode control grid.

Once bias voltage is lost and the control grid starts to become conductive vers cathode/ground, the 47k resistors do help to keep (grid-)current small.

And furthermore there are the two diodes at the input stage that do provide a voltage limiting effect.
In real practice, when the power amp is pushed quite hard into saturation, the voltage applied to the pentode control grids (via 47k) is effectively "limited" by these diodes.

 

[Wasnex]

"Full conducting" of output tubes might be critical if there was no 47k resistors present to each signal path that feeds bias/control-voltage from driver tube to pentode control grid.

Once bias voltage is lost and the control grid starts to be conductive vers cathode/ground, the 47k resistors do help to keep current small.

And furthermore there are the two diodes at the input stage that do provide a voltage limiting effect.
In real practice, when the power amp was pushed quite hard into saturation, the voltage supplied to the pentode control grids is effectively "limited" by these diodes.

Pretty sure Mr Hughes is describing how this specific problem manifested in the following section of the video. (Remember he is talking about the original 69s)


His description indicates a massive problem that forever changed the design of the SVT.

 

[boroman]

@ThisBass so, do you recommend going for 8.2k at R49 if my wall voltage is usually between 240-250volts now? (I checked my rehearsal space today and it measured almost 247v). I can easily do this mod in the next days as I have my amp opened already for tweaking, and I want to gig with it regularly this year

 

[ThisBass]

IIRC then the completely gain structure for the driver stage was considerably lowered when the Stones tour was finished.
I don't remember exactly where I did read this information. It might have been an article that was published at the bass gear magazine where Mr. Hughes was interviewed about the SVT story.

During the tour, just to fix the issue on the fly, why not just spend two diodes that do help to keep the signal small, just to tame the issue at some content that was especially caused bei Keith Richards special fashion style of playing?
(of course that's nothing but guessing regarding the two diodes)

Bill does also claim that the 6550 tubes did sound the same as the 4146 tubes.
At the other hand side there is so much discussion about different sound with different 6550 tubes
GE, Sovtek, RCA, JJ, Svetlania winged C, Genalex and many others more, its hard to believe that 6550 shall sound the same like 4146 as he claims

 

[ThisBass]

@ThisBass so, do you recommend going for 8.2k at R49 if my wall voltage is usually between 240-250volts now? (I checked my rehearsal space today and it measured almost 247v). I can easily do this mod in the next days as I have my amp opened already for tweaking, and I want to gig with it regularly this year

IMO that's not so bad of an idea.
Narrow to 250 VAC line voltage, I think that's quite a lot of voltage for a device that was originally designed for 220 VAC (give or take a little bit).

If I was in your shoes I'd try to check how large the supply-voltage really is, respectively the voltage at the plates. If critical high, enlarge R49

 

[boroman]

I juest checked it and I have already original 8.2k in there (1972 SVT). Measuring up to specs, but a bit "burned" here and there. So if my plates will read super high, should I put 10k? I'm running TungSol reissues there but I think there's no specs available for that tube (all point to 1960s spec sheet which is wrong for new production). I was a bit worried when looking on EH's and JJ specs, that had max 660V (Sovteks have max 800v)

 

[Wasnex]

IIRC then the completely gain structure for the driver stage was considerably lowered when the Stones tour was finished.
I don'r remember exactly where I did read this information. It might have been an article that was published at the bass gear magazine where Mr. Hughes was interviewed about the SVT story.

During the tour, just to fix the issue on the fly, why not just spend two diodes that do help to keep the signal small, just to tame the issue at some content that was especially caused bei Keith Richards special fashion style of playing?
(of course that's nothing but guessing regarding the two diodes)

Bill does also claim that the 6550 tubes did sound the same as the 4146 tubes.
At the other hand side there is so much discussion about different sound with different 6550 tubes
GE, Sovtek, RCA, JJ, Svetlania winged C, Genalex and many others more, its hard to believe that 6550 shall sound the same like 4146 as he claims

The way I have seen it described is the headroom of the first driver section is reduced so it cannot push the second driver section into massive overload/blocking distortion. My guess is the gain is a bit lower through the first driver section as well, but the actual intent was to reduce headroom.

In the video I referenced, Mr Hughes did not actually say the 6146 and 6550 sound the same; he said they sound equal. The implied meaning is not completely clear, but I assume he means they sound equally good. In all honesty, I think it's possible he's being a bit evasive...and the intent is to let the listener hear what they want to hear without actually giving an answer.

My 69 definitely sounds different than my 87, but I have no way saying it's because of the 6146s.

Actually the way my 69 sounds changed noticeably with the new custom transformer I recently installed. Previously it was running a 6550 transformer and had a resistor network to drop the 350V screen supply down to about 220V. With this setup, the 69 was a bit more compressed and not quite as loud as the 87. There were other tonal difference, but fairly subtle. Both amps sounded like an SVT and both sounded great!

The new custom Heyboer transformer runs at 237V on the screen supply. I believe most of the original PTs run at about 230V. I asked for 235V and they got very close.

With the new transformer, the 69 seems noticeably more powerful and has fuller, and more bloomy lows. The lows stay clear and focused rather than getting woolly and loose. Also it seems to have a bit less of the characteristic SVT grind in the mids. It still sounds like an SVT, but think there is now more noticeable differences between the 69 and 87.

I don't have a preference for either, so I could say they sound equal...but they definitely do not sound exactly the same. Depending on the circumstances or my mood, I might choose one or the other.

 

[Wasnex]

IMO that's not so bad of an idea.
Narrow to 250 VAC line voltage, I think that's quite a lot of voltage for a device that was originally designed for 220 VAC (give or take a little bit).

If I was in your shoes I'd try to check how large the supply-voltage really is, respectively the voltage at the plates. If critical high, enlarge R49

FYI the schematics say 120V/240V and 50-60hz. The 69 schematic runs only 120V 50-60hz.

I do agree that many of the power supply voltages seem to run higher than the schematic. AFAIK the heater supply sits right about where it should 6.3V. In 69s, Plate, Screen, and Bias supplies run significantly higher than the schematic. In 6550 based amps, I believe Plate and Bias supplies run high, but the Screen supply typically sits right at 350V which is what the schematic indicates.

Schematics show 660V at node A. But my research suggests about 690V at node A is common with the the amp biased to spec and idling.

 

[ThisBass]

I juest checked it and I have already original 8.2k in there (1972 SVT). Measuring up to specs, but a bit "burned" here and there. So if my plates will read super high, should I put 10k? I'm running TungSol reissues there but I think there's no specs available for that tube (all point to 1960s spec sheet which is wrong for new production). I was a bit worried when looking on EH's and JJ specs, that had max 660V (Sovteks have max 800v)

Speaking for the preamp tubes, I did kinda testing in the past how much of plate voltage such a rather typical preamp tube will handle before it quits working.
Cant't tell exactly anymore cause its too long time ago, I think the tube I used for testing did quit working at plate voltage higher than 360..370 Volts.
I wouldn't expect any tube takes 360 Volts at the plate before it quits working, some may take a little higher, others a little less.

Regarding power tubes, there is no way to reduce plate voltage for the 6550 tubes.
A couple of years ago there was some discussion in Germany about JJ 6550 tubes in the SVT cause the plate voltage exceeds the manufacturers maximum rating for plate voltage. In the meanwhile the JJ 6550 has been proved as a reliable 6550 tube even when used in the SVT power amp.

 

[ThisBass]

FYI the schematics say 120V/240V and 50-60hz. The 69 schematic runs only 120V 50-60hz.

True, the schematic reads 120 VAC (respectively 240 VAC)

The question is, does all of the voltage readings in the schematic apply to 110/220 VAC line voltage or to 120/240 Volts.
Personally I think its the maximum line voltage rating the amplifier can be driven safely without damage.

The original question asked was if some difference in supply voltage at the preamp may have an effect to the voicing (channel one and/or channel two).
Personally I think, no it doesn't matter nothing at all. At least nothing of an effect that was audible (in real practice) to me myself, even if the SVT may power a "high definition" cabinet there was no audible difference.

I do agree that many of the power supply voltages seem to run higher than the schematic.

Right, that was my thinking too when I did the mod to R49.
However, we did have some noticeable line voltage raise that was introduced in the EU in the year of 1987.
In the meanwhile our line voltage at the wall may even exceed 240 Volts!

 

[ThisBass]

In the video I referenced, Mr Hughes did not actually say the 6146 and 6550 sound the same; he said they sound equal.

Probably there is a language barrier on my side. To me "the same" means nearly identical meaning as if something was "equal".
In my understanding of words I might tell that "the same" and "equal" have "the same" meaning or alternatively an "equal" meaning.

The implied meaning is not completely clear, but I assume he means they sound equally good. In all honesty, I think it's possible he's being a bit evasive...and the intent is to let the listener hear what they want to hear without actually giving an answer.

To me as a German guy I'm probably more used to direct words. Sometimes folks (in your country) try to tell something by a "bit evasive" behaviour.
That's to me kind of a language "slang" that I still try very hard to comprehend.

 

[Wasnex]

Probably there is a language barrier on my side. To me "the same" means nearly identical meaning as if something was "equal".
In my understanding of words I might tell that "the same" and "equal" have "the same" meaning or alternatively an "equal" meaning.

To me as a German guy I'm probably more used to direct words. Sometimes folks (in your country) try to tell something by a "bit evasive" behaviour.
That's to me kind of a language "slang" that I still try very hard to comprehend.

You do a lot better with American English than I did with Deutsch, and you may have interpreted what he meant correctly.

 

[beans-on-toast]

Reference voltages apply to power transformers operating at 100/120/240VAC.

In the industry, the reference voltages are taken at the specified VAC line voltage. It's stabilized so there is no drift while taking readings. Since resistors were +/-10% voltages of individual amps can vary. The readings are compared to the design calculations. Techs typically use a variac to fix the line voltage at the nominal voltage on the schematic before taking readings. In North America that would be 120VAC 60Hz. Otherwise, the voltage is set for whatever voltage and frequency the amp is jumpered for.

These days, voltage regulators/stabilizers can be used when testing and blueprinting amps under development. They can dial in the mains voltage and the line frequency. An example is AC Power Supplies - KIKUSUI AMERICA, INC.. Bands tour with devices like this so they can use the line voltage that their amps are set at and biased for.


The Ampeg user manuals when the nominal line voltage was 117VAC used to specify that the amps could operate in the 105-125VAC range. It's been adjusted when the reference became 120VAC. Of course variation in the operating voltage affects the bias voltage. Bias can be optimize for whatever the local line voltage is to help with the performance of the amp.

 

[ThisBass]

I do remember that in the past (before 1987) many products that were made in EU offered an option to select between 220VAC and 240 VAC.
Until 1987 240 VAC was common standard in England (respectively in Great Britain) while 220 VAC was more common across all the other EU countries.
(edit: as far as I remember 220 VAC 50Hz was even common standard in the Sowjet Union and all of its "sister states. Can't tell exactly but 220 VAC may still remain the common line voltage in Russia)

Made in England in the 70's, Hiwatt's line voltage selector switch.