Embedded Files
In Fall of 2011, I taught a course on "Music Amplifier Circuits". This was geared toward senior undergraduates in an electrical engineering degree program. The point was to convey circuit design and analysis from the perspective of a guitar amplifier design. I also wanted to expose the current generation of students to valve technology and design art - this is sadly a "lost art" in the academic world. It was my first semester teaching this class, so it was a a learning experience for all.
In Fall of 2011, I taught a course on "Music Amplifier Circuits". This was geared toward senior undergraduates in an electrical engineering degree program. The point was to convey circuit design and analysis from the perspective of a guitar amplifier design. I also wanted to expose the current generation of students to valve technology and design art - this is sadly a "lost art" in the academic world. It was my first semester teaching this class, so it was a a learning experience for all.
In order to ground the discussion, provide first-hand experience, and solidify the purpose of the class I had the students build (and play) an amp. After some discussion, we settled on a Weber Tweed Bassman kit. The 5F6a circuit kit turned out to be a great choice. We opted for a complete kit and about five students worked on it.
In order to ground the discussion, provide first-hand experience, and solidify the purpose of the class I had the students build (and play) an amp. After some discussion, we settled on a Weber Tweed Bassman kit. The 5F6a circuit kit turned out to be a great choice. We opted for a complete kit and about five students worked on it.
In preparation for the kit, we discussed tube circuits using the excellent Blencowe and Megantz books as well as numerous other sources. The class and the project went well and the result was a great sounding Bassman amp with a few tweaks that I'll describe below.
In preparation for the kit, we discussed tube circuits using the excellent Blencowe and Megantz books as well as numerous other sources. The class and the project went well and the result was a great sounding Bassman amp with a few tweaks that I'll describe below.
There are numerous sites describing Bassman clone amps - a good example is (http://music-electronics-forum.com/t9395/). The students were EE majors and most played guitar. It was great to hear them actually play (I mean test) the amp in the final class sessions.
There are numerous sites describing Bassman clone amps - a good example is (http://music-electronics-forum.com/t9395/). The students were EE majors and most played guitar. It was great to hear them actually play (I mean test) the amp in the final class sessions.
The Weber kit comes with a schematic and layout wiring diagram. The rest is up to you. The kit was complete, the parts are mostly good quality, and the build process was relatively smooth. The few hitches we encountered are related here in hopes of helping others building this kit or facing similar issues.
The Weber kit comes with a schematic and layout wiring diagram. The rest is up to you. The kit was complete, the parts are mostly good quality, and the build process was relatively smooth. The few hitches we encountered are related here in hopes of helping others building this kit or facing similar issues.
Upon opening the box, we were struck with the question of "where to begin?" That was a good moment to review all the components and do some planning of the entire amp-building process. The mechanical assembly is the first step.
Upon opening the box, we were struck with the question of "where to begin?" That was a good moment to review all the components and do some planning of the entire amp-building process. The mechanical assembly is the first step.
The chrome chassis and mechanical parts are well made, but there are some issues (see below). We discussed a few modifications that required adding chassis holes, so we drilled holes and mounted the three bias test point jacks (see pictures below).
The chrome chassis and mechanical parts are well made, but there are some issues (see below). We discussed a few modifications that required adding chassis holes, so we drilled holes and mounted the three bias test point jacks (see pictures below).
After mounting all the main mechanical components and the blank eyelet circuit board, we wired the filaments (twist and route), and AC power sections. Power up and tube insertion confirmed all was well.
After mounting all the main mechanical components and the blank eyelet circuit board, we wired the filaments (twist and route), and AC power sections. Power up and tube insertion confirmed all was well.
Then the DC supply was built using the vertical cap board inside the chassis. Caps were soldered to the board before the board was mounted. A main "star" power ground was created with a loop of solid wire near the power cap board (see picture). After testing DC power up to the standby switch, the remaining DC supply was wired and tested.
Then the DC supply was built using the vertical cap board inside the chassis. Caps were soldered to the board before the board was mounted. A main "star" power ground was created with a loop of solid wire near the power cap board (see picture). After testing DC power up to the standby switch, the remaining DC supply was wired and tested.
The input stages were wired next with shielded cables and without relying on chassis for ground connections (see pictures) and tests confirmed the input stage worked - we had an amplifier!
The input stages were wired next with shielded cables and without relying on chassis for ground connections (see pictures) and tests confirmed the input stage worked - we had an amplifier!
The tone stacks were wired next and tested. The phase inverter stage was next and we confirmed its operation. Prior to the output stage, we constructed and mounted the bias adjustment board. The pots (see pictures) were mounted to be accessible through two of the speaker jack holes. We added a small piece of eyelet board for the bias connections (see pictures) and completed the output stage wiring.
The tone stacks were wired next and tested. The phase inverter stage was next and we confirmed its operation. Prior to the output stage, we constructed and mounted the bias adjustment board. The pots (see pictures) were mounted to be accessible through two of the speaker jack holes. We added a small piece of eyelet board for the bias connections (see pictures) and completed the output stage wiring.
We tested the output sockets without tubes to be sure all voltages at all pins were as expected. Bias pots were set to max neg voltage to minimize initial currents. Power tubes were inserted and power was applied and bias currents were matched near the minimum possible - around 30ma for each tube.
We tested the output sockets without tubes to be sure all voltages at all pins were as expected. Bias pots were set to max neg voltage to minimize initial currents. Power tubes were inserted and power was applied and bias currents were matched near the minimum possible - around 30ma for each tube.
The whole process was relatively painless and we never blew a single fuse. We used a variac on the AC line for safety, but we never had a problem that required low AC voltage debugging.
The whole process was relatively painless and we never blew a single fuse. We used a variac on the AC line for safety, but we never had a problem that required low AC voltage debugging.
Issues we encountered are described below:
Issues we encountered are described below:
1 - The brass grounding plate mounts on the inside of the chassis using the holes for the pots and input jacks. However, the holes for the jacks are not exactly aligned with the chassis holes so we had to re-drill the brass plate to match the chassis holes. Brass is soft, so it's easily done, but you do need a drill and the right size bit. You could probably file the holes to the right size, if you don't have a drill.
1 - The brass grounding plate mounts on the inside of the chassis using the holes for the pots and input jacks. However, the holes for the jacks are not exactly aligned with the chassis holes so we had to re-drill the brass plate to match the chassis holes. Brass is soft, so it's easily done, but you do need a drill and the right size bit. You could probably file the holes to the right size, if you don't have a drill.
2 - Soldering ground connections to the brass plate is almost impossible once the plate is mounted to the chassis since the chassis forms a huge heat sink that makes soldering near impossible. The brass plate grounding method is a poor grounding technique anyway, so it's best to avoid it altogether and run actual wires to the circuit board (eyelet board) to make ground connections.
2 - Soldering ground connections to the brass plate is almost impossible once the plate is mounted to the chassis since the chassis forms a huge heat sink that makes soldering near impossible. The brass plate grounding method is a poor grounding technique anyway, so it's best to avoid it altogether and run actual wires to the circuit board (eyelet board) to make ground connections.
3 - The eyelet board mounting holes don't align with the insulation board or the chassis holes. We found two mounting holes that lined up reasonably well and simply had to drill two more holes to establish four chassis mounts of the eyelet and insulator board. The boards are soft material and there is plenty of room on the boards and chassis, so adding holes is not a problem, but we had to avoid drilling holes in areas the transformers block.
3 - The eyelet board mounting holes don't align with the insulation board or the chassis holes. We found two mounting holes that lined up reasonably well and simply had to drill two more holes to establish four chassis mounts of the eyelet and insulator board. The boards are soft material and there is plenty of room on the boards and chassis, so adding holes is not a problem, but we had to avoid drilling holes in areas the transformers block.
4 - The fuse holder is plastic - as most are - so it's difficult to tighten to the chassis without stripping the threads. The threads stripped right away and our fuse holder is in place, but loose.
4 - The fuse holder is plastic - as most are - so it's difficult to tighten to the chassis without stripping the threads. The threads stripped right away and our fuse holder is in place, but loose.
5 - As on most clones, the AC polarity switch is mounted, but not connected to anything.
5 - As on most clones, the AC polarity switch is mounted, but not connected to anything.
6 - We did not like the idea of four RCA jacks as speaker connections. Fortunately, all four holes are exactly the right size for phone jacks, so we mounted two phone jacks for speaker connections and had two holes left that were then used for the bias pots.
6 - We did not like the idea of four RCA jacks as speaker connections. Fortunately, all four holes are exactly the right size for phone jacks, so we mounted two phone jacks for speaker connections and had two holes left that were then used for the bias pots.
7 - To ensure safety, chassis ground for the AC line (green wire) was made with a crimp connection lug (also soldered) and carefully bolted to the threaded post that is welded to the chassis. (Exposing students to 425v is frowned upon.) The kit lacked a nut for this purpose. None of the provided hardware fit the post, but it's a standard thread so a suitable nut was easily found in a salvage box.
7 - To ensure safety, chassis ground for the AC line (green wire) was made with a crimp connection lug (also soldered) and carefully bolted to the threaded post that is welded to the chassis. (Exposing students to 425v is frowned upon.) The kit lacked a nut for this purpose. None of the provided hardware fit the post, but it's a standard thread so a suitable nut was easily found in a salvage box.
8 - We opted to mount the power supply caps on the vertical cap board inside the chassis. We mounted the caps on the board first and then carefully bolted the board to the chassis. We used the same vertical post that was used for AC ground. The cap grounds were all wired together and soldered to the board mounting bracket - thereby tying DC power ground to the chassis. That vertical post became the ground tie point for AC mains ground and DC ground and chassis ground.
8 - We opted to mount the power supply caps on the vertical cap board inside the chassis. We mounted the caps on the board first and then carefully bolted the board to the chassis. We used the same vertical post that was used for AC ground. The cap grounds were all wired together and soldered to the board mounting bracket - thereby tying DC power ground to the chassis. That vertical post became the ground tie point for AC mains ground and DC ground and chassis ground.
A few modifications and choices were made to make the amp more flexible and simplify construction. Nothing substantial was done to alter the basic signal path. We wanted to preserve the Bassman tone, character and circuit behavior.
A few modifications and choices were made to make the amp more flexible and simplify construction. Nothing substantial was done to alter the basic signal path. We wanted to preserve the Bassman tone, character and circuit behavior.
1 - We opted to use the 125v primary. Since most US mains are in the 115-120v range, our secondary voltages are slightly reduced, but the lower voltages increase the safety and reliability margins, which although slight, can not hurt. The low voltage secondary (red/white) wires were left unused.
1 - We opted to use the 125v primary. Since most US mains are in the 115-120v range, our secondary voltages are slightly reduced, but the lower voltages increase the safety and reliability margins, which although slight, can not hurt. The low voltage secondary (red/white) wires were left unused.
2 - We wired the 5v filament for the rectifier tube, although we currently use the solid-state copper top, which doesn't need a filament supply.
2 - We wired the 5v filament for the rectifier tube, although we currently use the solid-state copper top, which doesn't need a filament supply.
3 - We opted to mount the eyelet and insulator boards in the chassis and then solder the parts and wires to the mounted board. We also routed all wires above the board, rather than between the eyelets and insulating board. If a wire or connection breaks (or is connected incorrectly) between the boards, it just creates a nightmare to find it and fix it. Routing wires and making connections above the board was easy and I recommend this to anyone building this amp.
3 - We opted to mount the eyelet and insulator boards in the chassis and then solder the parts and wires to the mounted board. We also routed all wires above the board, rather than between the eyelets and insulating board. If a wire or connection breaks (or is connected incorrectly) between the boards, it just creates a nightmare to find it and fix it. Routing wires and making connections above the board was easy and I recommend this to anyone building this amp.
4 - We mounted R1-R6 on the input jacks and used shielded cable to connect them to the input tube grids. This is a common wiring change that reduces noise and tendency for oscillation.
4 - We mounted R1-R6 on the input jacks and used shielded cable to connect them to the input tube grids. This is a common wiring change that reduces noise and tendency for oscillation.
5 - R14 and R15 were mounted at the volume pots, and shielded cable is used to connect them to the grid of V2a.
5 - R14 and R15 were mounted at the volume pots, and shielded cable is used to connect them to the grid of V2a.
6 - A 1/4 inch hole was drilled near the normal volume control to mount a SPDT center-off mini toggle that allows the normal channel to operate as "normal" (no cap) or "bright" (91pf) or "slightly bright" (68pf). This provides 3 voices from a single switch without having to unplug cables and change inputs. The caps are switched in/out between pins (input and wiper) on the volume control R11. See the picture below for details.
6 - A 1/4 inch hole was drilled near the normal volume control to mount a SPDT center-off mini toggle that allows the normal channel to operate as "normal" (no cap) or "bright" (91pf) or "slightly bright" (68pf). This provides 3 voices from a single switch without having to unplug cables and change inputs. The caps are switched in/out between pins (input and wiper) on the volume control R11. See the picture below for details.
7 - A small cap (0.1uf /200v) is placed in parallel with the filament winding to filter out HF noise from the AC-mains.
7 - A small cap (0.1uf /200v) is placed in parallel with the filament winding to filter out HF noise from the AC-mains.
8 - A piece of the spare cap board with three eyelets was cut and bolted to the chassis near the vertical cap board. A stiff wire was looped between the eyelets and wired to the filter cap board ground. This set of loops are exposed wires for easily making ground connections during assembly. Think of these loops as the main "star ground" for the rest of the system. The cap board ground is very difficult to access or solder to once it's mounted securely, so these loops make an accessible point for ground connections.
8 - A piece of the spare cap board with three eyelets was cut and bolted to the chassis near the vertical cap board. A stiff wire was looped between the eyelets and wired to the filter cap board ground. This set of loops are exposed wires for easily making ground connections during assembly. Think of these loops as the main "star ground" for the rest of the system. The cap board ground is very difficult to access or solder to once it's mounted securely, so these loops make an accessible point for ground connections.
9 - Adjustable bias is a necessity, IMO. However, there is no adjustment method supplied in the kit. The schematic shows a means of adding a pot, but this would only provide one adjustable voltage for both output tubes, therefore requiring a matched tube set. Matched tubes are more expensive and often not really "matched" very well.
9 - Adjustable bias is a necessity, IMO. However, there is no adjustment method supplied in the kit. The schematic shows a means of adding a pot, but this would only provide one adjustable voltage for both output tubes, therefore requiring a matched tube set. Matched tubes are more expensive and often not really "matched" very well.
It's more reliable and flexible to simply create two adjustable bias voltages - one for each tube - and adjust each tube independently to its target bias current. To make bias adjustments easy, we built a small perf-board with two mini-pots and mounted the board so the pots are exposed behind two unused RCA jack holes. (Recall, we used two phone jacks for speaker connections, so two holes are unused).
It's more reliable and flexible to simply create two adjustable bias voltages - one for each tube - and adjust each tube independently to its target bias current. To make bias adjustments easy, we built a small perf-board with two mini-pots and mounted the board so the pots are exposed behind two unused RCA jack holes. (Recall, we used two phone jacks for speaker connections, so two holes are unused).
To simplify bias current measurements, we also drilled holes and added two test points between the output tube sockets along with a third test point as a ground connection. The two test points connect to 1-ohm 2-watt resistors that are inserted into the cathode-ground paths of each power tube. Each millivolt measured at a test point corresponds to 1 milliamp of cathode current.
To simplify bias current measurements, we also drilled holes and added two test points between the output tube sockets along with a third test point as a ground connection. The two test points connect to 1-ohm 2-watt resistors that are inserted into the cathode-ground paths of each power tube. Each millivolt measured at a test point corresponds to 1 milliamp of cathode current.
Common (and safe) bias currents for Bassman 6L6 tubes are between 20 and 50 ma per tube. We biased each tube to 30 ma (30 mv at the test points) in the interest of conserving tube life, and they sound fine. (The bias voltage obtained from the transformer is relatively low, so setting a significantly lower bias current was not possible.)
Common (and safe) bias currents for Bassman 6L6 tubes are between 20 and 50 ma per tube. We biased each tube to 30 ma (30 mv at the test points) in the interest of conserving tube life, and they sound fine. (The bias voltage obtained from the transformer is relatively low, so setting a significantly lower bias current was not possible.)
10 - One side effect of two separate bias voltages is a need for an additional eyelet to mount R9 and R10 separately. The stock eyelet board does not have a spare eyelet in that area, so another piece of the unused cap board was cut out. This piece has the two needed eyelets and it was secured with one of the four eyelet board mounting screws that was conveniently nearby (see picture below).
10 - One side effect of two separate bias voltages is a need for an additional eyelet to mount R9 and R10 separately. The stock eyelet board does not have a spare eyelet in that area, so another piece of the unused cap board was cut out. This piece has the two needed eyelets and it was secured with one of the four eyelet board mounting screws that was conveniently nearby (see picture below).
The image below shows the input jacks with resistors mounted on them. The resistors feed into a 2-conductor plus shield cable that runs to the V1 grid pins. The cable shield is only connected at the jack end. The added bright switch is visible next to the jacks, along with its connections to the volume pot. Note that all the wires and components are routed and visible on top of the eyelet board.
The image below shows the input jacks with resistors mounted on them. The resistors feed into a 2-conductor plus shield cable that runs to the V1 grid pins. The cable shield is only connected at the jack end. The added bright switch is visible next to the jacks, along with its connections to the volume pot. Note that all the wires and components are routed and visible on top of the eyelet board.
The image below shows most of the eyelet board and pot connections. Note the added pyramid-shaped added eyelet board mounted with the same screw holding the eyelet board. The added board connects resistors R9 and R10 (via the thin grey and green wires) to separate bias voltages produced on the bias board.
The image below shows most of the eyelet board and pot connections. Note the added pyramid-shaped added eyelet board mounted with the same screw holding the eyelet board. The added board connects resistors R9 and R10 (via the thin grey and green wires) to separate bias voltages produced on the bias board.
Also visible are the resistors from the volume pots tied to the shielded cable connecting to the V2A grid. The shield is grounded only at the pot end. Note there are no soldered connections to the brass plate. All ground connections are wired to the eyelet board. The brass plate is just added shielding in our amp.
Also visible are the resistors from the volume pots tied to the shielded cable connecting to the V2A grid. The shield is grounded only at the pot end. Note there are no soldered connections to the brass plate. All ground connections are wired to the eyelet board. The brass plate is just added shielding in our amp.
The power supply is shown below. Note the vertical board holding the main filter caps. To the left of the grommet is an added strip of eyelet board with three eyelets supporting a solid wire loop. The added eyelet strip is secured by one of the transformer mounting screws. This loop is wired to the power cap ground and makes for a convenient main "star" ground point where other ground connections are made.
The power supply is shown below. Note the vertical board holding the main filter caps. To the left of the grommet is an added strip of eyelet board with three eyelets supporting a solid wire loop. The added eyelet strip is secured by one of the transformer mounting screws. This loop is wired to the power cap ground and makes for a convenient main "star" ground point where other ground connections are made.
A 470K resistor was added on the back of the cap board as a bleeder for the power caps. This could be removed now, but it was useful when the amp was under construction and there was little or no load to bleed the DC supply. Note the added cap to filter HF noise from the filament supply was mounted on the pilot light.
A 470K resistor was added on the back of the cap board as a bleeder for the power caps. This could be removed now, but it was useful when the amp was under construction and there was little or no load to bleed the DC supply. Note the added cap to filter HF noise from the filament supply was mounted on the pilot light.
The rear of the power cap board is visible below. Grounds for the bleeder resistor and all the power caps are connected to the board bracket, which is bolted to the chassis on the same post used for the AC-mains ground lug connection. The green wire visibly connected to that bracket connects to the "star" ground wire loop shown above.
The rear of the power cap board is visible below. Grounds for the bleeder resistor and all the power caps are connected to the board bracket, which is bolted to the chassis on the same post used for the AC-mains ground lug connection. The green wire visibly connected to that bracket connects to the "star" ground wire loop shown above.
The bias board is shown below. It's a simple perf board with two screwdriver-adjust mini-pots mounted on the rear of the board (not visible). The grey caps provide some additional filtering on the bias supply to ensure low ripple.
The bias board is shown below. It's a simple perf board with two screwdriver-adjust mini-pots mounted on the rear of the board (not visible). The grey caps provide some additional filtering on the bias supply to ensure low ripple.
There are four wire connections to the board: bias supply, ground, V4 bias, and V5 bias. To the left are the insulated bias test points (red) and the 1-ohm 2-watt resistors between them and the ground point (black - in the middle), which is connected to the star ground.
There are four wire connections to the board: bias supply, ground, V4 bias, and V5 bias. To the left are the insulated bias test points (red) and the 1-ohm 2-watt resistors between them and the ground point (black - in the middle), which is connected to the star ground.
The picture below-left shows the bias test point jacks that were added between the power tubes. There is one test point per tube (red) and a ground (black) for convenience. The image to the right shows the bias pots visible through the holes intended for RCA jacks. Two phone jacks are mounted in the remaining two holes for speaker connections. These jacks are also visible in the image above. The unused hole in the center is a hole I mistakenly added and have no current use for.
The picture below-left shows the bias test point jacks that were added between the power tubes. There is one test point per tube (red) and a ground (black) for convenience. The image to the right shows the bias pots visible through the holes intended for RCA jacks. Two phone jacks are mounted in the remaining two holes for speaker connections. These jacks are also visible in the image above. The unused hole in the center is a hole I mistakenly added and have no current use for.
The final combo amplifier is shown below with the chassis exposed.
The final combo amplifier is shown below with the chassis exposed.
The modifications to the output stage for test points and the bias board are shown below. The output stage is modified to add 1 ohm resistors between the cathodes of the power tubes and ground. The test points (TP1-3) make the voltages across these resistors accessible. Note also the separation of the 220K resistors that normally connect to one bias voltage (E). Now separated, they connect to separate voltages from the bias board - E-R9 and E-R10.
The modifications to the output stage for test points and the bias board are shown below. The output stage is modified to add 1 ohm resistors between the cathodes of the power tubes and ground. The test points (TP1-3) make the voltages across these resistors accessible. Note also the separation of the 220K resistors that normally connect to one bias voltage (E). Now separated, they connect to separate voltages from the bias board - E-R9 and E-R10.
The Bassman tone is pretty well known. You've heard it on hundreds of recordings. While each instance or variation of the Bassman design is slightly different, most have a classic sound that is recognizably "Bassman". Check out the "Enough" song on the music page. It was recorded using this amp.
The Bassman tone is pretty well known. You've heard it on hundreds of recordings. While each instance or variation of the Bassman design is slightly different, most have a classic sound that is recognizably "Bassman". Check out the "Enough" song on the music page. It was recorded using this amp.
Updates:
Updates:
This page relates to the original 5F6a Tweed Bassman, but after several years of playing with this amp, I extended it's capabilities, and after you finish viewing this page you may be interested in the new version I call the Bassman Redux.
This page relates to the original 5F6a Tweed Bassman, but after several years of playing with this amp, I extended it's capabilities, and after you finish viewing this page you may be interested in the new version I call the Bassman Redux.
Amp-Aid offers kits for Tremolo and Reverb/FX/Bias for the Bassman and I installed both in the Bassman Redux. Great additions! Some clips are on the Amp-Aid site.
Amp-Aid offers kits for Tremolo and Reverb/FX/Bias for the Bassman and I installed both in the Bassman Redux. Great additions! Some clips are on the Amp-Aid site.
A 3-position switch creates 3 tone stack options to reduce bass and remove the boominess of the Bassman.
A 3-position switch creates 3 tone stack options to reduce bass and remove the boominess of the Bassman.
Page updated
Google Sites
Report abuse