Bradley Labs Guitar Amplifiers

Q & A

Q:I have read that when the output stage is configured in the “push pull” arrangement all even order harmonics are cancelled. This is in conflict with your explanation of why the Iron Horse sounds “different”.


Q:  Why did you use a 12DW7 "specialty" tube for the input pre-amp V1?  There is very little historical background on the 12AX7 high mu section of the tube.  What would happen if I replaced V1 with a 12AX7 or other dual triode with the same pin out?


Q: What is the Volume Quench function all about?


Q: What was your motivation for adding the "Feedback" control and how does it work?

Q: Would you modify the basic design to more closely fit a user’s specific requirements?

Q: Have you considered operating the output in Class AB2 or possibly switching between AB1 and AB2? 

Q: Why do you specify Output Power with a resistive Impedance that is 1.5 times the indicated front panel output impedance? 

Q: How hard is it to change the line voltage selection to another voltage?

Q: The specifications state that you can get an extra 10 Watts of power by placing the Impedance switch to twice the speaker cab impedance when using 4 tubes, using the same logic, I would expect you to state that you could get more power out when using 1 tube by switching the Impedance switch to half the speaker cab impedance.

Q: I was listening to the sound clips and couldn’t associate the “Sound” of the Iron Horse with any other amplifier I’ve heard. What do you attribute the different sound to?


Q: Your amplifier has a single tube position as well as Pentode, Triode and Ultra Linear modes of operation. The Output Transformer seems too small to operate in the Class A mode and Triode and Ultra Linear modes need a very clean B+ supply due to the output tube having a low output impedance. How do you get around these seemingly conflicting operational modes?

 

Q: I've heard of Pentode and Triode, but what does the Ultra Linear switch position do?

 

Q: What make tubes are you using?

Q: Why are the 2 transformers field-aligned to each other?

Q: Are you concerned about the mini toggle switch handles breaking easily?

Q: What are the cover options?

 Q:I have read that when the output stage is configured in the “push pull” arrangement all even order harmonics are cancelled. This is in conflict with your explanation of why the Iron Horse sounds “different”.

A:  You are correct in your understanding of the "push-pull" configuration.  To explain why this occurs, consider the following: 

Even order terms all return positive results, and odd order terms retain their original polarity. For example when you consider the second harmonic, it is represented by the squared voltage. Squaring any polarity number, positive or negative, always returns a positive number. Now take the third harmonic which is represented by the cubed voltage. Positive numbers return positive cubes and negative numbers return negative cubes. This is true of any odd order term.  Now consider how the push pull stage operates. The signal is split in two, with one side representing the input signal and the other side representing the inverted input signal. Next the two signals are presented to the two output tubes. There can be parallel connected pairs for more power as long as the number of tubes remain the same on each drive signal. The output of the tubes now go to the output transformer and get added out of phase. The always positive even order terms cancel, while the odd order terms add together. Now getting back to the question. The cancellation can only take place if both signal paths are exactly matched. Slight imbalances of the  phase splitter will allow more of one phase than the other letting a percentage of the even order harmonics through. The second harmonic is the majority of distortion in the sweet spot as explained in my response earlier. The phase splitter in the Iron Horse is purposely miss matched about 20% to enhance this effect. For pure second harmonic bliss, configure the Iron Horse to the single tube class A1. No even order cancellation  is present in this output configuration.


 Q: Why did you use a 12DW7 “specialty” tube for the input pre-amp V1. There is very little historical background on the (12AX7) high mu section of the tube. What would happen if I replaced V1 with a 12AX7 or other dual triode with the same pin out?

 

A: This is a good question which uncovers one of the more interesting aspects of engineering. Replacing V1 with a 12AX7 would show no difference in operation if the Drive selection were placed in the HiZ position. A 3dB drop in boost gain would be observed in the LoZ position as compared with the 12DW7.  The engineering decision made was based on the need for two different tube types needed for the optimum operation of two different functions. If one has a preference of  pre-amp tubes by all means swap it out. In the broader sense any of the three signal chain tubes V1,V2 & V3 can be replaced with 12AT7, 12AX7 or any dual triode with the same pin-out and filament voltage without incurring any damage. The probable outcome would be less gain. The output tubes can be replaced with variations of the 6V6 such as the 6Y6 or for a really big change in sound try a 6L6. The 6L6 will not overload the power transformer filament winding as it is rated at 6A which leaves 1.3A per output tube. The 6L6 is rated at .9A. Leaving the Bias potentiometer set to the 6V6 bias adjustment will provide about the same output power as the 6V6 tubes. The sound however will be quite different. To be complete, the rectifier V8, 5AR4, can be replaced with other pin out compatible variations such as the 5Y3 and the king of rectifiers, the 5U4 for different droop characteristics. The 5U4 filament draws 3A, the power transformer’s rated rectifier winding design limit. Using 6L6 tubes or a 5U4 rectifier will be too tall to fit in the head cab enclosure so experimentation must be performed on the open chassis version of the Iron Horse.   Custom head cabinets are an option should the customer request a custom tube-compliment.


 Q: What is the Volume Quench function all about?
 

A: The Quench mode came about early in the prototype phase of the Iron Horse. To be quite truthful, I didn’t realize the usefulness of the Master Volume function until Matt pointed it out to me. My use of it was to hold the settings on a multiple channel amplifier so you could have a convenient place to reduce volume on all channels globally. Matt explained that it was a means of cranking up your pre amp gain without blowing people out of the house. My response was, you could do the same thing by placing a Power Attenuator between the Amplifier and the Speaker. Although in the Master Volume case we are allowing the preamp tubes go into limiting only, whereas using a Power Attenuator allows both the preamp and the power amp tubes to go into limiting. Unfortunately a Power Attenuator would mask out our Feedback adjustment due to the uncoupling of Amplifier output Impedance through the Attenuator to the Speaker. I didn’t want to compromise that function.

As a temporary experiment, I wired in a Master Volume pot and let Matt use it for awhile. Interestingly he came back with only a limited range of adjustments. After thinking about it for a few days I decided to try an idea that came to mind. I wired up the second section of a dual, single shaft pot into the Master Volume position of the Amplifier. The extra pot was wired backwards so it was minimum at maximum Clock wise rotation. The “Quench” mode was born.



First Impressions of the "Quench Mod":  When I first plugged in, dialed in a sweet spot and was completely, totally blown away! This was way better than the master volume pot, and the master volume pot was very good. I was immediately hearing early EVH, early ZZ Top with the unique Iron Horse voicing that I’ve come to know and love. I called Don on his cell phone and made him listen to my riffing as he was driving down the 101 freeway. For a guy that loves his ‘clean sound’ I think he was even sort of impressed with the beautiful sound of the searing distortion. 

The quench mod with the HiZ/LoZ setting and Input Hi/Lo Input Switch

I first tried the mod with my Charvel San Dimas dual humbucker guitar. I turned the amp all the way up and dialed it into the “Quench” region. I had that sound, that searing, out of control, sustaining distortion sound. I achieved this sound by utilizing the Hi input setting and HiZ gain switch almost by accident. When I plugged in, I did not change these settings as I would normally to get the amp to clip a bit sooner. I had simply left it all to my standard default clean settings, and when I switched both of the switches to their “Lo” settings, my ears heard just a bit too much distortion, as I liked the cleaner setting better for note separation and definition.

The next test was with my Stratocaster. The pickups are fairly low output, so they generally don’t do the best job of driving an amp into distortion. Using the same settings as my first test with the Charvel, I had the distortion, but I was not getting the sustain I was looking for. I flipped the HiZ Switch to the LoZ setting and was rewarded with more sustain. I finally took the input switch and switched it to low and there it was….that sound. I had that late 60s crazy Strat sound! The amp settings almost made my single coils approach humbucker fatness. I kept the Strat plugged in for a long time and I did not want to let the moment pass. We all search for that magic guitar moment and I was reveling in mine! This definitely took the limitations of the single-coils completely out of the picture.

For both guitars, I got what I was looking for: beautiful, musical distortion and more importantly: sustain. The quench mod is a unique approach and there is nothing subtle about it. You dial in your over the top sound, then back things down with your guitar’s volume knob.  

 What was your motivation for adding the "Feedback" control and how does it work?

 A:  The inclusion of a Feedback adjustment came about after trying to decide how much feedback to use. The different configurations of Triode Ultra Linear and Pentode all have their own effects on over all output dynamic impedance, thus affecting how much feedback is needed. I could have set a Damping Factor goal and put in fixed Feedback values in the selection switch, but that would take away the experience of a user finding their own favorite adjustment for the particular speaker setup used. See “Notes the on Character Section of the Iron Horse” for more details on Damping Factor and how it affects the speaker performance.

 Q: Would you modify the basic design to more closely fit a user’s specific requirements?  

A: There are a few modifications I would consider, however some may prevent the Iron Horse from fitting in the standard Pelican case or require some extra holes drilled into the chassis.  

Some Possibilities:

-Class AB2 45W with new Output Transformer

-6L6GT 50W with new Output Transformer (Head Cab option only)

-Re assigned Presence or Feedback pot to become a Master Volume

-Switch in back to allow 2 Master Volume presets (Head Cab option only)

-Master Volume Pot on back (Head Cab option only)

 Q: Have you considered operating the output in Class AB2 or possibly switching between AB1 and AB2? 

 

A: Yes I have. The complexity of Class AB2 didn’t seem to be worth the trouble. As a side note, Class AB2 was used extensively in the Juke Boxes of an earlier era as well as Modulators used in AM transmitters. Both instances need high peak power in bursts as in Music Power for the Juke Box and conversational speech for the Modulator. Notice the high power output is only needed in bursts, average power remained low so tube life was not much different than continuous output at lower power levels available in Class AB1.

 

For the wider audience, output tubes running in Class AB1 represent 2, or some multiple even number of tubes, operating with Grid to Cathode voltage in the negative region, less than zero volts. Once the Grid to Cathode voltage is allowed to go positive, the transition into Class AB2 is crossed. The benefits of Class AB2 are almost twice the Plate current for a given Plate voltage when the Grid to Cathode voltage is allowed to go 15 to 20 Volts positive. This translates into twice the available Output Power for a given plate voltage swing.

 

It can be compared to a Super Charger used on an Internal Combustion Engine. One of the reasons Class AB2 is not used is because the output tube is called upon to produce twice its maximum design output power. The unknown user use profile would cause early tube wear out if high output powers were called upon frequently. Just as a Turbo Charged engine would quickly wear out bearings and piston rings if full power were used extensively.

 

Another reason is once the grid to cathode voltage goes positive, the grid starts to conduct current, just as a diode would. The Phase Splitter, which drives the output tubes, is designed to drive a high impedance load. Once the grid to cathode voltage goes above zero volts, the Phase Splitter must supply current. The Phase Splitter is configured as a common cathode differential pair, which can only sink current. The plate resistors are the current sources needed to drive the output tube grids. The net outcome is that the Phase Splitter starts to clip once grid current starts to flow. The more you drive the Phase Splitter the harder it clips, never being able to get its output above the level needed to go into Class AB2. All Guitar amplifiers are designed in this manner.

 

It may be noted that some designs use cathode followers between the Phase Splitter and the output tubes, but in all cases these are AC coupled. The AC coupling prevents the grids from going positive. The harder you drive the grids, the more negative the average grid to cathode voltage goes, causing the output stage to remain in Class AB1.  There are ways to avoid this drive inadequacy, namely use a Cathode Follower after the Phase Splitter with its cathode connected directly to the output tube grid or use an inter-stage transformer driven by a power tube.

 

The second part of the problem is the Output Transformer. When the Output Tubes start to use this additional current capability, the output Peak to Peak Plate Voltage will increase. An optimally designed output stage will have an output plate voltage swing that approaches twice the B+ supply. The additional current will now simply cause the Output Tubes to clip. No additional output power will be gained. In order to utilize the additional available current a different turns ratio is needed to make the Output Transformer half its original impedance. This allows the same plate to plate voltage at twice the current resulting in twice the output power.

 

Of course this implies an entire design that is either Class AB1 or Class AB2 depending on the Output Transformer used as well as the modified Output Tube drive circuitry. One could argue that an Amplifier as versatile as the Iron Horse should have the capability of switching between Class AB1 and AB2 just to complete the package. The cost of a dual primary Output Transformer along with the complex switch and Grid drive circuitry would have driven up the selling cost by at least $300.

 

Early on in the design phase, a breadboard of a Class AB2 design was built. This entailed a modified Grid drive circuit and a half impedance Output Transformer. The output power did indeed increase to approximately 45Watts. Not quite the 50 Watts I was expecting due to starved plate current at low plate voltages near clipping levels. The extra head room was nice for that clean sound I prefer. The subtle distortion at moderate listening levels was now gone moving up into the higher output levels. This would be more preferable in a larger listening area. All in all I think I made the right choice in disregarding this interesting mode of operation, as probably no one would use it, yet everyone would have to pay.


 Q: Why do you specify Output Power with a resistive Impedance that is 1.5 times the indicated front panel output impedance? 

A: Early in the design of the Iron Horse, I was faced with the decision of what impedances I should design the output transformer to match. The Primary side of the output transformer is constrained by the output tube plate voltage and maximum plate current at 0V control grid voltage. The secondary side is simply the impedance you want to deliver your maximum power into. As we know from the speaker impedance graph, there is no fixed impedance to design to. I decided to put the power where it was needed, namely the low frequency end where the speaker is least efficient. A typical Guitar Speaker will have an impedance peak in the vicinity of the open #6 E string at 83 Hz, where the impedance can easily climb to 8 times the rated speaker impedance. A conservative multiplier of 1.5 times the rated speaker impedance will maximize the power delivered to the speaker in the low frequency range with minimal impact on the mid range where the speaker is more efficient. The net effect is a more efficient use of the available tube output power to maximize speaker output Sound Pressure Level.

 Q: How hard is it to change the line voltage selection to another voltage?

 

A: This is a factory wiring modification. The Power Transformer has all international voltage selections available, so It’s an easy wiring job at button up time, however, it is not an easy change once configured for the country it will be used in. My suggestion would be, if you plan on using the Horse in many countries, contact Bradleylabs and we can discuss the options available. A simple solution would seem to be a 110/220VAC toggle switch on the back. Although simple in implementation, the Amplifier will be seriously damaged if the selection switch is placed in 110VAC and 220VAC is applied. Additionally safety regulations are different in various countries, with Grounding being the most difficult. The best solution would be to drag around a converter transformer.

 Q: The specifications state that you can get an extra 10 Watts of power by placing the Impedance switch to twice the speaker cab impedance when using 4 tubes, using the same logic, I would expect you to state that you could get more power out when using 1 tube by switching the Impedance switch to half the speaker cab impedance.

 

A: My friend you are correct!  I just scratched up a couple of pages with equations, graphs and diagrams and lo and behold, the power does indeed double when switching to half the 2 tube output impedance. For the example I gave in my previous explanation of 2 to 4 tubes, namely An optimized 20 Watt Audio Output Transformer designed for 2 tubes having a 10,000 Ohm center tap to 8 Ohm secondary driving an 8 Ohm load and a 1 dB roll off at 50 Hz and 15 KHz will reduce its output power to 10 Watts and retain its original 50 Hz and 15 KHz frequency response when driven by 1 tube and using the 4 Ohm output tap. The roll off stays the same due to 1 tube driving half the primary sees the same loading as 2 tubes driving the center tapped primary. The specifications are correct as stated when leaving the output impedance selection the same for 1 as required for 2 tubes. We get an extra 2 Watts by setting the Output Impedance to ½ the speaker impedance when using 1 tube Class 1A1. The Specifications will now include your observation.

 

 Q: I was listening to the sound clips and couldn’t associate the “Sound” of the Iron Horse with any other amplifier I’ve heard. What do you attribute the different sound to?

 

A: Wow another tough question. The “Sound” as you can probably imagine can be attributed to all the different available modes of operation. There’s bound to be some combination in there that is new and different. The truth is, in the basic 4 tube Pentode mode with non distorted volume, the Horse is the cleanest amplifier you’ll probably ever hear. This is no fluke. The use of 6V6 output tubes biased at the “sweet” spot in their Grid voltage, Plate current curve is the major contribution. This has other connotations that cause problems with the tube’s operation. The “sweet” spot bias voltage is quite low which in turn causes large plate current to flow at all times. This would cause the design to require low plate voltage to keep the plate dissipation below the maximum allowable value. This in turn would limit the output power available. As an example the big brother to the 6V6, the 6L6, has upwards of 600 V on the plates. The grid bias is on the order of 60 V to keep the tubes from burning up. This high grid bias forces large excursions of the grid to get full power out. This in turn stresses the phase splitter preceding the output tubes. The Horse, on the other hand, operates with about 400 V on the plates and about 15 volts grid bias. The small grid bias, whether developed by the self bias resistors or the fixed bias supply, allows the phase splitter to modestly drive the grids. The subtle “Square Law” response of the 6V6 in this region of operation gives rise to melodic octave harmonics. The operation at this small value of grid voltage would cause very large plate currents to flow. The plates would glow red hot if something clever wasn’t done to limit this current. Fortunately we have another grid available. The Screen grid is typically tied to the same supply potential as the plate. If the screen voltage is instead varied in a controlled manner both high plate voltage and low control grid voltage can be had at the same time. This is the “new” sound you are hearing.

 Q: Your amplifier has a single tube position as well as Pentode, Triode and Ultra Linear modes of operation. The Output Transformer seems too small to operate in the Class A mode and Triode and Ultra Linear modes need a very clean B+ supply due to the output tube having a low output impedance. How do you get around these seemingly conflicting operational modes?

A: This sounds like a request to give away the family jewels!  Fortunately I encourage detailed questions of this nature, as I have at many times been frustrated with the lack of information available due to company proprietary secrets. So with that out of the way, let me explain the single tube operation. It is one of my favorite design conquests. First let me explain to the wider audience why those Class A output transformers are so big. A single tube operating in the Class A mode draws a nominal plate current that is half the peak plate current. This current is always flowing in the Output Transformer Primary. The Transformer Core thus, has a Bias. This causes the core to easily saturate. When that happens the magnetic coupling between the Primary and the Secondary vanishes, additionally the Primary impedance goes to a very low value. In other words, the output clips.

The core is made larger to prevent the core from saturating. The extra core material can be up to twice as much as needed for a standard 2 tube Push Pull output stage. The solution is simple once you think about it, use one of the remaining output tubes as a current sink on the other Primary winding. This balances the current in the core so no bias is present. This same tube can be used to inject out of phase ripple from the non regulated B+ supply for the Triode and Ultra Linear modes. Single tube Triode with minimal feedback is one of my favorite settings for early break-up. Listen to Matt's "Class A Mayhem" clip to hear what I'm talking about.

 Q: I've heard of Pentode and Triode but what does the Ultra Linear switch position do?

A: Ultra Linear is a concept that originated in the 1930s and later popularized in the 1950’s. It is a form of distributed loading of the output tubes. As an example, the Pentode configuration relies on a fixed, well regulated Screen Grid voltage, the Triode configuration relies on the Screen Grid being directly tied to the Plate. The Ultra Linear configuration on the other hand, relies on the Screen Grid being tied to a tap on the primary side of the output transformer. The benefits of this arrangement are the same power output capability of the Pentode with the low distortion benefits of the Triode. The Triode configuration will only provide about half the power capability of the Pentode, but has superior distortion characteristics. The dynamic output impedance of the Triode and Ultra Linear are also very low as compared with the Pentode. The nominal Output Transformer tap for Ultra Linear operation is optimum at 43% for most Power Pentodes like the 6L6GC, however the lower power 6V6GT has an optimized tap of 22.5%. These ratios have been verified extensively in the early 1950’s and will hold true if the physical dimensions of the Cathode, Screen Grid and Plate remain the same. Conventional theory holds that the optimum tap percentage matches the Screen Grid distance percentage between the Cathode and the Plate.  

  Q:  You state that you can get an extra 10 Watts of power when using the 4 output tubes by increasing the amplifier output impedance selection to twice the speaker impedance. I've read that it is not that simple because the output transformer is designed for a specific impedance. How do you get around the problem mentioned in the articles I've read?  

A: Output Transformer design is a specialty due to the conflicting constraints of Frequency Response and Power handling. Power Transformers on the other hand, have a fixed frequency, or at most, a very limited frequency range, namely 50 or 60 Hertz. Power and Turns Ratios needed for different output voltages are well understood. The designer of a power transformer simply has to insure that the core doesn't saturate and there are minimal losses of input power across the primary winding when the transformer is lightly loaded.

The same holds true for Audio Output Transformers at the lowest specified operation frequency. This translates into how big the transformer is physically due to the size of wire and the size of the iron core for a given output power. The trick Audio Transformer designers use to extend the frequency response is to interleave the Secondary with the Primary. The higher frequencies rely less and less on the Iron Core and more and more on the mutual coupling of the two windings due to proximity.

The limiting factor is inter-winding capacitance. This will dominate at higher frequencies and roll off the frequency response. At the other extreme we have Primary Inductance, as the frequency goes lower the inductive reactance will, at some point be less than the Source impedance and again the output will roll off. That said, here is the conflict, More windings yield more inductance and a lower operating frequency, but inter-winding capacitance limits the high frequency response. More windings also have more loss due to the resistance of smaller wire size. Increasing wire size to reduce losses makes the transformer bigger. Less turns and more Iron Core give better high frequency response due to less inter-winding capacitance and better low frequency response due to more permeability in the core material. Bigger is better! 

Now back to the question: An optimized 20 Watt Audio Output Transformer designed for 2 tubes having a 10,000 Ohm center tap to 8 Ohm secondary driving an 8 Ohm load and a 1 dB roll off at 50 Hz and 15 KHz will extend its 1 db roll off frequency range to 25 Hz and 30 KHz and increase it's output power to 40 Watts when driven by 4 output tubes and using the 16 Ohm output tap. This is due to the lower drive impedance driving the fixed input inductance and capacitance. The only problem that will arise is core saturation at the low end and increased current on the windings. These are simple things to fix:  More Iron and heavier wire. You will notice the Iron Horse has an ample supply of Iron!
 

 Q: What make tubes are you using? 

A: New Tung-Sol Re Issue matched Quad 6V6GT, 12AX7, JJ Electronics 12DW7, Sovtec 5AR4.

 Q: Why are the 2 transformers field-aligned to each other? 

A: The Power Transformer null field is aligned towards the input circuitry. The Output transformer alignment is such that the power transformer field is nulled at the Output transformer's output winding. This arrangement was extensively studied and verified before committing to the chassis layout. The chassis is Zinc plated 18 Gage Cold Rolled Steel which re-directs any remaining Power Transformer fields away from the sensitive input circuitry wiring. This as well as the Output Configuration Matrix switching is covered in the forthcoming applied for patent.

 

 Q: Are you concerned about the mini toggle switch handles breaking easily?

 

A:  The Mini toggle switches are within the border perimeter of the base, thus protecting them from accidental harm. These switches are used in high quality lab equipment from leading manufacturers. Should one accidentally break due to misuse, another can be shipped directly to you with wiring instructions included.

 Q: What are the cover options?

 

A:  We offer a full head-cabinet for the touring or gigging musician.  Additionally, we offer a Pelican 1520 case that is included with the Iron Horse.  For home or studio use, we've designed the amp to be visually appealing in its current state with out covering its unique design.