REFERENCE: Power Transformer Testing

Testing an Unknown Power Transformer

Every now and then I end up with a transformer on the bench that I know came from valve gear, but I did not take enough notes when I pulled it.  That was the case here.  I knew this one came out of an old Zenith radiogram, but that was about it, so the job was to work backwards and identify the windings properly.

Naturally, anything involving a power transformer deserves caution.  Unknown windings, mains voltage, and potentially lethal secondary voltages are not something to rush into.  It is worth stopping and thinking before connecting anything, or even attempting this.  

What I Knew Going In

In this case, I knew the donor unit ran on 240V and used a pair of cathode-biased 6V6, along with preamp and phase inverter valves, plus a valve rectifier.  

That gives a rough idea of the sort of transformer it is likely to be, but only in a broad sense - it does not confirm the actual heater voltage, HT voltage, or current capability, nor whether it actually works to begin with.

What a Typical Valve Amp Transformer Looks Like

A typical valve amp power transformer usually has a primary winding connected to the wall supply, one or more low-voltage secondary windings for heaters, and a higher-voltage secondary for the HT supply.  

Where I Start

First up is a visual inspection.  Before I go anywhere near it with power, I want to know whether there are any obvious signs of trouble - burnt areas, dodgy repairs, brittle or damaged insulation, loose leads, corrosion, or anything else that makes the transformer look questionable.  Basically, I am looking for any reason not to go any further.  Now is not the time for blind optimism.  

If it passes that first check, I start with the multimeter on continuity, just to sort the loose leads into groups that appear to belong together.  Often the original colour coding will help here too, if it is still clear enough to trust.

A low-resistance winding is often a heater winding, and a higher-resistance winding is often HT, but resistance on its own does not prove the voltage.  Winding resistance mainly reflects the wire size and the number of turns, and practical transformer behaviour also depends on inductance, losses and loading.

So I use resistance readings to sort and narrow things down, not as final identification.

Warning: It is also worth checking each winding to the transformer core, frame or bell ends.  Normally, I want to see open circuit there. 

Spotting a Likely Centre Tap

Where resistance readings are especially useful is in spotting a likely centre tap on a HT winding.

If one winding measures, say, about 74 ohms end to end, and about 38 ohms from each end to a third wire, that strongly suggests a centre-tapped winding.  Each half measures about half the total, which is exactly what you would expect.  A large resistance also suggests that it is not a heater wire.

In the case of this transformer, the HT is the only winding with a centre tap.  The 6.3v heater does not have one, which is not uncommon on older transformers.  Trying to confirm a heater centre tap using the method is not reliable, due to the very low resistances that will be involved.  

The Better Test

Once the winding groups are sorted out, the next step is the one that really matters: apply a small known AC voltage to one winding and measure what appears on the others.

That is the proper way to identify an unknown transformer, because the measured voltage ratios directly reflect the turns ratios. 

It is also the point where you stop guessing and start mapping what the transformer actually does. If you put a known AC voltage on one winding and get a much higher voltage on another, you are clearly looking at a step-up relationship. If you get a much lower voltage, it is a step-down relationship.

This is the step that tells you which winding is the likely primary, which winding is the HT secondary, whether a winding is centre tapped, and what the heater winding voltages actually are.

That is a much firmer basis than ohms readings alone.

Bringing It Up on Mains (really, really carefully)

This is the point where things can get dangerous in a hurry.

Once I think I have identified the primary, I power it up and check the secondary voltages with no load connected. I use a Variac with a current meter, start with low voltage, and keep a close eye on the current draw. Some no-load current is normal, but I do not want to see anything excessive.

This is a completely hands-off test. The leads need to be properly secured on a clean bench before anything is powered up.  I want a clear view of the multimeter and the current meter, and I do not want to be moving probes or wires around while it is live.  

The Variac starts at zero and switched off.  Then I bring it up slowly, while watching, listening, and paying attention to how the transformer behaves.  Once I have the readings, the Variac goes back to zero and gets switched off.  Then I double-check that before touching anything.

At this point, I am not expecting the transformer to draw zero current  - some no-load current is normal. 

What I want to see is sensible voltages and sensible behaviour. 

If the transformer stays reasonably quiet and cool, and the secondary voltages look believable, that is a good sign. 

If it draws obviously too much current, gets hot quickly, or starts buzzing and complaining, I would stop immediately and reconsider all previous steps and the safety of the transformer. 

What the No-Load Voltages Tell You

The unloaded voltages are important because they confirm the winding functions far more reliably than resistance ever can.

A low-voltage winding that sits around the expected few volts AC is very likely a heater winding. 

A winding that gives you two equal voltages from each end to a centre lead is very likely a centre-tapped HT winding. 

Once that is established, you can start thinking sensibly about whether the transformer is suitable for the sort of amp you have in mind.

There is also a practical reason for checking the transformer unloaded before going any further.  Transformer behaviour changes under load. Whitlock points out that when a load is connected, the secondary current opposes the excitation flux and the primary draws additional current accordingly.  So if all you want at this stage is identification, unloaded testing keeps the picture simpler.

A Couple of Easy Mistakes

One easy mistake is to overstate what a resistance reading tells you.  It can suggest a heater winding or suggest a centre tap, but it does not confirm the operating voltage.

And one more: if the transformer came from old or unknown equipment, do not assume the original primary was intended for today’s wall voltage, or is in fact even vaguely suited to your wall voltage.

Bottom Line

For me, the sensible order is:

1. visual check
2. continuity,
3. resistance,
4. small AC voltage test to confirm,
5. then a careful mains bring-up and unloaded voltage check.

That way you use each test for what it is actually good at.  Continuity tells you what belongs together. Resistance helps sort likely windings and likely centre taps. AC voltage testing tells you what the windings really are. Then a careful live test confirms whether the transformer behaves sensibly at its intended input.

Resistance gets you close. Voltage confirms it.

⚠️ Caution — High Voltage:

Transformers can produce dangerously high voltages, even when powered by a seemingly low-voltage AC source. The windings in tube amplifier transformers are capable of stepping voltages up or down significantly. Always assume exposed leads may carry lethal voltages, and take proper safety precautions. Use insulated tools, never work on live circuits unless qualified, and ensure your test setup is isolated and safe.

GEC: Germanium Signal Transistors (1971 Semicondcutor Handbook)

From the General Electric semiconductor handbook, 1971

TONE STACK: James

The James tone stack is one of the more common bass/treble tone controls. You most often see it in hi-fi equipment, or in guitar amps that sit outside the usual Fender and Marshall mould - although both Fender and Marshall did use similar circuits on rare occasions.

Probably the best-known guitar amp examples are Orange amplifiers and some Ampegs.  Closer to home, Australian builders, such as Goldentone and Vadis used this style of tone control.  Marshall used it in the Artiste, and Fender used a related bass/treble arrangement in the blonde Twin, model 6G8.

It’s hard to mention the James circuit without mentioning Baxandall - same same, but different, so I won’t get too bogged down in that here.  The original Baxandall circuit was an active hi-fi tone control, incorporated into a gain stage.  The James circuit came a few years earlier and is passive: 1949 for James, compared with 1952 for Baxandall.

The James is a simple and useful design. It mostly leaves the midrange alone, while giving reasonably independent control over bass and treble. Think of it as two tone controls working in parallel, rather than a Fender-style tone stack where the controls interact heavily.

For this example, ignore RIN and RL. These simply represent the source and load connected either side of the tone control.

ORANGE, GRAPHIC MKII

link to TSC for frequency response

FILMOSOUND 385

Despite the impossibly difficult to read schematic with dual 3meg pots on a single tone control, this is actually a James eq.    You probably can't find 3m dual gang pots anymore, but you can adjust values to a 1m dual pretty easily - just divide the resistance by 3, and multiply the capacitance by 3.  

Link to TCS

AMPEG B15

FENDER TWIN, 6G8

note: this uses a tapped pot.  i.e. it has fourth lug - good luck finding one of those.

MARSHALL ARTISTE

MARSHALL: 1974X AKA the Marshall 18w one knob tone controls

Another simple one knob tone control, this time from the Marshall 18w amplifier, or the model 1974x, whatever you prefer...   

Marshall used two different tone controls in the amp.  A lot of folks build the 18w lite, which is just the normal channel of the amp, and they often use the trem channel tone control as it has a bit more range to it (18w lite II, it's a great little amp).  The trem channel is basically a Fender one knob tone control using 500k pots.

Normal channel - a little less bass than the trem channel, no top boost

Trem channel - more bass than the normal channel, a bit of top boost

Combined

FENDER: One knob tone control

This is the classic single-knob tone circuit made famous by a number of Fender amps—most notably the 5E3 Tweed Deluxe.  Despite not inventing it, Leo must have thought that he nailed it when he first dropped this into a guitar amplifier: it’s simple, clever, and very musical.  Despite using only a handful of parts, it offers surprisingly flexible tone shaping. Best of all, it barely saps any gain—there’s virtually no signal loss, which is rare for a tone circuit.

So how does it work?

At full clockwise, the 500p cap connects directly between lugs 3 and 2 of the volume pot.  That’s effectively a bright cap, allowing treble to bypass the pot's resistance and reach the output.  Turn it counter-clockwise, and you’re grounding high frequencies through a 5n cap, taming the treble response. It’s a beautifully balanced design.

Right now I’m mapping how different pot values affect the response.  Why bother?  Well, on some old amps I refurbish, the original pots aren’t always 1M.  This might be a handy reference in future, as I'm sure to forget at some stage.

All my tests use a 22nF coupling cap from the previous gain stage, and both the 500pF and 5nF caps in the tone circuit (though 470pF and 4n7 are perfectly valid substitutes).

1meg volume, 250k tone

1meg volume, 500k tone

1meg volume, 1m tone (5E3 sytle)

Big Muff Fonts - Vintage

Vintage big muff fonts, these are generally available for download.  Please note that EHX is known to be very protective of their brand. 

Arnold Boecklin

Devinne:  sustain, volume, tone

Futura:  on, off and made in U.S.A

Skjald:  Electro Harmonix

ITC Pioneer

Not exact, but pretty damn close.  The M needs a slight drop on one side, which would not be hard to do with fonts converted to vector.

CTS Speakers

Chicago Telephone Supply (CTS) originally built drivers for telephone network signaling and tone generation. By the late 1950s they expanded into loudspeakers, supplying the booming tube-amp market. 

Their speakers, often marked with the EIA code "137," were used extensively by Fender, Ampeg, and many other amplifier companies from the 1950s through the 1970s.

Early CTS speakers used alnico magnets (square-back frames) prized for smooth breakup, while later ceramic-magnet versions (round frames) offered tighter lows at the expense of some warmth, but this is of curse a little subjective and varies with a lot of other factors.

Old CTS speakers do pop-up on eBay and Marketplace in Australia, and it can be hard knowing what you're looking at - are they hi-fi or guitar speakers?   The advertisement below might help.

REFERENCE: Output Transformer Testing

How to Test an Output Transformer

Got an old output transformer with no markings or a transformer in an amp that your are not sure about?  Maybe one used for a 100v line system?  Don’t toss it just yet — with a few simple tests, you can figure out what it is, how it’s wired, and whether it’s usable for your amp build. This guide walks you through the process using a multimeter, a low-voltage AC source, and some basic calculations.

Do not do this with the amp plugged in, with the power tubes in, or with the speaker connected.  Remember - dangerous voltages in amps...  make sure the filter caps are drained of DC before touching anything.

1. Identify the Windings

If it's in the amp, this will be simple, but otherwise, a multimeter can be used to find the different windings:

• Primary winding (plate to plate): typically has a resistance of 50–300Ω.
• Secondary winding (speaker side): usually very small, maybe even less than 1Ω.

If there's a centre tap on the primary, it's likely a push-pull transformer.  The measurement between the centre tap and the two plates is half the total plate to plate value.

2. Measure the Turns Ratio

Apply a known low AC voltage (like 1V) to the secondary and measure the resulting AC voltage across the primary. Then calculate:

Turns Ratio = Vprimary / Vsecondary

Example: If you apply 1V AC to the secondary and measure 22V across the primary, then the turns ratio is 22:1.     

You can also do the reverse, and apply a larger voltage across the primary (like 20v) and measure the secondaries.   This is probably safer, as the AC voltage will be stepped down, not up.  I normally use 15v, cause that’s heading towards the upper limit of one of my signal generators.     

3. Estimate the Primary Impedance

Once you know the turns ratio, you can estimate the reflected primary impedance using:

Zprimary = (Turns Ratio)2 × Zsecondary

Example:

Zprimary = 222 × 8 = 3,872Ω

4. Check for Shorts or Opens

Use a multimeter to confirm continuity across each winding. Also make sure there's no continuity between any winding and the core or mounting hardware. If there is, the transformer may have an internal short and should not be used.   Also watch out for any high values - if it reads a few meg, that’s a bad sign. 

5. Check Frequency Response (Optional)

If you have a signal generator and oscilloscope, you can sweep an audio signal through the transformer and observe the output for frequency roll-off or resonant peaks.  Note that a low e on guitar is about 80hz and most speakers drop off pretty fast after 5kHz.   A lot of Hammond transformers intended for guitar amps list 75hz to 15kHz as their frequency response

What you are looking for here is significant changes to the size of the sine wave displayed on the scope.  i.e. when the wave decreases in size, the AC voltage is dropping, meaning that the frequency response is changing by a noticeable amount.   It's not as accurate as a spectrum analyser, but it's close enough for a guitar amp, considering the kind of speaker and cab that it will be hooked up to.

6. Visual Clues

• A center tap on the primary usually means it’s for a push-pull amp.
• A gapped core is a sign it’s for single-ended operation.
• Larger transformers with thicker wire usually handle more power and lower impedance loads.

Reference: Typical Load Impedances

Tube Configuration	Typical Primary Impedance
6V6, single-ended	5k – 8kΩ
6V6, push-pull (pair)	8k – 10kΩ
6L6, single-ended	3.5k – 5kΩ
6L6, push-pull (pair)	4k – 6.6kΩ
EL84, push-pull (pair)	~8kΩ
6L6, push-pull (quad)	1.7k – 2.2kΩ

Tips from the Old Masters

“A 1-volt signal applied to the speaker winding should give you a good idea of the turns ratio just by measuring the voltage on the primary.”
— Jack Darr, Electric Guitar Amplifier Handbook

The Radiotron Designer’s Handbook also provides detailed transformer theory, including reflected impedance formulas, equivalent circuits, and response testing using low-voltage AC signals — all foundational concepts that support this method of transformer identification and testing.

From the Radiotron Designer’s Handbook:

Transformer impedance is reflected from secondary to primary in proportion to the square of the turns ratio:
Reflected Impedance = Load × (Turns Ratio)2

The handbook also discusses transformer response across frequency and how to analyze bandwidth using -3 dB points, supporting AC test methods for audio applications.

Source: Radiotron Designer’s Handbook, 4th Edition

⚠️ Caution — High Voltage:

Transformers can produce dangerously high voltages, even when powered by a seemingly low-voltage AC source. The windings in tube amplifier transformers are capable of stepping voltages up or down significantly. Always assume exposed leads may carry lethal voltages, and take proper safety precautions. Use insulated tools, never work on live circuits unless qualified, and ensure your test setup is isolated and safe.

BREADBOARD: How To

Breadboards are one of the most effective learning tools when it comes to electronics.  Not only are they great for prototyping and testing circuits, but they also offer a hands-on way to fine-tune and troubleshoot designs—perfect for testing guitar pedals.

In this post, I’ll walk you through how I use a breadboard.  While this approach may not be for everyone, it’s a method that’s worked well for me, and it could help you get more comfortable with circuit building. 

Testing transistors for the Colorsound hybrid fuzz box.  This cheap breadboard doesn't have the power rails connected as per normal - I had to jumper them.

HOW TO USE A BREADBOARD (For Guitar Pedals & Electronics Beginners)

So, where to begin?  You will need:

• a breadboard
• power supply - bench top, battery or pedal power supply will do
• components to play with
• jumpers, which can be solid core wire, or you may have a bunch supplied with the breadboard
• input / output jacks, ideally with an on/off switch, or better yet a test box

You also need to be able to read a schematic and have some basic understanding of how the circuit works.

PLANNING TO BREADBOARD A PEDAL

I don’t typically pre-plan my breadboard layouts.  I’ll take a quick look at the schematic and then dive right in.  Early on, I tried using DIYLC (software for creating vero layouts) to map out my builds, but I found that breadboarding is a hands-on, three-dimensional process.  Once I started placing components on the board, my plans usually went out the window—so I’ve learned to just dive in and adjust as needed.

SO HOW DO THEY WORK?

Breadboards are relatively simple. They consist of rows and columns of holes that are electrically connected in a grid pattern. Along the edges of the board, there are vertical power rails—one for positive voltage and one for ground. You can buy fancier breadboards with additional features, but they all work in roughly the same way: you place your components and make connections using jumpers.

COMPONENT PLACEMENT ON THE BREADBOARD

When I build a circuit on a breadboard, I aim for a logical, linear layout.  I don’t try to make the circuit as compact as possible.  Instead, I leave a little bit of space between components to allow for easy adjustments and troubleshooting.  Generally if you're breadboarding, you will be swapping components in and out and this means leaving room for clumsy fingers.

For simple circuits, the main goal is making sure everything fits on the board and is properly connected. Once you get into more complex builds, you might need to think about space a bit more, but for testing pedals, spaciousness is often your friend.

For some circuits, I'll build most of it on vero, and then test some components on the breadboard.  There might be a particular section that I want to play with and test, without having to build the entire circuit on the breadboard.  I often use little clamps for this.

TESTING THE CIRCUIT

Usually I feed the circuit a sine wave from my signal generator and test that it's working as expected on one of my scopes.  You could easily use an audio probe to do this as well.

For larger or more complicated circuits, I test as I build, making sure that each stage works.

After it's confirmed to behave as expected, I usually plug in a guitar and run it to my amplifier, as real world testing is the always the best.  Just need to be a little cautious here - I have an amp that I'm not attached to for this purpose.  I've never killed an amp in testing, but there's always a chance right?   There is often a bit of popping and unsettling noises when touching parts on the board - just be aware of this.  

When testing, I usually have a selection of parts to swap in and out to see how they effect the sound - in some cases I'll use a pot instead of a fixed resistor to sweep values.  When I find a setting that I like, I'll take the pot out of the circuit to measure the value and replace it with the nearest value fixed resistor.

For transistors and caps, it's just a case of swap them and see what happens.  

I usually have a few pots sitting in the parts draw with solid core wires soldered on for breadboarding.  It's handy to be able to swap a few around to see what works better for you in the circuit.

TROUBLESHOOTING

OK - you've put all your parts on the breadboard and it looks fine, but it doesn't work.  What next?  Same as testing a regular circuit, check your voltages and probe the circuit to see where you went wrong.  

It's often a component leg in the wrong hole, or a possibly a poor connection somewhere.  Poor connections can be a thing with cheap or dusty boards, quite often on the power or ground rails.  I normally tap and wobble parts to see if they make noises, maybe reseat transistors.  In short, if it doesn't feel secure, it probably isn't a good connection.  Although this could also be a sign that I need to spend more on my breadboards instead of getting cheap ones for a few dollars from China...

PRE-BUILD TESTING

In some cases I breadboard the entire circuit that I plan on making on a breadboard first, using the components that will be used in the final build.  This is usually only for circuits that I know will be tricky to tune.  Example: a MKI Tone Bender.   In some cases I’ll just use the breadboard to test the transistors and not worry about the rest of the circuit.  

OPTIONAL:  COMPONENT LABELS

If you're like me and don't have the best eyesight, small masking tape labels on the legs of components is sometimes helpful.  Especially when you start pulling out small metal film resistors and leaving them all over the bench.  

I can read 4 band, 1w carbon films easy enough, 1/4w metal film is a tad tricky for me.  Sometimes I avoid using 1/4w metal film for this reason.  Same can be said for some caps - greenies are sometimes hard to read.  I sometimes write the value on with a marking pen.

This humble little transformer is a favourite valve amp OT among some DIY folks in Australia, as it's cheap at only $27 and readily available from Altronics.   It's sold as a 20w 100v line transformer, but it also works really well as a 8k P-P output transformer.   Based on real world testing, it's thought to be closer to 15w, rather than 20w.

Here’s how you connect it: just follow the primary side and pick an output impedance that works for you (4, 8, or 16 ohms).

P: 1.25w 
CT:  5w
P: Common

These are perfect for replacing transformers in lower-power amps. I’ve even used one in an 8w push-pull PA head, and it sounds great.  Some old PA heads have output impedances that were meant for a bunch of 100v line speakers, not a single 8 or 16 ohm speaker. 

There's a few more versions available in the line with higher and lower wattages, but this one is probably the most commonly used.  The M1130 is larger, and with an impedance mismatch on the secondaries, it can be taken up to a 4k plate to plate load suitable for a pair of 6L6 not running too hot - they're used in the valve Heaven Lamington II.

Further reading:  Ozvalve amps

ALTRONICS M1120 OUTPUT TRANSFORMER

Frequency Response: 30Hz - 20kHz ±3dB
Secondary Taps: 4, 8 & 16 Ohm
Power taps: 1.25W, 2.5W, 5W, 10W, 15W, 20W
Frequency Response: 20Hz - 20kHz ±3dB

HOW IS THIS CALCULATED?

Note the transformer’s rated voltage (100v). Square this voltage: 100 squared is 10,000.

Divide 10,000 by the wattage on each tap. The smallest wattage has the largest P-P value.

10,000 divided by 1.25 equals 8,000. Repeat for the rest of the taps.

To calculate the centre tap (CT), divide 8,000 by 4 and match it to the relevant tap. In this case, the CT is the 5w tap.   

If for some reason 4K P-P was required, then connections would be as follows:  P 2.5W, CT 10W, P common.   

ALTRONICS M1120
PRIMARY	Watts	P-P Value	CT
6 (P)	1.25	8,000	2,000
5	2.5	4,000	1,000
4 (CT)	5	2,000	500
3	10	1,000	250
2	15	667
1	20	500
C (P)

Which is conveniently confirmed by the specs on the box.

and here's a youtube clip explaining the theory

 

UNCLE DOUG: Valve Amplifier Explainers

I could just bookmark these in my browser, but hey...   here they are.   If you struggle with complex explanations of how valve amplifiers work, here's a collection of easy to understand valve amplifier basics from Uncle Doug. 

Don't let the pets and other distractions put you off - there's really good information in these videos.  

Phase Inverters

Output Transformers

Power Transformers

Power Supplies

Triodes & Pentodes

Single-ended amplifier biasing

Double ended biasing (push-pull)

Quad biasing

Cathode Bypass Capacitors

Negative Feedback

Tone Stacks

Resistors

Introduction

Power & wattage requirements

Grid leak, grid stopper

Cathode bias resistor

Plate resistors

Capacitors

How to Measure Output Power 

ENCLOSURES: Cutting Holes for a Battery Draw

I know I could probably pay Tayda to do this, but they don't always have the enclosures that I want, and part of me likes going down the hillbilly route with a hand drill and a file.

Here's my process for cutting out holes for battery draws.  It's actually pretty easy to do.  

STEP ONE:  MARK OUT

So first things first - make sure that you know where you plan on putting the foot switch and other parts, to make sure there's room for the battery draw.  

Use the battery draw as a guide to trace out the length / width etc, using a square.   

STEP TWO: DRILL THE OUTLINE

I use a small drill bit to get as close to the inside edge of the line as I can, trying to keep the holes quite close together.  I don't have a fancy pedestal drill, so this is all done with a regular battery powered hand drill.

STEP THREE: REPEAT, BUT BIGGER

Then I use a larger bit to drill around the smaller holes, and in some cases this breaks through the edges of the smaller holes, creating a bigger gap.

Once this is done, metal shears finish the job - it's usually doesn't take much effort at all.

STEP FOUR: GET BUSY WITH THE FILE

As the title suggests, time to get busy with the file.  Check progress as you file, because aluminium is quite soft and the file will do its work quickly.  Check the hole with the battery case occasionally, just to make sure you are on track.  Watch the corners too - most of the hole can be the right size, but if the corners aren't, it will not fit, leading you to remove more material than required.

STEP FIVE: DRILL MOUNTING HOLES

I like to drill the holes for the battery holder at this stage, using the case itself as a guide.

NEARLY THERE

After drilling out small pilot holes for pots and jacks etc, it's time for the stepped bit.  Keep an eye on the stepped drill bit, as they tend to clog and this can cause issues.  I find slow and steady drill speed with a little pressure works for me.

I also like to double check the size while drilling, using the components that will go on the enclosure, to make sure that I don't accidentally overshoot it.  

Hopefully it looks something like this when you're done.  I've given it a light sand and a wash in warm soapy water by this stage.