Paul Ruby Amplifiers

vacuum tube amplifier faq
for diy amp builders
home gallery services parts yard sale contact links tech info
After lots of reading and typing on the AX84 BBS, I thought an FAQ would be a good idea. Thanks to all who contributed questions and answers! Before reading the FAQ, please read the safety info at Drifter amps. Also, many thanks to Kursad for much of the material below, both questions and answers.

I've finally created a turret installation tutorial.

My old pedal board page still exists, by the way. Also, I just saw this very cool pedal supply link posted at AX84.

Required Reading...

Musician Shock Hazard: We all need to know this.
Safety info at Drifter amps is of key importance.
Tube Amp FAQ at GEO: If it ain't here, it's there!
Aiken Amplification Has excellent DIY articles in the Tech Info section.

Questions...

I've fallen and I can't get up:

My amp is built. How should I power up for the first time?
How do I isolate an unknown problem (divide and conquer)?
Why isn't my amp stable (squeals, weird oscillations, etc)?
What causes hum and how do I reduce it?
What causes hiss noise and how do I reduce it?
B+ voltage is too high/low. Why? Should I bother? How?
How do I determine turns ratio of an unknown OT/PT?

Tone:

Why does my amp sound harsh?
Why is my amp "farting out"? Just what is amp farting anyway?
How do resistors & capacitors on the cathode affect tone?
How does location of the tone stack affect tone?

Theory:

What is Earth and Signal ground?
What is Star ground?
How do I wire my filaments to avoid hum?
How much can I exceed the wattage spec of a hammond OT?
When do I use twisted cables or shielded cables?
How do I select transformers for my amp?
How do I build an amp using published "typical operation point" data?
How do I bias a cathode biased amp?
How do I bias a grid (fixed) biased amp?
PCB vs turret board vs point to point (pros and cons)?
How do I get started building my first amp?
How do I position transformers so they don't interfere?
How do I determine speaker impedance?
How do I determine speaker polarity with 9V battery?

Components and Circuits:

What kind of hook-up wire and shielded wire is best?
How do I interpret the colors on resistors and caps?
How do I add a speaker-simulated headphone output or line out?
How do I add an effect loop?
How do I add reverb to an amp?
How do I add a pilot or standby light?
How do I build a dummy load or resistive attenuator?

My amp is built. How should I power up for the first time?

To get started you MUST read the safety info at Drifter amps. Tube amp voltages are LETHAL! The rules are well stated and I don't need to repeat them. Read them!!

The following is a simple checklist to follow for your first power up of a new amp. The goal here is to protect your valuable circuitry rather than have it go "poof" due to a simple wiring mistake. Unfortunately, we've all given into the temptation to "fire it up" as soon as the last solder joint is cold. That's a mistake. I lost a brand new Hammond 269EX. Talk about bursting your balloon of excitement!! Following this checklist should help get rid of some pins lurking next to your balloon. This debug procedure actually starts prior to completing the final solder joints...

The starting point is to do some basic circuit checkout PRIOR to soldering the transformers into the rest of the circuit. If you've already soldered in your trannies, take a minute to desolder the secondaries from the rest of the amp. Iron is expensive and it's worth some effort to protect it. Trust me!

Start with no tubes installed, MAINS UNPLUGGED, trannies NOT soldered in and get out your ohmmeter. Also remove the pilot lamp if you have one. Use your ohmmeter to verify that the connection points for the PT secondaries are open circuits. With no tubes and no pilot lamp, the 6.3VAC and 5VAC secondaries should be open circuits, no continuity. If you used a tube recto, then the HV secondary of the PT should also read as an open circuit. With solid-state diodes, there is continuity to the filter caps. So, you will see an initial low resistance that increases with time as the filter caps charge. Actually, some meters use a pulsed current to measure resistance, so use the continuity setting on your meter for this. Also, check your speaker jack with and without a speaker plugged in to make sure it isn't shorted.

If all is well, solder in your tranny's low voltage secondaries. Don't yet solder in the HV secondary. You can also solder in your OT secondary.

With no tubes installed and switched to standby, install your pilot lamp and use a 250mA slo-blo mains fuse. Click on the power for 1 second just to see the pilot come on nice and bright. This is primarily checking the 6.3VAC supply line is not shorted and is properly connected to the lamp. If the lamp did not come on, check to see if the fuse blew. If not, try another lamp and do the 1-second power thing again. If the fuse blows, there is a short on the PT or mains. If the fuse survives, then it's likely that the 6.3VAC is not properly connected to the pilot lamp. Use your AC voltmeter to check for 6.3VAC (actually more like 7VAC with no tubes installed) at the lamp and all the tube sockets. If you have a recto tube powered by 5VAC, then also check that you have 5VAC at the recto tube socket now. This will be up around 6VAC without the recto tube installed as yet.

In the following steps, B+ is going to go high. If your amp circuit does not include a bleeder resistor, you should install one at least for the debug phase. Solder a 100K, 2W resistor across the first filter cap (from B+ to ground). You can use four 470K, 1/2W resistors in parallel if you don't have a 100K 2W resistor.

Assuming you now have a nice bright pilot light, hook your AC voltmeter to the tranny HV secondary (which is still not soldered into the circuit). Turn the power on just long enough to get a reading to verify it is correct. You should get a value 10 to 20% higher than the rated output voltage. If you get a value less than the rating, shut down the amp and check the fuse. If you get a proper value from the HV secondary, power down and solder the secondary to the recto tube or diodes. Install the recto tube (if you have one). Now replace the 250mA slo-blo mains fuse with a 1A slo-blo (or whatever your amp calls for). Hook your VOM to the HV secondary again. BE VERY CAREFUL at this point, your B+ will charge up for this power up.

Before powering up LOOK AT YOUR RECTO DIODES AND FILTER CAPS!!! You ABSOLUTELY MUST have the diode and cap polarities correct. This is critical yet is a very common error. If either the diodes or caps are wired in reverse, you WILL destroy the caps, diodes and PT!!

Did you check the recto diodes and caps? Now walk away, watch some TV, come back in 20 minutes and LOOK AT THEM AGAIN! :) Sorry, but this is important.

Power up the amp and watch the following things as quickly as possible and roughly in this order: pilot lamp comes on brightly; HV secondary goes to nearly the same value as it did with the lines unsoldered; recto tube filament lights up right away; Recto tube plates do NOT glow red (overheat). If any of these three does not happen, shut off the amp immediately and find the problem by looping back to the beginning of this checklist. If these check out, power down. B+ MUST be discharged to safely continue messing with the amp guts. Your bleeder resistor will take at least a minute to bring B+ down to safe levels.

Now hook your DC voltmeter to B+ and ground. Power up again and check the B+ voltage. With no tubes installed, all the filter caps will charge up to the same voltage. If you do not have a choke, the voltage should be very close to 40% higher than the raw AC. Assuming your measured, say, 600VAC across the full secondary in the above steps, then each half is delivering 300VAC. B+ will be ~40% above this, which is ~420VDC. If you have a choke, the DC voltage depends on the size of capacitance prior to the choke. With no cap prior to the choke, the first filter cap should char to about 90% of the AC volts. With the same 300VAC on each half, the first filter cap should have ~270VDC. If there is a cap prior to the first filter, then the DC volts out will depend on the load current. With no load (as there is now without tubes installed), even a small capacitance will allow B+ to charge up all the way to the same as the no-choke case. As an aside, you can use this pre-choke cap to adjust B+ under load to some point between the 90% and 140% of the VAC.

KEEP IN MIND that every time you power up from now on that B+ will be high. In all the following stuff, allowing B+ to bleed is implied at each power down.

If all is well, check that B+ is at the OT on all the primary taps. Without the power tubes installed, the OT primary should be at B+. If not, something is wrong at the OT. Power down immediately and check for shorts of the OT primary. This should not be the case, however. An OT short should have been caught by now by checking B+ levels in the previous steps. This is really just a final sanity check to really make sure the trannies aren't going to be killed by any mistakes. Leave the amp on for a few minutes and make sure neither tranny is getting warm. The OT should stay dead cold and the PT should get just a little warm supplying the pilot lamp and tube recto. If your amp is fixed bias, make sure the negative voltage is being generated and check that it gets to the power tube grids and is adjustable with the bias pot.

OK, finally ready to put some tubes in. Power down and install all the signal tubes. You also need to hook up a speaker or dummy load for the OT. I actually suggest you hook up speaker rather than a dummy load, but preferably an old or less valuable speaker to get started. Turn volume and gain pots all the way down and tone controls to center. If your amp is fixed bias, turn it all the way down to start (most negative grid voltage possible. Power up and, again, watch for the following signs: pilot lamp comes on brightly; all tube filaments light up right away; tube plates do NOT glow red (overheat) this time paying attention to power tubes. If your amp is fixed bias, check power tube current and bias appropriately. Do this soon after power up to make sure all is well. You should be able to hear a little hiss or hum from the speaker. Of course, hopefully this is low level requiring your ear up next to the speaker to tell. If hiss and hum is loud at this point, there are problems. If there is dead silence, something is likely wrong, too. But, let's hold off on the AC signal debug until later.

Now is time for DC bias assessment. With one hand behind your back, measure and write down the B+ levels at each filter cap. Also write down cathode and plate voltages at all stages and also the screen grid voltage at the power tubes (if applicable). Compare all the DC voltages to those expected. In general, the triode gain stages should have ~1V on the cathode, 0V on the grids and ~1/2 B+ on the plates. If the DC voltages are not in the ballpark (within 50% of the general rule just stated), take some time to check the circuitry of the offending stage. If you have a cathode follower in the signal path, the grid should be at the plate voltage of the previous stage and the cathode should be about a volt higher.

The moment of truth arrives! Plug in your guitar and turn up the gain and volume knobs slowly and see how it goes. With a lot of luck, you'll hear the guitar sounding great with no hum or squeal. However, such a case is the exception and even then is typically reserved for experienced amp builders. To err is human and problems at this point are to be expected. Don't get discouraged. Jump down to the debugging stuff below...

top | home

How do I isolate an unknown problem (divide and conquer)?

An amp is composed sub-circuits. Even if the amp does not work, it's highly probable that most of the circuits work. Isolating a problem to a specific circuit is the primary task. If you've just finished building an amp, you really should follow a basic checkout procedure for the first power-up. Heck, even if this is an old amp that's gone south, I suggest you start by going through the basic checkout procedure for the first power-up. That process will take you all the way up through DC bias debug. It's worth it. Do it!

top | home

What causes hum and how do I reduce it?
How do I wire my filaments to avoid hum?

Hum can come from many places, all realated to the same root cause. Our amps draw power from your local utility company as AC, alternating current. At 50 or 60hz, the AC from the utility is in the audio range and you'll hear it if it gets into the signal path of your amp. So how does it get into the signal path? Hum enters the signal path from the heater (filament) wiring, poorly filtered B+, improper ground wiring or ground loops, right out of the air as radio frequency interference coming into unshieled signal lines, via direct magnetic coupling frm PT to OT, or from ancient amp wiring connecting tube filaments directly to the AC mains (no tranny). You'll need to tackle each of these in your amp to keep hum out of your ears.

Hum From Ground...

Hum from ground is usually a buzzy, raspy sound. The solution is to follow proper star grounding. The sound is similar to the RF source of noise getting into unshielded input lines or ground loops. Since it's a big task to convert an amp to star ground if it isn't already, you may want to clean up line shielding first. Also, be sure you don't have ground loops outside the amp, too.

Hum From Unshielded Input Lines...

Hum from unshielded lines is usually a buzzy, raspy sound. If your amp does not hum or buzz with the guitar cord unplugged (e.g. the input lines are hard grounded by the input jack), then the problem may be your guitar or guitar cord picking up the noise. Or, it may be your input lines running from the hacks to the tubes. You need to determine which it is first.

One quick way to see how bad your input wiring is is to simply plug a short patch cord into your amp's input and see how much that hums. This simply releases the grounding of the iput jack so you can hear what is getting into the line. Don't touch the tip of the dangling patch cord or your body becomes an antenna grabbing all noise from the air and sending right to your amp. The result is painfully loud buzz. If hum is as bad or worse than the hum with your guitar plugged in, the problem is probably your input wiring. If it is better than with the guitar, then the guitar is the likely source of the noise. Your guitar's pickups are very good at picking noise out of the air from other electrical equipment near by. Shut off stereos, PCs and especially PC monitors.

Still suspect the input wiring? The key thing is that the input stage is a very sensitive node for noise. This is especially true once past the 68K ohm grid stopper resistor in series with your input lines. As such, the 68K grid stopper really should be mounted right on the tube socket with ZERO lead length between the socket lug and the resistor. Your shielded wire then runs from the other side of the resistor to the input jack. The shield of the cable is only grounded at the jack end. Cut the shield off the tube end and use a hunk of heat shrink to make sure it doesn't short to anything near by.

Note, I use shielded input lines on ALL my amps and mout the 68K resistors on the tube socket as the first thing when getting rid of hum. I don't even bother to debug first. It's cheap, easy insurance against hum, noise or feedback (squeal). Just do it! Read more about hook up wire here. Hum From Filament wiring... Hum from filaments is characterized, typically, by low frequency 60hz (50hz outside America). It's not usually a "buzzy" sounding hum. On a scope, it will look reasonably like a sine way by the time it gets to the speakers.

If you have an ancient (er... vintage) amp that connected tube filaments directly to the AC mains, you very likely have a severe hum problem. As an aside, the tip off for this sort of thing is that the tubes have oddball numbers, like 50L6 or 35C5. The first number of the tube designation is the heater voltage. If your tubes are taking 50V or 35V, chances are they are wired in series and hooked right to AC mains. This is a real problem in old amps made as cheaply as possible so be warned when hunting around on ebay...

The only solution that I can ensure effective for old amps with heaters wired to AC mains is to convert the power distribution to modern standards. That gets rid of hum and, more importantly, brings the amp up to modern safety standards. This means adding a power transformer to the amp to isolate the mains, wiring the tube heaters in parallel and switching to standard tubes: 6L6, 5AR4, 12AX7, etc. Obviously, the choice to dive into a vintage amp like this is up to you. But, if your goal is a playable amp, this is the way to go. You want a collector's item? Stick it in a closet.

Hum from the heater (filament) is the easiest to take care of. No, you don't need DC filaments! The filament is powered directly by AC from the power transformer in your amp. The heater in the tube, of course, is hot and can source electrons, just like the cathode. If the heater is negative with respect to the cathode, electrons from the heater will be jumping from the heater to cathode. They way to stop this is to keep the heater from going substantially negative, or better yet make it positive, with respect to the cathode. This is easy to do!

Warning!! All amps have some form of DC referencing of the filament winding from the PT. If you're trying to get rid of hum in an old amp, you MUST find the existing DC reference connection and determine it's type. If you decide to use an alternate referencing method, the existing reference MUST be disconnected or you risk toasting your PT!! Read ahead to learn the methods so you know what to look for.

First, check to see if your filament winding has a center tap. If so, that's the lead we're going to play with to reference your filament voltage. If not, get two 100 ohm, 1W resistors, wire them in series connecting the two ends fo this to each of the filament lines. The node at the junction of the two resistors (the end of the resistors NOT connected to either filament line) is now your "center tap." It's not really a center tap but serves the same purpose for the remaining discussion...

The easiest thing to do is grond the center tap (CT) of the filament winding. Each filament wire will still go negative w.r.t the tube cathodes, but only about -9V at the negative peaks of the AC wave form. That's good enough in most amps to keep hum out of the signal path. Very high gain amps, however, will get some hum from this. Even though only a little is getting in, the high gain preamp amplifies the little hum to something audible. High gain amps need a positive DC reference for the filaments.

A better, but slightly more difficult, solution than grounding the CT is to connect it to some clean, positive DC voltage source. If the power stage in your amp is cathode biased with a nice big bypass cap on the cathode resistor, you're set! Just tie the CT to your power tube's cathode. There's somwhere between 9VDC and 35VDC at this node. It's just as free and easy as ground, so use it instead of ground if you have it.

If your amp is not cathode biased, then you need to find or make an alternate DC reference. Maybe your amp has 12VDC availble for realy switching or something. If you have it, use it. The last resort is to make a DC reference voltage. In this case, add a resistor divider from your preamp B+ stage. Wire on end of a 470k 1/2W resistor to preamp B+. Wire the other end to an 82K 1/2W resistor. The other end of the 82K resistor goes to gound. The junction of the two resistors will have between 30 and 50VDC. Add a 100uf, 100V filter cap from ground to this node, negative side of the cap to ground. Now, connect your filament CT to your new source of ~40VDC. Done!

Hum From B+...

Hum from B+ ripple is characterized by 120hz (or 100hz outside America). The key difference from filament hum is that the frequency is doubled by full wave rectification of the AC mains. This is a dead give away using an O-Scope for debug.

The filter caps on B+ in your amp have the job of smoothing out the DC B+ voltage in your amp. The plate resistors of the riode gain stages are tied right to B+ and ALL ripple on B+ becomes part of the signal delivered from the stage. This is why B+ goes through stage of R (resistor) C (cap) filters from the PT down to the input gain stages. By the time B+ gets to the input stages it must be DEAD FLAT. Fortunately, if you're debugging an amp or building from a known schematic, chances are that the issue is a leaky or otherwise bad filer cap causing B+ ripple. Note that the problem may also be excessive current draw from B+ somewhere. The filter cap values are chosen for the proper load. Excessive load will cause more ripple. You'll certainly want to keep this in mind if debugging a B+ ripple problem.

top | home

How do I determine turns ratio of an unknown OT/PT?

The first thing to do is check the transformer for shorts or opens. There should be no continuity between any primary wire and secondary wire or any wire to the transformer core. Check this with a DC ohm meter. Note that DC resistance of the windings is meaningless in determining the impedance. This is just to look for gross faults before hooking anything up to the tranny. Though rare, I have seen trannies with an explicit wire attached to the core intended to be grounded. The core and casing of the tranny should be grounded but that's usually taken care of just by bolting it to the chassis.

The way to determine turns ratio is to impose low voltage AC (~10 Volts) on the primary and measure the output voltage at the secondary. I happen to have a 12VAC tranny that I use for the test. Using low voltage for the test is important in case you hook it up backwards. For example, if you hooked 120VAC (e.g. raw mains) to the secondary, you'll end up with a few thousand volts on the primary. So, use 12V.... Please...

The impedance ratio is the square of the voltage ratio or the turns ratio. Divide input voltage by the output voltage and square the result. Note that votlage ratio and turns ratio are the same. For a push-pull OT, measure the impedance ratio across the full primary to secondary and use that for determining proper match to the tubes.

Sean Weatherford offers the following good advice to go along with this... It is a VERY good idea to put a fuse in series between the low voltage source transformer and the transformer being tested. If you don't want to use the fuse, put a light bulb of the appropriate voltage (6 or 12V) in series between the two. If the bulb lights brightly, the tranny is bad. The bulb will limit the current through the tranny to a non fire hazard level.

top | home

What is Earth and Signal ground?

There are two grounds: signal and earth.

Earth ground (green from mains) is for safety and reference. It references everything electrical, literally, to the earth, which we call 0V. Of course for safety, if anything other than 0V somehow shorts to anything that could touch a human, it also sinks current to blow the fuse rather than let it kill us. Earth ground connects to the chassis to reference the system to earth. There will be currents such as RF picked up by the chassis and such that flow and connecting the earth ground away from the star ground keeps this sea of RF, EMI crud coming in out of the signal ground (at least that's the hope). Of course, the earth ground must also be a VERY solid connection to the chassis for the safety aspect. In the event of a malfunction, that connection may need to carry enormous current to hold the chassis at earth (safe) voltage until a fuse blows (dead guitarists ruin a live show).

Signal ground is referenced to the earth ground via the chassis but the primary goal is to not allow signal current to pass through earth ground lines and visa versa. Within any given electrical thing like your amp there will be no current between the earth ground and the star ground (except maybe some EMI pickup by the signal ground lines, like the shield of your guitar cable). The chassis connection from star to earth is purely a referencing connection. But again, it's also for safety, as that same connection will carry huge current from the circuitry to the earth in the event of malfunction. So, the single connection of the star ground to the chassis should also be very solid. Key thing is SOLID.

The best physical place to make the connection from the signal ground to the chassis is debatable. But, to avoid noise getting into your signal path, the best place is at the input jack. This is because your guitar and guitar cord shield make a perfect antenna for RF noise from neon lights or dimmer switches. You goal is to get that noise from the guitar shield onto the chassis as soon as it enters your amp. Else, the noise will continue along to the input tube and gets into your signal path. You may have seen discussion of this online. For existing amps that don not have the input jack grounded, some folks suggest connecting a cap on the input jack from the shield to the chassis to serve the purpose of getting RF coming in on the guitar cord shunted to the chassis. This is a good suggestion for existing amps. But, for a DIY amp, you can instead make your single chassis connection at the input jack.

OK, that optimizes our individual piece of equipment. But, it leaves open the possibility of ground loops when stringing multiple things together. Each thing has it's own mains ground with a reference connection to signal ground. So, the mains power line and the line cords between items form loops. Some amps have a "ground lift" allowing the quick elimination of ground loops by removing the mains ground connection. This SUCKS. Throwing that switch eliminates the safety of mains ground and is NOT needed to solve ground loop problems. The argument is that "if there was a ground loop, then there must be another path to earth, so there's no safety issue lifting the ground." That's crap. Do you trust the cheesy patch cords between your foot pedals to protect you against a B+ short to chassis?? So what's the solution? Think about your mains as extensions of the star ground principle. Keep your equipment physically localized and keep all the mains cords running parallel and tight to each other (there's no surface area to any ground loops and thus no noise pickup). Run them all back to one group of sockets on a wall rather than plugging things in willy nilly around the room. For pub gigs, this is an easy policy to follow. For the big guys, use isolation transformers and run balanced XLR to break the ground loops.

top | home

What is Star ground?

Star ground is covered very well at Aiken Amplification. So, read that as well as my stuff.

Some folks ask about grounding rules or best methods. The following statements cover the basic guide lines.

. One and only one connection from circuit ground to the chassis.
. No Ground Loops. Only one path should exist for any ground current.
. Tight, isolated loop for PT, recto and 1st filter cap.
. Ground circuits to their respective B+ filter cap negative terminal.
. Ground your OT secondary and the OT itself.

Note that there's nothing particularly better about a star or buss or pot ground. As long as the implementation follows the above guidelines it'll work fine. A star ground is often suggested for new DIY folks because it's idiot-proofed against ground loops. But, a buss or pot ground can meet all the guidelines when implemented with care. Note that a chassis grounding scheme can also meet the guide lines and function very well. However, I personally avoid chassis grounding becuase it's easy to end up with an issue that is difficult to fix once an amp is complete. It's easy to implement a star or buss ground, so that's what I do.

Star grounds are not directly connected to the chassis. Only ONE connection to the entire chassis should be made. The actual connection point is discussed above.

The physical location of each star ground is not very important. But their electrical location in the circuit is very important. Your board layout should have the following basic layout running either right to left or left to right:

PT-HV, recto and first filter cap; Power amp and second filter cap; final preamp stages or PI and filter cap; initial preamp stages and filter cap.

In solid state recto amps in particular, it is very important to have the PT-HV, recto diodes and first filter cap in a very tight loop off to the far end of the chassis. There are very large ground current spikes in this loop and the loop itself forms an AM transmission antenna. So, you want this loop far away from your gain stages and with the smallest possible cross-sectional area. This is why wreck's and clones have the power supply on a separate circuit board on the far end of the chassis. In my latest builds, I mount a small board directly to the PT mounting bolts including the recto and first filter cap. A single, heavy wire then runs from the first filter cap negative terminal to the second filter cap negative terminal. That's the spot for a primary star or the power-amp end of a ground buss.

I Like to have a star ground for each of major section and I have them on the board where the supporting components are located. The number of star grounds is set by the number of filter caps. Every time you add a resistor and filter cap to the B+ supply line, just consider the ground side of the filter cap as another star ground in the path.

Given this, you should have a group of turrets (or eyelets) on the far right (or left) that form the first star. Then a group of turrets for the power stage, etc...

Again, the number of filter caps determines the number of individual star grounds. One wire connects from one to another IN THE SAME ORDER as the B+ is segmented. ALL the cirucits supplied by a particular B+ section are grounded at that section's star.

Now, there's almost always a volume pot between the major section where it is fed by the previous and feeds the next. Which to you ground it too? Well, it doesn't matter if you've done a good job with the grounding. But, I ground POTs to the section sharing a DC path to ground, which is always the next stage.

Form a metal picture of this: Several clusters of turrets along the bottom of your tag board, each forming a star ground with one wire connecting each.

If you have followed this so far and understand, read on. If still confused, go back up and reread until you have the metal picture and understanding firm.

Now alter your mental picture and you'll know how I do the grounds in my amps. Picture each of the clusters of turrets spread out so that there's just one long line of ground turrets along the bottom of the board with one lone wire connecting them. Now there's no clear physical distinction to the individual stars, BUT, if the board is well laid out, the components will naturally follow the proper order of grounding. You just need to have the correct mental picture and run ground lines from your pots and such to the correct location along this line. Also, the filter caps per stage are ALWAYS wired in the path closest to the HV supply and the components of this stage are away from the HV supply, relative to the filter cap for the stage. In other words, I stretch out my star grounds into a buss. On some amps, the buss is on the board. In others, it is on the pots.

As stated previous, I make my single connection from circuit ground to the chassis at the input jack (or near it). This happens to be exactly what is happening when using a pot ground buss. Think of the pot buss as one big chassis connection and one end of it is at the input jack.

Finally, you need to ground your OT secondary. I'm often surprised to see folks online suggesting otherwise. First and foremost, an ungrounded OT secondary means your speaker lines will rise to B+ in the event of an OT fault. That's a safety issue and is not acceptable. Second, if your amp has NFB the OT secondary must be grounded at your power amp star point for proper ground referencing. Finally, even if you don't have NFB an ungrounded OT secondary will be capacitively coupled to the primary and can thus have a high AC voltage swing. This high AC voltage swing can make its way back to the input path and cause squeal. And, the OT core should be grounded to the chassis via the mounting bolts.

top | home

What kind of hook-up wire and shielded wire is best?

A general guideline I suggest for amp building is shielded wire to tube grids and to feed/return from control pots. This is good practice to protect against noise. If you really know what you're doing, this can be avoided with excellent circuit design, layout, star grounding, tranny placement, chassis grounding, etc... But, I consider myself knowledgable and I do not take such risk. All my amps have shielded wire to tube grids and control pots. It's just plain good practice.

Some folks argue that shielded wire "kills the top end." If that were so, then how can you afford to use an 18' shielded cable from guitar to amp?! The capacitance added by short lengths of shielded cable inside your amp is very small. Of course, if you have a lousy layout with long cable runs and you choose the wrong cable with high capacitance then, yeah, you'll kill the highs. The key thing: short runs via good layout; low capacitance per foot; flexible, easy to work with.

The cable I use is great. Very thin and easy to use. Low capacitance per foot. The exact numbers are:

The important part of the number above is RG-174U. This is the stuff Hoffman Amps sells. Go to the "Board Building Parts" section of the online store. Other folks have used George L cable. That's the kind used to make patch cords. It's good cable, low capacitance and will work fine. But, it's more expensive and thicker (so may be more difficult to work with). George L cable is also available at Hoffman Amps on the "Jacks/Plugs" page. Heck, if you have some old guitar cables, CD-ROM cables from a PC or RCA audio cables, you can cut those up, too...

I've recently started using TEFLON shielded wire. It's great!! Very thin, easy to use and impervious to soldering iron bumps. You can even solder directly to the shield, right against the inner insulation. I buy all my teflon wire from http://www.apexjr.com/. Do a BBS search for "use this wire" to read more...

For generic hook-up wire, there's plenty of opinions. The key thing is that you need high voltage, 600V, insulation. Beyond that there isn't much else to be particularly worried about. But, here's my own opinion. I use teflon insulation, 18 gauge, stranded and tinned wire in a bunch of colors. Teflon insulation is great for DIY because it won't melt when bumped with a soldering iron, which you WILL do. 18 gauge is more for ease of use and keeping the wire where I want it. Smaller gauges of stranded, teflon tends to just go where it wants and is harder to use. Some folks use solid core because it stays where you bend it better. Heavier gauge stranded core will do the same. Either way is fine. But, I really do suggest teflon.

Teflon, stranded, 18 gauge is also good wire for making the Tone Brewery cables. It's flexible enough for a chassis-to-chassis cable, doesn't nick or cut easily and won't melt if it touches a hot tube. In fact, given the cable is external to the chassis, I think teflon is really a requirement for safety.

top | home

How do I interpret the colors on resistors and caps?

This one is easy. Ampwares has the codes on their site.

If you use a mac, checkout Lyndon Stolton's Resistor Value Calculator.


top | home
content & layout © copyright paul ruby 2001, 2002, 2003, 2004
all rights reserved