Building a vacuum tube amp from scratch

[Planobilly]

This is my first try at building everything myself. Programming the G Code to cut the chassis on my old Linux computer took a bit of thinking...lol Here is a short video.



A few photos so far. I will post the schematic when I finish it.

Cheers,

Billy

 

[Svein]

So - what is the schematic of this amp?

 

[Planobilly]

Hi Svein,
I have been thinking about this project for a while now. I wanted to build a chassis that could be used for several different designs and would lend itself to experimentation and ease of change. I am installing six Noval and four Octal tube sockets. The power transformer measured no load output is 731 VAC rated at 150mA, 367VAC center tap to each side for the HT. It has a 6.3 VAC @ 5A and a 5 VAC @ 2A tap. That will power most designs up to perhaps 50 plus watts.

I had in mind to build the power supply a bit like the Mesa Boogie Dual Rectifier with a switchable solid state bridge rectifier.
I had in mind to build two turret style circuit boards, one based around a Fender bassman http://schematicheaven.net/fenderamps/fender_bassman50.pdf
and the other based around a Fender Deluxe Reverb, http://schematicheaven.net/fenderamps/deluxereverb_II.pdf

All of this gives me the ability to use 6l6GC or 6V6 or 5881 or KT66 tubes with small changes to the basic circuit.

Well...that is the general idea at the moment. I also thought some of the guys here on the site might like to suggest some ideas. Nothing is set in stone at the moment.

The idea also was to see if I could build everything myself. There are a lot of different skills needed from woodworking/metalworking to computer programming to electronic design and others. How, for example do you make a instrument panel cover with nice writing? Easy I guess if you have a CNC engraving machine...I may be able to get and modify some software that will run my CNC mill for that.

None of this is turning out to be simple but it is a fun challenge and so far everything is going well. A few small mistakes in the machining that no one but me would know were there. I have not got G Code programming down very well...lol...at least I am not dealing with a thirty horse power machine that if you crash it can knock the building over..lol

Cheers,

Billy

 

[Svein]

Are you thinking of building a hi-fi amplifier or a high-distortion (guitar) amplifier?

Also, I suggest checking out http://www.r-type.org/articles/art-010q.htm.

 

[Planobilly]

Hi Svein,

This will be a guitar amp. I play guitar so it has that value also for me. I have been repairing guitar amps for a while now and I want to learn more about how they are designed. In this learning process to date it has been hard to determine what is fact and what is is simply not true. If someone knows little or nothing about vacuum tube for example, how do you sift through all the nonsense I see posted around the internet and related to me from others in person to get to a real understanding?

I want to experiment with different resistor types to hear for myself what the differences if any are. For example there is a brand of carbon film resistors called Takman which are about one dollar each. Do they really change the sound or is this just sales and marketing nonsense? For this amp it makes no difference what a resistor cost. I am not constrained by having to sell the amp. This investigation has nothing to do with a commercially viable product.

It is obvious or should be that a guitar amp designed to play rock, country, and other forms of popular music must be able to produce sounds associated with those styles. To me, a guitar amp is as much a "sound effects device" as it is a sound reinforcement device. There are styles of music like jazz for example that require a very different type of guitar amp that has minimal distortion. I own what I think is the best jazz amp ever made which is a Roland JC 120 made in the mid seventies. I have never personally played another transistor jazz amp which I thought sounded better. Well...I have not played everything that exist so I am open to something new...lol

One of the ideas I would like to explore is this question of high distortion. Can a lower distortion amp be designed to produce the sounds necessary to play popular music.

Many people like the sound of very high quality tube stereo amps. There is a trend toward producing those amps today along with a resurgence of the use of vinyl records.

Tubes are said to have a "warm" sound, whatever that means. I assume I understand what people mean when they say that. I sometimes use that term to describe the difference between a tube guitar amp and a transistor guitar amp.

A long post to answer your question "Are you thinking of building a hi-fi amplifier or a high-distortion (guitar) amplifier?"...lol

So...yes a guitar amp and I don't know how much or how little distortion needs to be there. Also thanks for the link as I had not read that one.

I don't have the skills yet to set down and design an amp on my own. That means I am relegated to copying existing designs and trying to modify those designs to better suit my needs. I understand that this is a very "old school" way to learn about design or determine the effect of components. I know that design software exist and most everything is modeled that way now days. I like to build stuff and this is for my own amusement and amusement...lol

Cheers,

Billy

BTW...I placed an order today for all the remaining parts that I don't have on hand except the output transformer.

EDIT: I read the info in the link and went back to the start and read the whole article. It described the function of valves in a easy to read manner. Thanks

 

[Svein]

By all means build a tube amplifier, but be aware of some points:

• Due to the heater, tubes run hot
• Tubes wear out and must be replaced often (I know, I worked as a TV technician while studying. I always carried a large bag full of tubes)
• You say that "Tubes are said to have a "warm" sound". MOSFET power transistors have the same characteristics as pentodes and therefore inherently have the same "sound". Since they can deliver much higher current than tubes, you do not need an output transformer. Here are some schematics: http://circuitswiring.com/300w-mosfet-power-amp-ocl-hifi-class-ab-by-k1530j201/
• If you just want high power and controlled distortion, I suggest a Class D (digital) amplifier with a signal processing stage in front...

 

[Planobilly]

Hi Svein,

My basic interest is in tube electronics. In the world of guitar players, tube amps have always been the platform of choice. This holds true today as in the past.

Sweetwater is one of the largest seller of guitar amps in the United States. They sell 264 makes and models of of transistor guitar amps and 614 makes and models of tube guitar amps. Tube guitar amps are by far the most popular for what ever the reason.

I repair guitar amps as a hobby and to make a few bucks. I am retired and have no real need to work other than to stay busy, 95% of the guitar amps I repair are tube.

I am not sure what goes on in other major citys, but here in Miami there are very few people who know how to repair tube guitar amps. In fact, electronics repair of all consumer electronics has been in serious decline for some time now.

Thanks for the reminder that "tubes are hot"! Every now and then the tubes remind me to put on my gloves...lol

I am still looking into the basic circuit design I want to use for this project. I will post something when I make up my mind.

Cheers,

Billy

 

[Planobilly]

Here is the general design idea that I will build from. Finally made up my mind...lol
I have in mind to add several things to this basic design. There exist a output transformer that can be used with both 6V6 and 6L6GC tubes...still thinking if I want to spend the extra $200 to install that...lol
This is a design Fender came out with around 1965 also known as a Blackface.

 

[Bandit127]

Well done picking a design. I have to say that it is a complex one for a build with the vibrato and reverb circuits adding up to the equivalent of 3 preamp stages (in terms of wiring it up and getting it right).

I have just picked up an old Fender transistor amp that blew up and was rebuilt as a valve amp in 2009 by someone called Jason Sansome. He based it on the 5E3 Deluxe circuit and he clearly knew his stuff. The standard of the build is exceptional and the modifications are mostly sympathetic - and some surprising (like the EL84 output tubes). It sounds awesome even at living room volumes and it considerably louder than the rhythm guitarist's VOX AC15 at band practice.

The Fender sound isn't for me though so I am considering a Marshall 18W Plexi circuit for a build project. I will start that once I have re-tubed and biased the mongrel Fender which is going to the rhythm guitarist, so next month hopefully.

Good luck with your build. I will keep an eye on this thread with interest.

 

[Planobilly]

Hi Bandit,

Yes, the reverb and vibrato does add a bit of complexity, not to mention the fact I am building this as a head cabinet and have the issue of where to place the reverb tank. I like the sound of those two effect as produced by that type circuit. I am not so keen on other methods I have listen to for the same type of effect.

I just finished up a JTM 45 for someone but I have never built a 18W Plexi. I have repaired several 18W Marshall re-issues lately.

I am going through the AB763 circuit design at the moment to look for issues I may want to change. I will install both the GZ34 rectifier tube and a switchable solid state bridge rectifier. I am not sure at the moment if the if the 16uf filter/smoothing capacitors are the best solution and I have not got into the effect of the new added solid state rectifier as this will most likely change the filter cap requirements.

Fender used a power transformer that has a 45 VAC tap in the HT secondary and I have a transformer that does not have that tap. I can build a single diode circuit off the HT secondary of the transformer I have for the bias supply like Marshall does without issue...I assume...lol...not the end of the world to buy more iron if need be.

The major changes will include the use of power scaling based on a Mosfet design IF I have room to mount the two Mosfets on the chassis in a way to provide a good heat sink. If not, I will at least add a post phase inverter master volume control. The idea of being able to use either 6V6 or 6L6GC tubes appeals to me but there is a good bit of cost associated with the multi tap output transformer.

This whole project is seriously pressing the limits of what I understand about electronics and other subjects like G Code for example...lol

I am sure some folks on the site will help me figure out the stuff I don't understand.

Cheers,

Billy

 

[Planobilly]

Update,
Building the turret board and the placement the components in the chassis has required a bit of rethinking and adjustment. I changed from the original power transformer to another one which will require an adapter plate as it has a smaller cut out dimension. More adventures in G Code for the CNC mill...lol I have yet to master the tool offset issues yet. If you want to mess things up, use a computer. If you really want to mess things up, have that computer direct the actions of a four axis milling machine...lol

I still have not made up my mind about the placement of the filter capacitors. Fender typically placed them on top of the chassis and contained them inside a cover plate on a small board. I may try to place them (there are five 16mf 475V caps) inside the chassis under the potentiometers...not sure yet.

Here is a photo of the almost finished turret board. The transformers will arrive tomorrow and I want to place everything before I drill the mounting holes in the turret board. There has been more time and work involved in defining a step by step build process than I assumed there would be. The electronic issues are proving to be easer to deal with than the logistics of the construction.

Cheers,

Billy

 

[Bandit127]

General layout dictates that the high volts are placed opposite the low volts. Power transformers and power tubes are in the opposite corner to the input.

I can see that putting the filter caps outside the chassis (as in most Fender amps) could provide insulation to the low voltages in the preamp circuit. However a lot of amps put the filter caps on the board inside the amp so it can't be that critical. Certainly the circuit boards I have seen are generally laid out in order of increasing voltage so the filter caps are placed at opposite end of the input.

Are you are logic blocked on G41 and G42 for the mill? Program G40 (tool comp off) and adjust your target size for half the mill diameter.

 

[Planobilly]

Hi Bandet,

The issues with the mill started from the beginning. Some years ago, having no experience with CNC, or manual milling, I acquired a mill that was to small to be very useful to begin with.
At that time, I did not have the time to learn how to program. Little by little over the years I have learned some things. 99% of the things I want to do are pretty simple and many things can be done manually.

The computer that controls the mill is based on a very old version of linux and needs replacing. It is not so much that I do not under G41 and G42. The issue lies in the program interface that has a screen to deal with tool offset and tool length which I have not quite come to understand yet. To further complacate the issue is the fact that the program I am using does not understand all G Codes. I know the answer to all this. I just don't want to spend the money.

As I use CNC more it is becoming evident that the amount of work required to produce only one part is rarely worth the effort unless there is no other way to do it.. Drawing a part in AutoCad or Solid Works and getting a CAM program to produce the code is a ton of work. That is why I am just writing the code for simple stuff. Once a reasonable code is produced the process of verifying that it actually works and adjusting speeds and feed rates also takes up considerable time and effort. Then there is the fixtures that must be built to hold the part. Lots of stuff for just one part.

Truth is that building amplifiers really requires a considerable amount of shop equipment and space. More than I have, and more than I am willing to acquire! Anyone who is married will also have factor in that consideration!...lol My wife is wonderful to deal with but having said that I still have to deal with her...lol

I seriously doubt that any of these problems are addressed in electronics schools !...lol We live in a world of specialization and I assume it is not very normal for one person to even want to acquire all the many skills needed to build all the various parts. This is just a fun project using the equipment and skills that I have and that I am learning. I do see examples of people who build their own vacuum tubes. Now there is a hot topic!

Cheers,

Billy

 

[dlgoff]

As I use CNC more it is becoming evident that the amount of work required to produce only one part is rarely worth the effort unless there is no other way to do it.. Drawing a part in AutoCad or Solid Works and getting a CAM program to produce the code is a ton of work. That is why I am just writing the code for simple stuff. Once a reasonable code is produced the process of verifying that it actually works and adjusting speeds and feed rates also takes up considerable time and effort. Then there is the fixtures that must be built to hold the part. Lots of stuff for just one part.

To me, this is more exciting than the actual amplifier. Just curious; you need a "CAM program" to translate you AutoCad output files? Your CNC machine doesn't accept .DXF files? Oh and I do understand how easy it would be to tear up your tools and your work without any "speed and feed" data.

Thanks for keeping us up to date on your project. Love it.

 

[Planobilly]

Hi Don,

The vast majority of CNC machines understand G Code. There are some other proprietary codes used by certain manufacturers of CNC machines. G Code is the standard. As you can see below there other standard codes, letters,symbols, and numbers involved, all of which are commonly referred to as G Code Here is the first few lines of code that could be used to cut a square in a piece of sheet metal.
%
G20 (inches)
M6 T1 (Change Tool: Diameter: 0.1250 in)
M3 (Start Spindle)
M7 (Flood Coolant On)
G0Z0.5000
G0X0.0000Y0.0000Z0.5000
G0X1.0625Y1.0625 (Origin 0)
G1Z-0.0500F10.0 (Depth 0)
G1X1.9375Y1.0625F10.0 (leg 1)
G1X1.9375Y1.9375F10.0 (leg 2)
G1X1.0625Y1.9375F10.0 (leg 3)
G1X1.0625Y1.0625F10.0 (leg 4)
G0X1.0625Y1.0625 (Origin 1)
G1Z-0.1000F10.0 (Depth 0)

Everything in parentheses is ignored by the machine. The % symbol tells the machine that a program is about to start. The second line G20 tells the machine to use inch measurements. If the second line would have been G21 that would have told the machine to use metric measurements. As you can see there are other codes such as M codes.
Here is a very SIMPLE list of codes and their functions...lol

From Power Automation America

NC Programming Codes (Note, When all this got started the term NC or numeric code was used and later the term Computer Numeric Code came into use)

NC Programming as per ISO (DIN 66025) and RS274

G-Codes simple definition
G00 Rapid traverse
G01 Linear interpolation with feedrate (Note, what this tells the machine is to use linear interpolation and the feedrate must be intered such as F10 as in the example I bolded and underlined above. F means feedrate and 10 is the numeric rate in inches per minute, at least for my machine. Some industrial machines can move at a rate of around 35 MPG! Being driven by 20 or 30 HP at 35 MPH you certainly better know where you have told the machine to do...lol )
G02 Circular interpolation (clockwise)
G03 Circular interpolation (counter clockwise)
G2/G3 Helical interpolation
G04 Dwell time in milliseconds
G05 Spline definition
G06 Spline interpolation
G07 Tangential circular interpolation / Helix interpolation / Polygon interpolation / Feedrate interpolation
G08 Ramping function at block transition / Look ahead "off"
G09 No ramping function at block transition / Look ahead "on"
G10 Stop dynamic block preprocessing
G11 Stop interpolation during block preprocessing
G12 Circular interpolation (cw) with radius
G13 Circular interpolation (ccw) with radius
G14 Polar coordinate programming, absolute
G15 Polar coordinate programming, relative
G16 Definition of the pole point of the polar coordinate system
G17 Selection of the X, Y plane
G18 Selection of the Z, X plane
G19 Selection of the Y, Z plane
G20 Selection of a freely definable plane
G21 Parallel axes "on"
G22 Parallel axes "off"
G24 Safe zone programming; lower limit values
G25 Safe zone programming; upper limit values
G26 Safe zone programming "off"
G27 Safe zone programming "on"
G33 Thread cutting with constant pitch
G34 Thread cutting with dynamic pitch
G35 Oscillation configuration
G38 Mirror imaging "on"
G39 Mirror imaging "off"
G40 Path compensations "off"
G41 Path compensation left of the work piece contour
G42 Path compensation right of the work piece contour
G43 Path compensation left of the work piece contour with altered approach
G44 Path compensation right of the work piece contour with altered approach
G50 Scaling
G51 Part rotation; programming in degrees
G52 Part rotation; programming in radians
G53 Zero offset off
G54 Zero offset #1
G55 Zero offset #2
G56 Zero offset #3
G57 Zero offset #4
G58 Zero offset #5
G59 Zero offset #6
G63 Feed / spindle override not active
G66 Feed / spindle override active
G70 Inch format active
G71 Metric format active
G72 Interpolation with precision stop "off"
G73 Interpolation with precision stop "on"
G74 Move to home position
G75 Curvature function activation
G76 Curvature acceleration limit
G78 Normalcy function "on" (rotational axis orientation)
G79 Normalcy function "off"

G80 - G89 for milling applications:
G80 Canned cycle "off"
G81 Drilling to final depth canned cycle
G82 Spot facing with dwell time canned cycle
G83 Deep hole drilling canned cycle
G84 Tapping or Thread cutting with balanced chuck canned cycle
G85 Reaming canned cycle
G86 Boring canned cycle
G87 Reaming with measuring stop canned cycle
G88 Boring with spindle stop canned cycle
G89 Boring with intermediate stop canned cycle

G81 - G88 for cylindrical grinding applications:

G81 Reciprocation without plunge
G82 Incremental face grinding
G83 Incremental plunge grinding
G84 Multi-pass face grinding
G85 Multi-pass diameter grinding
G86 Shoulder grinding
G87 Shoulder grinding with face plunge
G88 Shoulder grinding with diameter plunge
G90 Absolute programming
G91 Incremental programming
G92 Position preset
G93 Constant tool circumference velocity "on" (grinding wheel)
G94 Feed in mm / min (or inch / min)
G95 Feed per revolution (mm / rev or inch / rev)
G96 Constant cutting speed "on"
G97 Constant cutting speed "off"
G98 Positioning axis signal to PLC
G99 Axis offset
G100 Polar transformation "off"
G101 Polar transformation "on"
G102 Cylinder barrel transformation "on"; cartesian coordinate system
G103 Cylinder barrel transformation "on," with real-time-radius compensation (RRC)
G104 Cylinder barrel transformation with center line migration (CLM) and RRC
G105 Polar transformation "on" with polar axis selections
G106 Cylinder barrel transformation "on" polar-/cylinder-coordinates
G107 Cylinder barrel transformation "on" polar-/cylinder-coordinates with RRC
G108 Cylinder barrel transformation polar-/cylinder-coordinates with CLM and RRC
G109 Axis transformation programming of the tool depth
G110 Power control axis selection/channel 1
G111 Power control pre-selection V1, F1, T1/channel 1 (Voltage, Frequency, Time)
G112 Power control pre-selection V2, F2, T2/channel 1
G113 Power control pre-selection V3, F3, T3/channel 1
G114 Power control pre-selection T4/channel 1
G115 Power control pre-selection T5/channel 1
G116 Power control pre-selection T6/pulsing output
G117 Power control pre-selection T7/pulsing output
G120 Axis transformation; orientation changing of the linear interpolation rotary axis
G121 Axis transformation; orientation change in a plane
G125 Electronic gear box; plain teeth
G126 Electronic gear box; helical gearing, axial
G127 Electronic gear box; helical gearing, tangential
G128 Electronic gear box; helical gearing, diagonal
G130 Axis transformation; programming of the type of the orientation change
G131 Axis transformation; programming of the type of the orientation change
G132 Axis transformation; programming of the type of the orientation change
G133 Zero lag thread cutting "on"
G134 Zero lag thread cutting "off"
G140 Axis transformation; orientation designation work piece fixed coordinates
G141 Axis transformation; orientation designation active coordinates
G160 ART activation
G161 ART learning function for velocity factors "on"
G162 ART learning function deactivation
G163 ART learning function for acceleration factors
G164 ART learning function for acceleration changing
G165 Command filter "on"
G166 Command filter "off"
G170 Digital measuring signals; block transfer with hard stop
G171 Digital measuring signals; block transfer without hard stop
G172 Digital measuring signals; block transfer with smooth stop
G175 SERCOS-identification number "write"
G176 SERCOS-identification number "read"
G180 Axis transformation "off"
G181 Axis transformation "on" with not rotated coordinate system
G182 Axis transformation "on" with rotated / displaced coordinate system
G183 Axis transformation; definition of the coordinate system
G184 Axis transformation; programming tool dimensions
G186 Look ahead; corner acceleration; circle tolerance
G188 Activation of the positioning axes
G190 Diameter programming deactivation
G191 Diameter programming "on" and display of the contact point
G192 Diameter programming; only display contact point diameter
G193 Diameter programming; only display contact point actual axes center point
G200 Corner smoothing "off"
G201 Corner smoothing "on" with defined radius
G202 Corner smoothing "on" with defined corner tolerance
G203 Corner smoothing with defined radius up to maximum tolerance
G210 Power control axis selection/Channel 2
G211 Power control pre-selection V1, F1, T1/Channel 2
G212 Power control pre-selection V2, F2, T2/Channel 2
G213 Power control pre-selection V3, F3, T3/Channel 2
G214 Power control pre-selection T4/Channel 2
G215 Power control pre-selection T5/Channel 2
G216 Power control pre-selection T6/pulsing output/Channel 2
G217 Power control pre-selection T7/pulsing output/Channel 2
G220 Angled wheel transformation "off"
G221 Angled wheel transformation "on"
G222 Angled wheel transformation "on" but angled wheel moves before others
G223 Angled wheel transformation "on" but angled wheel moves after others
G265 Distance regulation – axis selection
G270 Turning finishing cycle
G271 Stock removal in turning
G272 Stock removal in facing
G274 Peck finishing cycle
G275 Outer diameter / internal diameter turning cycle
G276 Multiple pass threading cycle
G310 Power control axes selection /channel 3
G311 Power control pre-selection V1, F1, T1/channel 3
G312 Power control pre-selection V2, F2, T2/channel 3
G313 Power control pre-selection V3, F3, T3/channel 3
G314 Power control pre-selection T4/channel 3
G315 Power control pre-selection T5/channel 3

G316 Power control pre-selection T6/pulsing output/Channel 3
G317 Power control pre-selection T7/pulsing output/Channel 3

Note that some of the above G-codes are not standard. Specific control features, such as laser power control, enable those optional codes.
M codes simple definition
M00 Unconditional stop

M01 Conditional stop
M02 End of program
M03 Spindle clockwise
M04 Spindle counterclockwise
M05 Spindle stop
M06 Tool change (see Note below)
M19 Spindle orientation
M20 Start oscillation (configured by G35)
M21 End oscillation
M30 End of program
M40 Automatic spindle gear range selection
M41 Spindle gear transmission step 1
M42 Spindle gear transmission step 2
M43 Spindle gear transmission step 3
M44 Spindle gear transmission step 4
M45 Spindle gear transmission step 5
M46 Spindle gear transmission step 6
M70 Spline definition, beginning and end curve 0
M71 Spline definition, beginning tangential, end curve 0
M72 Spline definition, beginning curve 0, end tangential
M73 Spline definition, beginning and end tangential
M80 Delete rest of distance using probe function, from axis measuring input
M81 Drive On application block (resynchronize axis position via PLC signal during the block)
M101-M108 Turn off fast output byte bit 1 (to 8)
M109 Turn off all (8) bits in the fast output byte
M111-M118 Turn on fast output byte bit 1 (to 8)
M121-M128 Pulsate (on/off) fast output byte bit 1 (to 8)
M140 Distance regulation “on” (configured by G265)
M141 Distance regulation “off”
M150 Delete rest of distance using probe function, for a probe input (one of 16, M151-M168)
M151-M158 Digital input byte 1 bit 1 (to bit 8) is the active probe input
M159 PLC cannot define the bit mask for the probe inputs
M160 PLC can define the bit mask for the probe inputs (up to 16)
M161-M168 Digital input byte 2 bit 1 (to bit 8) is the active probe input
M170 Continue the block processing look ahead of the part program (cancel the M171)
M171 Stop the block processing look ahead of the probe input part program segment (like a G10)
M200 Activate the handwheel operation in the automatic mode (to introduce an offset in the program)
M201-M208 Select the axis (by number from 1 to 8) for the handwheel operation
M209 Activate the handwheel operation in the automatic mode, with PLC control of the axis selection
M210 Deactivate the handwheel input while in the automatic mode
M211 Deactivate this handwheel feature and also remove the handwheel offset (if any)
M213 Spindle 2 clockwise
M214 Spindle 2 counterclockwise
M215 Spindle 2 stop
M280 Switchable spindle/rotary axis, rotary axis on, first combination
M281 Switchable spindle/rotary axis, rotary axis on, second combination
M290 Switchable spindle/rotary axis, spindle enabled, first combination
M291 Switchable spindle/rotary axis, spindle enabled, second combination


Note: Other machine functions, like tool change (usually M06) or coolant control, have their M-code value specified by the PLC application not by the CNC software. Most of the M-code values in above list are configurable.

Other M-codes (up to M699) can be handled by the PLC application based on the particular machine requirements.

The common process today is to design a part in a "computer aided design" program such as AutoCad, Solid Works, and many many others. Once that process is finished, that code is converted into G Code or other proprietary code by a " Computer-aided manufacturing" program such as Hipermill or CAM for Solid Works.

And of course, if the part is pretty simple, one can just write the G Code directly. Complex parts may require thousands of lines of g code to produce.

So...Don, I guess you now have more information about CNC than you ever wanted to know...lol

Cheers,

Billy

 

[Planobilly]

Whereas a chassis may at first glance appear to be just a "box" to hold a bunch of electronic widgets it is far from that when one has to consider how to construct that "box"! Aside from the physical/environmental demands one has to consider the electromagnetic implications of using ferromagnetic material as opposed nonmagnetic material in the construction, just to name a few of the constraints placed on the builder.

As this is the first "box" I have tried to build, I never imagined how truly complex the thinking process just to build the "box" would be.

I think I have all the "box" issues solved. Well...I assume...lol

I have a three ring binder that is getting filled up with notes, schematics, drawings, material list and all sorts of ideas and documentation.

At my age, if I don't write things down, I will lose track of what I was doing five minutes ago...lol

Onward through the fog!

Cheers,

Billy

 

[dlgoff]

The vast majority of CNC machines understand G Code. There are some other proprietary codes used by certain manufacturers of CNC machines. G Code is the standard.

Thanks for the education Billy. I had no idea that G Code is the standard. I probably deserve an infraction for not Googling.

 

[Planobilly]

You are welcome Don,

The more I try to explain things the better I need to understand it myself. I almost always learn something new trying to explain stuff. So...this is a two way learning street. Also G Code is not a common concept that I would expect anyone not directly involved in CNC machining to know much about.

Don't worry, payback is coming soon...lol, as I have several questions about the electronics part of this project...lol

At least for me this site is a fun place to learn new things and I probably deserve an infraction for being so verbose!...lol

Billy

 

[Bandit127]

I had an infraction once. It was for straying off topic... Err... Oops! Maybe that's #2...

Good luck with the electronics. There are one or two things you can do wrong that might pop a cap, sizzle a resistor or even fry a transformer, but only one thing that there may be no coming back from - so stay sharp with your hands/tools and the high voltages in there.

 

[Planobilly]

Thanks Bandit,

I will do my best to keep the HT on the board and out of my body! Actually safety is a big deal and everyone needs to be reminded frequently.

On another note, I made a mistake on the chassis front panel and have to redo it. Easy enough to make a new one with this chassis design. I always think if I never make a mistake, I am not working very hard! Part of the deal and to be expected on just about anyone off project.

I think I have all the materials on hand to finish except a .0033uf 600V cap and a 3.3M 3 watt resistor.

Thanks,

Billy

 

[Planobilly]

Hi guys,

Little by little things are coming together. I am making final placement decisions now. I hope they are correct. When building a kit all the decisions are made for you. Building from scratch involves a lot of decisions that are not so easy to think my way through. Photos as things set at the moment.

I changed the transformer so I had to make a adapter plate to make it fit. The black shiney turret board material did not stand up to the flux cleaner which dulled the surface. Unexpected issues make it difficult to get things correct on the first build.

The transformer and tube sockets are in place. The black filter cap cover and the output transformer and the choke and reverb driver I hope will be OK in these positions.

I will make a new front panel at some point I think. I still have not figured out how to make a cover plate with the lettering. It is pretty expensive to have it made, more than $100.

The chassis construction has been the most complex part of this project to date. I don't think getting the electronics working will be to hard. Making small adjustments to make things sound better will require some thinking and most likely a bit of help from others.

Cheers,

Billy

 

[Tom.G]

I still have not figured out how to make a cover plate with the lettering.

You have a CNC, find the software to do lettering and use an end mill to engrave. Paint. Let completely dry. Paint contrasting color to fill engraving and wipe off excess while still wet. When done carefully, this leaves the contrasting color in the engraving and the background is the first applied color. It's easier to get a clean job if the two paints are different chemistries, for instance Lacquer or Epoxy for the base color and Enamel for the fill color.

Second approach: Use colored anodized Aluminium and engrave to bare metal. (Doesn't look as classy though.)

 

[Planobilly]

Hi Tom,

Yes, I had considered that option. There exist engraving software that I can run. The issue is I would need to buy a high speed motor for the mill to do a good job. There is a high speed attachment designed for that purpose. It cost two or three hundred dollars by the time everything is said and done. If I remember correctly the spindle needs to turn somewhere around 10,000 RPM to do good engraving. There is plastic material with a top color and black under which is commonly used for the purpose of engraving face plates.

I see the commerical guitar amp manufacturers chrome plating or painting the face plate white with lettering I assume may be silk screened on or printed in some other manner.

It may be a less expensive solution just to have a face plate made. I am a ways away from having to deal with the issue.

Thanks,

Billy

 

[Svein]

When I did something similar long ago, I used an eccentric sander on the face plate (it made some pretty patterns). I then used Letraset for the printing and afterwards spray-lacquered the whole thing with clear lacquer. Amateurish, yes, but the total outlay was a couple of US dollars.

 

[Bandit127]

It is possible to find a custom faceplate cheaper.

This UK site will do one for £40 with all the options ticked. or £20 for something more standard.
http://modulusamplification.com/CUSTOM_FACEPLATES-W3.aspx

 

[Planobilly]

Thanks guys,

I am sure I will figure something out with the face plate. At the moment I have a few other issues to solve. The transformer does not have a center tap for the 6.3 volt heater supply like the original did. Not a game stopper but I will have to run two 100 ohm resistors to ground in parallel to form a quasi center tap. Also the turrets are a bit too small which requires me to rethink the the hookup wire I was planning to use. Just normal small issues to deal with.

In the meantime I have a engineering job to do to install a agro machine that came from Europe that runs on 240V 50Hz 3 phase. Plus I want to go to the Bahamas and go fishing...lol There needs to be priorities in this life...lol Work always has a habit of getting in the way of the important stuff...lol

Cheers,

Billy

 

[Svein]

The transformer does not have a center tap for the 6.3 volt heater supply like the original did. Not a game stopper but I will have to run two 100 ohm resistors to ground in parallel to form a quasi center tap.

Why not insert a small bridge rectifier etc. and use DC on the heaters? Then you can connect the negative polarity to ground (preferably through a resistor - I am somewhat paranoid about possible ground loops).

 

[Planobilly]

Hi Svein,

Actually I was thinking about that. I like the idea of using a DC supply and in particular to V1. I assume a 1N4004 bridge would work well. How much do you think think I would need to filter the DC? One reservoir capacitor and a filter cap would be OK? The 6.3 supply is rated at 3A.
I am assuming a full wave bridge rectifier. I don't know how to get around the voltage drop issue across the diodes which I assume would be around 1.2V or a bit more. In general the heater voltage needs to be + or - 10% of the 6.3V I assume.

The modern tube manufacturers don't give much data about heater voltages other than 12.6 in series and 6.3 in parallel. Old RCA data states .3A for 12AX7 and I don't recall for the 6V6. I think 6L6 heaters were around .9A. Those are max numbers I assume.

These are the typical issues with tube amps in general...complex transformers, high voltage, high heater current. Fun stuff!..lol

Billy

 

[Svein]

I am assuming a full wave bridge rectifier. I don't know how to get around the voltage drop issue across the diodes which I assume would be around 1.2V or a bit more. In general the heater voltage needs to be + or - 10% of the 6.3V I assume.

I have checked some suppliers, and come up with:

• You can get a full-wave bridge rectifier with a forward drop of 1V at 6A (http://www.onsemi.com/pub_link/Collateral/EN2799-D.PDF)
• Using Schottky diodes, you could get still lower
• There are several designs out there using MOSFETS in order to get rid of the voltage drop (http://www.ti.com/tool/TIDA-00858)
• And you could get fancy and create a DC-DC converter and thereby boost the output voltage to 6.3VDC (for example http://www.linear.com/product/LT3759)

 

[Tom.G]

Aaugh! Just put a 47 Ohm 1/2 Watt from each side of heater winding to Common, neatly twist the heater leads (or use a twisted pair cable) and run them in the corners of the chassis away from the low level signals. With around 4pF heater to cathode capacitance you won't have noticeable hum from the tubes themselves and the heaters will last a little bit longer than when run on DC.

If you have a DC bench supply capable of powering the heaters, try substituting that for the AC supply after you get the beast running and see if it makes any difference. You would probably only need to power the pre-amp stages with DC anyhow.

If you must have DC heaters, use a MOSFET rectifier circuit.