Page 1 of 3
#1
Well after building and tweaking a number of different amps over the past few years, a lot of which I tore down after a few days for parts since they weren't what I wanted, I finally think I found the one that I'm looking for.

I've always been a fan of the Vox AC30, but I've wanted more gain, or should I say a different kind of gain. I've played a Trainwreck Liverpool and loved the amp, but it also wasn't exactly what I was looking for. I've built a Trainwreck Express, but it wasn't my thing so I sold it to a friend, who had been looking for what he called a "JCM on Steroids".

The preamp section is a parallel 12ax7 that will achieve a gain of somewhere around 90. Normally I tend to shy away from high gain in the early sections of the amp, but in this case I'm immediately driving the tone stack from this, so it suffers a pretty heavy loss. That isn't a bad thing though, as I'm using an EF86 in a pretty high gain arrangement, its like a Vox input section but with a few tweaks.

The capacitor in the cathode circuit will be switchable for a gain boost/reduction. From the EF86 we go into the phase inverter, which will have removable negative feedback, possible a presence control but I have not decided upon that yet. Other than that a pretty standard phase inverter.

The output section is pretty much lifted from a Vox ac30. I split the cathode bias resistor into two instead of one. I'm working on a half power switch and this keeps each pair biased. If I used a single 50 ohm resistor, instead of two 100 ohm resistors. If a pair of tubes is removed, the remaining two will still see 200 ohm cathode resistances and not 100 ohm.

The amp will feature a 5AR4 rectifier tube. Since the Vox is not class A I'll be able to get some "sag" when the amp is pushed hard.

So far the amp lacks a master volume control, when it gets built I'll play around with it for awhile and decide if I need one. I tend to use the volume control on my guitar and pick attack more now so I don't want to put a master volume in if I don't need it.

Also, so far there are no plans for reverb or tremolo in this amp. I've never really used them much, and I'm also working on a design of an analog delay so I'll use it to add ambiance to my playing.

So recap
V1 12ax7
V2 EF86
V3 12ax7
V4-7 EL84
V8 5AR4

Controls: Treble, Bass, Mid, Volume
Boost Switch
Negative Feedback Switch
Half power switch
On/Off switch

I'm probably going to make this into a head cab, as I have some nice "flawed" flamed maple. It has some dark streaking in it which looks pretty nice.

So let me know what you guys think, or ask any questions.
Last edited by XgamerGt04 at Mar 19, 2011,
#2
Looks cool,

You may have problems with the EF86, they are very prone to noise and being microphonics. Usually they are used right at the input. You'll be hitting it much harder.

Half power switch is useless. Tried it and it did very little. Many TW guys say the same thing.
#3
Quote by kurtlives91
Looks cool,

You may have problems with the EF86, they are very prone to noise and being microphonics. Usually they are used right at the input. You'll be hitting it much harder.

Half power switch is useless. Tried it and it did very little. Many TW guys say the same thing.


That is actually why the EF86 is at that spot in the signal chain. The switchable cathode cap should keep the gain pretty tame when its not in the circuit. When its in the circuit noise may become an issue, but the quality of the EF86 tube can affect that. I'm going to pick up a NOS Mullard EF86 to see how it works out and probably a new JJ as well. Also, if I get 1V p-p out from the guitar then at full volume I have a 20V p-p signal on the grid of the EF86, so it will clip pretty early on. A good layout will also keep noise to a minimum.

As far as the half power switch it may or may not be included. I'm experimenting with a lot of different things, so i'm not sure what will happen in the end. This is probably a few weeks out from being built.

Edit: Forgot to mention, assuming that I can find somewhere that will sell me M6 laminations without a huge order I will be winding the power and output transformer for this amp. I've got access to Motor, Inductor, and Transformer winders at work as well as people who have been designing transformers for years. So the transformers will be quite stout.
Last edited by XgamerGt04 at Jul 16, 2010,
#5
Then maybe you should consider some college education on the matter. If it's any testament to the complexity of some of this stuff... at my school, it is not until the third year of EE that you learn enough to design useful transistor amplifiers... let alone huge 200-watt ones :lol:

one important thing to note is that even if someone gave you a complete schematic, it would probably not do you any good. the odds of it working right the first time are pretty slim... and the odds of it randomly blowing up are pretty high... in other words, you are going to have to know how to debug the circuit. a car amplifier is going to have quite a lot of components, so it's not going to be very simple to debug... you are definitely going to need a thorough understanding of its operation to do so. and if you are just trying to build it from someone else's schematic with no knowledge... well you're probably just wasting your time and money.
=============
Sightseeing Holidays in Australia
culinary destinations
#6
Can I ask why you're running the EF86 at a voltage gain of nearly 200? If you hit it with a 1 volt input signal, even with that big cut from the following divider you're still slamming the PI with a 15 volt input signal. I'm not sure if the switchable cap will make much of a difference.

I think you could get away with running that EF86 at a much lower voltage gain (like around 100) and have a much more reasonable result. Additionally, noise and microphonics worries would be mitigated considerably.

Just .02, looks like a cool build!
#7
I'm not really sure what the hell the guy above me is talking about, or who he is talking to. If its to me, well to put it simply I'm a senior EE student so I have more than enough knowledge to do what i'm trying to do. That, and this is an amp of my own design so I am anticipating problems with it.

Also, just as an update to the thread, I've been tweaking the design some more so I'll be posting new schematics soon. I'm going to go down to about 22 watts. I should be able to get the tone I'm after with that.

I'm also putting together a parts list and deciding what type of components to use.

So far it looks like this

Carbon Film Resistors
SBE 6PS OD caps
Illinois Power Supply Caps - I can't see spending 10 dollars a piece for Sprague Atoms when I can get all of my power supply caps for less than 10 dollars.
Sprague Cathode Bypass caps
Edcor Transformers

I'll post the updated schematics sometime later today.

The heater filaments are rectified with FREDs and a large resevoir capacitor and then regulated using a Zener Diode and an IRF1404Z Mosfet. Zener diode is bypassed with a .1uF cap.

So two EL84s are removed and the rectifier is being changed to a 5Y3. So the amp will have a NOS 5Y3 and some form of NOS EF86
#8
Quote by CECamps
Can I ask why you're running the EF86 at a voltage gain of nearly 200? If you hit it with a 1 volt input signal, even with that big cut from the following divider you're still slamming the PI with a 15 volt input signal. I'm not sure if the switchable cap will make much of a difference.

I think you could get away with running that EF86 at a much lower voltage gain (like around 100) and have a much more reasonable result. Additionally, noise and microphonics worries would be mitigated considerably.

Just .02, looks like a cool build!


The stage isn't running anywhere near a voltage gain of 200. Its voltage gain is at or below 100 at max. Take a look at the schematic again, and you'll see that the cathode bypass capacitor has a resistor in series with it. This allows a modest boost in gain.

At low frequencies, with the cap in the circuit, it will act as an "open" circuit, so only the 2.2K resistor is seen by the cathode.
At high frequencies it will act as a "short" circuit, so you will see the parallel combination of the 2.2k and 47k resistor.
In between you'll see some combination of both resistors and the cap, but it will not cause a gain of near 200.
#9
Quote by XgamerGt04
The stage isn't running anywhere near a voltage gain of 200. Its voltage gain is at or below 100 at max. Take a look at the schematic again, and you'll see that the cathode bypass capacitor has a resistor in series with it. This allows a modest boost in gain.

At low frequencies, with the cap in the circuit, it will act as an "open" circuit, so only the 2.2K resistor is seen by the cathode.
At high frequencies it will act as a "short" circuit, so you will see the parallel combination of the 2.2k and 47k resistor.
In between you'll see some combination of both resistors and the cap, but it will not cause a gain of near 200.


I understand what you're doing with the cathode, but what you're not considering is the total plate load. That multiplied by the transconductance gives you voltage gain.

A plate load resistor around 82K would get you closer to a voltage gain of 100, but it would still be a bit over. Essentially, somewhere around in the 60K area would be closest to 100.

The way your EF86 is configured now provides enough gain to drive a PI and subsequently an output section with just a single pentode preamp stage (input stage) being fed by a 100mV to 150mV signal. That tube can do a voltage gain of 100 without batting a proverbial eyelash.
#10
Quote by CECamps
I understand what you're doing with the cathode, but what you're not considering is the total plate load. That multiplied by the transconductance gives you voltage gain.

A plate load resistor around 82K would get you closer to a voltage gain of 100, but it would still be a bit over. Essentially, somewhere around in the 60K area would be closest to 100.

The way your EF86 is configured now provides enough gain to drive a PI and subsequently an output section with just a single pentode preamp stage (input stage) being fed by a 100mV to 150mV signal. That tube can do a voltage gain of 100 without batting a proverbial eyelash.


That only holds true if both the screen and cathode are fully bypassed. In this case, the cathode is never fully bypassed.

The gain of this stage is better represented by the following

Transconductance * Anode Resistor / (1 + Transconductance * Cathode Resistor)
#11
You're right, my mistake. I had a brain malfunction when looking at the schematic and for whatever reason I was doing math with a fully bypassed cathode resistor in mind!

It indeed looks like you'll be under 100 for the voltage gain.
#12
Quote by CECamps
You're right, my mistake. I had a brain malfunction when looking at the schematic and for whatever reason I was doing math with a fully bypassed cathode resistor in mind!

It indeed looks like you'll be under 100 for the voltage gain.


I've done the same thing as you, so I know what thats like. I'm thinking about changing the value of the 47K resistor to something smaller to get a slight more gain boost. I'll probably experiment with it once I actually build the amp to see what happens, or maybe even put a 50K pot in there to have variable boost.
#13
I like it, simple, but should get the job done rather nicely. The 250k treble pot, is this because the lower output impedance of the paralleled 12AX7 is low enough? I've typically seen 470k/500k after a 12AX7 stage. I'm kind-of answering my own question, but is that why you used it?
#14
Quote by blandguitar
I like it, simple, but should get the job done rather nicely. The 250k treble pot, is this because the lower output impedance of the paralleled 12AX7 is low enough? I've typically seen 470k/500k after a 12AX7 stage. I'm kind-of answering my own question, but is that why you used it?


Honestly, its just a normal fender style tone stack, with the values basically the same. I like them when it comes to tone shaping so I just stuck with those values. The blessing of using the EF86 in this configuration is that it can easily make up for all of the loss the tone stack causes plus more.

The input stage, with the parallel valves, does have a lower output impedance, but there is going to be a switch to change between parallel and single valve operation. Parallel valves also can sometimes lead to a bit darker sounding amp, and this amp will have a ton of high end, so I want the option to possibly fatten it up a bit.
#15
So I haven't had a chance to fully update the schematic as I made the decision to include a master volume, and make some changes in the boost circuit for the EF86.

The amp will now include a PPIMV control, and the EF86 will have the 47k resistor in the cathode circuit replaced with a potentiometer, and the switch removed. So with the potentiometer at 10, it will effectively be out of the circuit allowing for tons of gain. At 0 it will get slightly less gain than with just a 2.2K resistor and no cathode bypass cap.

I did design a front and back face plate for the amp, as well as come up with a name for the amp, Since its British influenced its going to be called the British Bomber. I think it fits in nicely with the name Mach 5 Amps, that I have started to use recently.

The face plates will be black plastic engraved to a white background. So all of the black text in these pictures will actually be white, and the white background will be black.
Attachments:
British Bomber Front Panel.pdf
British Bomber Back Plate.pdf
#16
As I'm getting closer to having this ready to be put together I thought I would update some more, and I have a few questions.

I'm going back to the 4 EL84 output configuration, after playing around with a few different vox amps I noticed that you never seem to have the same chime and sparkle in the 2 EL84 amps. I'm also modifying the power supply to include a choke.

So the power supply will be

Transformer - GZ34 - 32uF-15 H choke (~255 ohms)- 32 uF - 3.4K - 20 uF - 4K - 20 uF
This should give me voltages of about 326, 318, 295, and 275.

This is where my first question comes in. I typically dislike using cap cans in builds, but as I am attempting to keep the circuit board at 12" long it would be useful to use a cap can. The cap can would also just be for the plate and screen supply for the output power tubes. Are F&T multisections good caps, or do they suffer the same QC problem as JJ's and others?

The master volume has been scrapped. The one thing that I did learn while playing the trainwreck that I built was how to play an amp without one. Its probably gonna be hella loud but I have a few ways to quiet it down.

So components will be
Multicomp Carbon Film Resistors
Sozo Coupling Caps
F&T Power Supply Caps
EDCOR Transformers
JJ Tubes
Ohmite 5W resistors in the power supply and as screen and grid stoppers in power section.
PEC RV4 Potentiometers

As far as Sozo caps go, has anyone had any experience with the different types they offer. I've been hearing that the blue molded have great sparkle to them once broken in and that the vintage are great caps all around. While this amp will be used for some heavier stuff, I tend to play clean a lot so I like pristine clean tones.

I'm also starting to design the combo cab for this amp. I'm looking at it being a 2x12 with a Weber Blue Dog and Weber Silver Bell for the two speakers.

Overall I'm looking at around 700 for the amp in parts and some other little knick knacks and 500 for the cab. Not too bad for using quite a number of top of the line parts though.

As soon as I finish up my board layout, waiting to hear from Sozo what the dimensions for the caps are, I'll post that up in a pdf format as well as the planned chassis layout. I should also have an idea of the wiring in the chassis sometime in the next day or two.

After I get back from California, which will be around the beginning of the year, I plan on ordering the chassis and most of the hardware and getting to work with the mechanical work. Then as I get the money I'll order the board and other parts as well. The cab will be built sometime after the amp is working, so its still up for discussion if anybody has any good speaker ideas.

Sorry for the long post, but give me some ideas if you do read it all and have any suggestions.
#17
Are you doing a 4x12 speaker combination? I've always wanted to use Texas Heat and Swamp Thangs from Eminence. The Texas Heats are very top-ended and should be good for screaming solos, country grooves. Whereas the Swamp Thang is very low-ended and should be pretty sweet for chugging out. My only issue is, when I play clean I like a midboost, so that may have some drawbacks.
#18
It is going to be a 2x12 configuration. Its also going to be made as a combo amp. I want to include an AlNiCo Blue in there, so that is why i'm getting the Blue Dog. That should allow it to keep the voxy chime.

The other speaker is up for debate, but I've played a Blue Dog Silver Bell combo which was just amazing.

The speakers may change as it gets closer to building the amp, I'd love to hear some of these Eminence speakers to make my decision though. I wish I could go to a store that had like a "wall of speakers" or something
#20
I saw that part of the site. The reason the Weber speakers are getting a lot of consideration is that after I get the amp built and working I have a shop close to me that will let me take it in and play it on some of their cabs to decide the speakers. They have a cab there with the two Webers and everything that I play sounds awesome there.

That would be why I wish that a shop just had like a wall of speakers and some sort of "plug and play" cab. I've been to a store once that had that, and it was pretty amazing. Buying a lot of speakers is kinda expensive haha.

I did like the sound of the Texas Heat and the Swamp Thang though. I also really liked the Wizard. I still want that AlNiCo Blue sound though, which is hard to find.
#21
Parts ordered today, so this should start progressing soon hopefully.

I've got my circuit board layout done so as soon as I get it I'll print off the layout, take it to work and start populating that part. I have access to a drill press there so that is probably where I will do most of my machine work.

I'm going with a solid state rectifier and using UF4007 diodes now. All Edcor transformers.

So I should get my resistors and all of that tomorrow. The rest of the parts should all roll in by the middle of next week with the exception of transformers. They are supposedly made to order and says to give them 4-6 weeks but I've heard Edcor is much faster than that.

So this thread should start seeing some pictures pretty soon.
#23
Quote by blandguitar
What made you choose Edcor over say Hammond?


Edcor does all of their testing loaded, so when they tell you the voltages of the transformer you can be pretty sure that is what you are going to get.

Their prices are also very competitive with Hammond, often times cheaper. Hammond doesn't have a wide enough selection for me to choose from as I often want different transformers. I got my entire set of transformers for around 180 including shipping, just the power transformer alone from Hammond was

Overall their transformers tend to be better than Hammond. They are similar to Heyboer transformers, but its more of a pain to get transformers from Heyboer. With Edcor they have an online list that I can choose from and not have to deal with too many emails back and forth.

Quality wise Edcor is great too. They've been making transformers for a long time and I tend to use them as replacement transformers when I work on amps if the owner doesn't mid possibly needing new mounting holes.
#24
So I've started working on the board finally. I'm still missing the coupling capacitors and the two F&T supply caps. I also haven't put the silver mica caps on the board yet.





Those power resistors have some really nice thick leads. I'm a little disappointed with the Multicomp resistors. The ones I have at work had some really nice leads, but these are somewhat flimsy. Still better than the Xicon metal film caps though.

I should get the two F&T caps tomorrow, but the coupling caps I don't know about.
Last edited by XgamerGt04 at Jan 20, 2011,
#25
Looks good, but I don't think the pictures are big enough.

It is good to kind of learn about different brands though, because atm, I'm clueless.


████████████
████████████








RBY CYOA
#26
Quote by MonkeyLink07
Looks good, but I don't think the pictures are big enough.

It is good to kind of learn about different brands though, because atm, I'm clueless.



Shhhhhhhhhhhh.... I was resizing them : I wanted to give Jim a heart attack first though.
#28
Quote by end_citizen
Looks good. Where do you get your board material? I usually go to turretboards.com, but their prices aren't great on anything really.


When I have access to a full set of tools (table saw and drill press) I tend to buy my board from McMaster Carr. I tend to buy a 48" x 48" sheet and rip it down to 3" or 4" by 24" boards. As I don't have either of those I bought the board from turretboards.com

I had to get F&T Caps anyway so I got them all from that site.

This was actually my first time building a turret board. I've always used eyelets before, and its a bit more interesting. I had to redo two or three turrets.
Last edited by XgamerGt04 at Jan 20, 2011,
#29
Is there anything special about turret board that you couldn't just use a normal material instead?


████████████
████████████








RBY CYOA
#30
This is normal board material. It is G10 board (FR4) so its fire resistant and non conductive. Its also a pain in the ass to cut. The website that I bought it from is actually Watts Tube Audio. He just calls it turretboards because he sells a lot of boards.

If you mean prototype board, you could possibly use that. FR4 is very good for electrical isolation so you don't have to worry about it becoming conductive. I wouldn't use prototype boards due to the very high voltages that you have in a tube amp though.
#31
Oh, well that makes sense, I was thinking about it's conductivity, but not really about as a fire retardant.


████████████
████████████








RBY CYOA
#32
Quote by MonkeyLink07
Oh, well that makes sense, I was thinking about it's conductivity, but not really about as a fire retardant.


Yeah the fire retardant part is a good thing. Some of the resistors on the board will dissipate around 3 watts of power at idle, like the cathode resistor for the power tubes. At maximum signal it may get over 4 or so for slight periods of time. So I want a material that won't burn or catch on fire.

Its also a nice shade of blue though
#33
So the update for today.

I got the F&T caps in so I've added them to the board. I didn't really see a reason to take a picture yet so I didn't.

I also changed a few of the resistor types. I changed the phase inverter anode resistors to carbon composition as well as the anode resistor on the EF86 and the screen grid resistor. If I had a 47k carbon comp I would also change the anode resistor for my parallel input stage, but I don't think I have one.

I may have to order a GZ34 though as I have been unable to find the one I had, and if I do that I'll also get a 47k carbon comp resistor.

I'm scrapping my current master volume idea, so the amp will be pretty loud. I'm fine with that, but I'm also going to make a cut out for a potentiometer in the back should I decide to add in VVR later. I've got enough chassis room that I should be able to do that later if I want to.
#34
Things are moving along now. I did some machine work on the chassis and have all of the top mounted components attached. I just have to drill the grommet holes for all of the transformer wires.





The rectifier tube is pretty gigantic, that is one of the NOS 5U4G tubes that I got from a shop here in town. The transformers look like they are touching but there is a good bit of space between the two of them, and before I finish everything up I will put a small dowel or piece of cardboard between the two for extra insulation.

I've got an autocad mock up of the layout from a flat piece of aluminum so if I make any more they will be precision laser cut.

I'm looking at possibly adding a switch to the back of the amp to switch between a normal output stage and a UL stage. If I recall correctly EL84's would rather have 20-25% taps and this has 43%.

The power transformer in the picture was defective, the entire primary was wound without a center tap, a center tap added, and the regular number of turns for the other side was added. This resulted in a 945VAC output unloaded instead of a 630VAC, but I'll find a use for it. I should get the replacement transformer by Wednesday.

I also got my Sozo caps, they better be worth the wait.

I should hopefully have a chance to finish machining the chassis this weekend and wire up the heaters and output transformer wires.
#36
Did the remaining machine work tonight on the amp. I have one hole that needs to be enlarged so I'll grab the unibit we have at work for that tomorrow. I have to get something to clean up the burrs on the amp so that is my current need. Once I get that heaters and all will begin being wired up.

Just as a note to anyone doing their own machine work, get a uni-bit and keep it well lubricated with wd-40. This makes quick work of the metal and leaves a pretty clean hole.
#37
All major machining, other than the mounting holes for the circuit board and ground lugs has been completed.







Sorry about some of the pictures being blurry, I'm still getting used to the camera on my new phone. It can take some really nice pictures at times (8 Megapixel), but it has this weird auto adjust/focus/mess you up mode.

You can probably see that the master volume has no control in it. For now I intend to leave the amp without one and see how it goes. It will get quite loud I know, but I tend to use an attenuator between the amp and speaker to handle that. The 4 ohm output will also be left unhooked for the time being, I can't seem to figure out a way to keep it from shorting out to the chassis even with the isolating washer. I may just have to take the faceplate off and file the hole some to fix that issue.

There should be enough room in the chassis to keep everything nice and tidy without too much trouble. Later tonight I'm going to wire the heaters, and hopefully get a closer picture of the inside of the chassis. I have a little trick for the heaters that might help some people out. Replacement power transformer should get here on Tuesday so I might have all wiring completed in the next two weeks, depends on how coursework and work goes.
#40
Quote by kurtlives91
Looking real good!


Thanks... I had a few issues with the face plates. The drill press I used was a bit wobbly so I had a few misaligned holes. I also had a few holes on the face plate that seemed. Bit off center. I'm going to take the caps to work tomorrow to determine where the outside foil is.
Page 1 of 3