Featured post

What's this blog about?

Saturday, 16 December 2017

EL84 Amp II: progress post 1

Progress is happening with the new amplifier... this design is more modular as I have decided to design standard boards for tone controls, headphone output, and phono RIAA.
Having standard boards for these means I can easily accommodate future builds, and provide a "menu" of sorts.

The process has not been entirely smooth sailing, owing to the somewhat hit-and-miss nature of home PCB fabrication. Until now, my method has been to print the PCB design onto an iron-on transfer which then gets pressed onto the board (and then touched up with the etch resist pen) before going into the etchant.

This process has been unreliable and time consuming, and expensive, owing to the high reject rate. So a new technique was called for.

I've decided to move to a photosensitive board workflow. The design is printed onto transparency, which is then plaed over a light-sensitive board and exposed under a UV light, thereafter a two-step chemical process: Developing then Etching.

The first board I designed for this project is the tone control. This is using the same circuit as the previous project, except I had two changes:
  1. I needed to reduce the size of the board 
  2. I needed to put the tubes on the copper side of the board
So, I re-designed it to be 120mm X 65mm (down from 150 X 75) and attempted to fabricate the board... with less than spectacular results


Yeah. Not enough light. 


This board failed because I did not expose the photosensitive layer sufficiently.

Lesson learned, I did a second attempt, which looks much better. So I went ahead and drilled and stuffed it.

Result:

There are four topside wire links on this board. I always try to design with as few of those as possible. It's a challenge!



The copper looks a bit more messy than I'd like because the balance of exposure and development and etching was still not quite right, but this board is usable at least.


The tubes are on the copper side because of the customer's preferred aesthetic of having the tubes visible. This design will be applied to the other boards in this amp as well.

In the process, I have become a lot more familiar with the operation of my PCB software: namely DesignSpark from RS. Also its quirks and foibles, such as less-than-ideal behaviour when moving things around, and its ability to have "invisible" track that isn't visible in design but is when you print. As a result of this, the board above needs to have one track cut with the dremel and re-routed with a short jumper on the track side. Yeah I hate doing that!

Lesson: Inspect the board VERY carefully in print-preview before fabrication.

Or, to use an appropriate engineering axiom: Measure Twice, Cut Once!

I also built the bias boards for the EL84s. Owing to the amount of heat these produce, I am not mounting them on boards, but the voltage divider and potentiometers for the negative bias voltage, and the cathode shunt, can be put on a board. So drawing on my earlier design, these are the bias boards, made using the same technique:



Next up: A two-triode RIAA stage, I'm planning this on a board on 100 X 65mm.

There's a reason I want these boards as small as I can get them: The size of the chassis


Internal Dimensions 300 X 225mm

This chassis is going to represent a challenge to fit everything into it... this design will have 13 tubes: The RIAA stage, tone controls, headphone stage, as well as the amplifier itself. And size is a consideration since it will be packed up and sent overseas when it's finished.

Next update when I have more boards to show...

Monday, 4 December 2017

New project: Another EL84 Amp

Following an approach from a new customer, a new design has emerged...
This customer had a well-defined set of requirements:
  • Usage situation dictated an EL84 PP design would be suitable
  • MM Phono peamp required
  • Line-level inputs required
  • Tone controls required
  • Headphone output required
  • Remote control volume adjustment required
Fairly rapidly I decided this amp could be based on the previous EL84 amp I made at the start of the year, with some additions.

Power Transformer

Firstly, I intend using an off-the-shelf power transformer. The custom-wound transformers are handy, but they're an industrial product, and as such the aesthetics in their design limit their use in a piece of equipment where they are going to be on display. Sadly the manufacturer was unwilling to work with me on this aspect, so my transformers will have to come from Canada now, instead of being locally made.

The power transformer I selected for this job is the Hammond 370FX. 172mA at 275v, 3A at 5v and 5A at 6.3v, with a 50v bias tap. Everything I need.


Tone Control

The previous tone control worked well enough for it to be included in this project without modification. Except I'll redesign the circuit board.


Headphone Stage

This customer was adamant this amplifier have a headphone socket. This was a non-negotiable requirement since their musical taste is not shared with other members of their household. This is provided by an ECC99 SRPP-based OTL design borrowed from the internet. It simulates well in LTSpice down to 32 Ohm headhones, and will drive into 16Ohms as well, although with greater distortion and a lower level.


Printed Circuit Boards

This amplifier will be designed on several PCBs:
  • Power Supply
  • Phono Stage
  • Tone Control
  • Headphone stage
  • Bias adjusters for EL84s
The circuit boards will differ from the previous tone control board in that the tube sockets will be on the opposite side to the discrete components, to facilitate the boards being mounted upside-down in the chassis, allowing the tubes to rise from the top of the chassis as in a point-to-point design.

The EL84s will be chassis-mounted, as in the previous EL84 design on this site.


First stage of development, we have a circuit.

Full circuit. Click to enlarge, right-click to download.


By way of explanation:

V1 + V2 are the MM cartridge phono amplifier stage. RIAA equalisation is given by the RC network giving NFB to the stage

V3 is a Cathode Follower, necessary because the preceding phono stage has a high output impedance, and also to provide additional current capability to any line-level signals at the input, to drive the tone stage.

V4 provides around 20dB gain to compensate for the losses in the tone stage, restoring the entire stage to unity gain. This is the same circuit as the previous "Tone Control" project on this site

V5 is the gain stage for the amplifier proper
V6 is the concertina phase splitter. This needs an elevated heater.

This stage encompassing V5 and V6 is borrowed from the Fisher X-100.

V7 and V8 are the PP output stage with the EL84s, running in Ultralinear configuration into Hammond 1650E output transformers. Fixed bias is employed with the cathode resistors providing the reference voltage for adjustment.

V9 is the gain stage for the headhone amplifier, V10 and V11 the SRPP current driver stage to power the headhone output.

The power supply will incorporate the same 30-sec startup delay on the B+ as the previous amplifier projects on this site.


Owing to the current capability of the low-voltage secondaries, we have a split, with some of the tubes receiving DC heater voltage and others receiving an elevated AC.


Parts are ordered, next stage is PCB design. to be continued....

Wednesday, 29 November 2017

Tone Control Finished

After much waiting on parts, the tone control is now finished and in service.

The 250K Potentiometers took three weeks from order to arrival, in the meantime I'd been using components of the wrong value, so the characteristics were not correct.

Also the front panel has been an epic test of the patience to get the printing onto it. Several techniques were tried:

  • Using thermal transfer film - the same method I use for making PCBs - with the iron. Result: Design and lettering failed to transfer cleanly.
  • Using a cold-transfer method with a laser-printed design and chemistry (mix of alcohol and acetone). Result: A highly flammable and volatile mix of chemicals, complete failure to transfer lettering
  • Print onto paper, transfer paper, transparency (smooth and rough side), experiment with printer settings regarding toner etc - all to no avail.

In the end, the method that was the least dreadful involved covering the front panel with adhesive masking tape, and using a laser-cutter to cut the outline of the letters, then peeling them off with the tweezers to create a stencil, through which several coats of black spray paint were applied, before peeling off the adhesive then applying several coats of clear lacquer to protect the paint.

The results are not fantastic, but they are tolerable in the face of the spectacular failure of the previous methods attempted.

High on my To Do list is to devise a better method of printing onto aluminium.

Anyway, this is the device as completed


All assembled and in service


Full frontal. Yeah a bit of a Star Trek vibe in the labelling. The LED is just a power indicator


The transparent acrylic top gives a nice view of the insides.

Also there's a few extra photos here if you need to see more.



The circuit as built in the end. Note I changed the R and C values in the treble arm to even up the response on both sides



What LTSpice (Circuit Simulator) says this circuit should do at various control positions


Observations
  • This circuit works extremely well; listening tests reveal a completely neutral sonic signature, and that is using the cheap Shuguang 12AX7 tubes (all I had to hand)
  • The circuit is "quiet as the grave" - hum and hiss are inaudible even with the amplifier on maximum volume and ears pressed right up to speakers
  • The boost and cut levels measured on the oscilloscope (see earlier post) match closely with the predictions in LTSpice
I am very pleased with this circuit since it was my first attempt at designing an audio circuit on a PCB. Previously, my PCBs were limited to power supplies.

Waiting to be done: Distortion and noise floor measurements. Rainy Day activity.

This unit is now in service in my listening room, sitting between my RIAA stage and integrated amplifier.

This unit is fitted with the same very important SAF* Modifier as the integrated amp: A power-pass port, to allow the main amp to turn on the tone control and RIAA stage, so that multiple power switches don't need to be toggled to play some vinyl


* SAF := Spousal Acceptance Factor

Thursday, 9 November 2017

FFT Tests on the tone control

With the tone control board and power supply built, but the case still in the machine shop, it seemed a good opportunity to run some performance tests. The site I got the circuit from only had Spice simulations rather than measured results. So I don't know if this is the first time actual test results from this circuit are available.

Anyway, my testing method was to run a sine sweep into the input from 50Hz to 40kHz, then connect input and output to the oscilloscope, running the output into a 100K dummy load (to simulate the volume control of the amplifier it would be running into)

I ran FFT transforms at tone flat, full treble lift, full treble cut, full bass lift and full bass cut. The results are below. 

Please ignore anything below around 70Hz; my FFT process loses all resolution at that frequency.

There are two traces on each plot; the yellow trace is the input signal, for reference. The blue trace is the output from the tone control.

First up - the testing setup


Ain't it beautiful? Careful where you put your fingers!
The 9V is for the signal relay; if 9V is present the contacts close and the circuit is engaged. If no 9V, then the relay bypasses the circuit and bridges the input and output. When the case is made I'll wire up the 9V supply properly but for now it's Energizer-power


Tone controls flat. As mentioned in the text, ignore the region below 70Hz; my FFT process doesn't have resolution there


Bass full cut


Bass full lift


Treble full cut


Treble full lift


In each graph, scale is dB on the millivolt, 5dB per vertical division, yellow is input trace (for reference) and blue is output. (My FFT math capability to reference one signal off the other doesn't work properly, hence the presentation of both signals).

For completeness, I ran the same tests on the other channel. Results were identical.

Conclusion

This tone control is working exactly as anticipated, and is not far from the predictions on the source site.

Something seems to have gone right.

Tuesday, 7 November 2017

Gestation photos of the Tone Control project

Just a few photos of the project's gestation. Click each to make it bigger :)


The circuit I decided on. Simple tone control (bass and treble) with feedback. Uses an initial cathode follower and a final gain stage to compensate for the losses in the tone control section. Because it's an Active tone control, it doesn't need audio taper pots. Linear ones work fine. I managed to find some with a centre detent as well!


I decided to make this on a PCB just because I hadn't done this before. So being a complete novice at PCB design, everything is done by hand. No auto routing or anything else. This was my initial sketch


Then I designed the board. Measuring the components I intended to use by hand with the micrometer. Yeah, when I said "First Principles" I meant it.


The board, ready for etching. The design was made according to the size of the board blanks I had to hand, to avoid having to do any cutting. The transfer process involves printing the design with a laser printer onto a plastic sheet, then using a clothes iron to transfer it on to the board blank. Usually this requires a bit of touch-up with a Sharpie-pen but in this case it was almost perfect.


Work in progress. Starting to build the board. I discovered to my dismay that I did not have all the resistors I need, so there are some unfilled holes in this board. 

Also there's the power supply board which I made the same way, it's very simple and boring.


This whole project started because I bought the power transformer from the local auction site for $9. It has 213 - 0 - 213v secondaries, plus 6.3v. This gives a nice B+ of around 300VDC.

Also I had two 12AX7 tubes left over from a previous project - these are horrible cheap Chinese ones, but they work OK. Sufficient to test it, if I like it I might put some JJ or EH ones in.

More photos later when I get the case back from the CNC and laser etching...

Saturday, 28 October 2017

KT88 Amp Freq Response measurements

With the big KT88 amplifier nearing one year old, it occurred to me that I had never performed a frequency response measurement on the whole amplifier.... I'd done plenty of them when I was prototyping the preamp and driver stages, but I hadn't done anything other than cursory checks with the oscilloscope over the entire amplifier.

Today I set about rectifying that omission. My method was simple. Move the amplifier to my (newly tidied) workbench, attach the function generator to the input, dummy load to the output, set up a sine sweep from 50Hz to 50hKz at 200mV RMS, open the taps until the amp is producing about 5v RMS on the output, then start the FFT process.

This resulted in the following:




So the yellow trace reveals that the input signal is not exactly flat... about 1dB down by 30kHz. This is why I showed the input signal, for a reference.

The blue plot is the speaker terminals. the load is an 8 Ohm carbon resistor on a heatsink.

So what does this tell us? I read this as the amplifier being reasonably flat (within 1dB) to 40kHz whereafter it drops rather quickly. 

My 'scope does have a "Math" facility where I can supposedly reference one signal against the other. This functionality seems buggy however, which is why I avoided it.

I am happy with this frequency response.

Monday, 16 October 2017

New project - tone control

After building the last two amplifiers, I've had several months of not building anything, while I wait for the next firm customer. Lots of people expressing interest, but no-one ready to put any money down... yet.

So in the meanwhile I've been keeping an eye on the local auction site and pouncing whenever anything that looked suitable for a future project came up.

Latest score was a small power transformer with 213-0-213 secondaries (so good for 300V B+) and a handy 6.3v also.

Size suitable only for a preamp, this transformer cost the grand total of $8.00

With a couple of 12AX7s left over from my last project, thoughts turned to the possibility of making a headphone amplifier, nice idea except I didn't really need one.

Then I saw a site with some tone controls. My phono cartridge is a little down on treble so it seemed like a good idea to build a tone control based around a pair of 12AX7s. Idea being to put this between my RIAA stage and the integrated amp.

This is the schematic:



(Heaters don't need elevating; it's an error in the notes)


Just to challenge myself, I plan to build this on a PCB rather than on a chassis with point-to-point. The reason for this decision is so that I can at some future point take this PCB and transplant it into something else, if needed. Or alternatively, make up another one quickly if needed.

The next challenge is to arrive at a PCB layout. Being a complete novice at PCB layout design, my preferred design method is the same as the constipated accountant: Work it out with a pen and paper.

Or in my case, because I am not a complete luddite - my Surface Pro computer with the drawing pen.

So. This is the concept drawing of the PCB - obviously I'll duplicate it for stereo - and there will also be a pair of relays to disconnect the inputs and outputs from the circuit and tie them together for a tone bypass.



Next step will be to measure the components and design the PCB.

Work in progress.

Just for reference, my Phono preamp I am using is the 3-triode "Little Bear"


This machine uses 6N2 tubes, two for gain and one on the output as a cathode follower.

I've tested its RIAA response using the FFT function on my USB oscilloscope:


Blue is input 5mV RMS Sine sweep,  yellow is output. Scale is dB on the millivolt. Ignore the input below 70Hz, hum due to unshielded leads.

From this it's clear that the RIAA stage is behaving itself, so the blame for the slight treble loss from the turntable must be with the cartridge. An Audio-Technica ATS-11 with a band new Shibata stylus that cost considerably more than this tone control will.

(The treble loss is ascertained by listening tests by doing an A/B comparison with the same music played from vinyl and a digital source, synchronized at playback. Other than that, the vinyl sounds fantastic.)