In a Spin

Aside from admiring, owning and using them, your humble servant is now active within the global economy of Leslie Speakers, having jointly developed an accessory which is commercially available. More of that sordid, mercenary semi-professionalism later. First, true to my Ham credentials, I'd like to share some details of my new Leslie interface, just for the love of it.

Readers have learned how I got my first Leslie, an 825, from Bob, just down the M56 in Chester. A couple of weeks before Christmas, I got another Leslie from another Robert, a little further west in Holywell. I'm pretty sure that the relative remoteness of Chester and (particularly) Flintshire kept the prices down to my beer-budget levels.

The new (old) Leslie is a 125 - similar to the 825 in that is has only a single speaker (but that's about to change - watch this space). The 125 is, however, critically different in at least three important respects. Firstly, it has a veneered wooden enclosure (rather than the ugly "Tolex" vinyl covering of the ProLine 825) - it LOOKS like a Leslie. Secondly, it has a valve(/tube) amplifier, with all the romance that entails. Thirdly, it uses a different interface to the outside world.

The 825 uses Leslie's "9 pin" interface, as described in this previous post. It achieves speed control by a simple, low voltage switching scheme, ready for plug 'n play connection to my homebrew half-moon switch.

The 125 uses an older interface standard, called "6W". The 6 indicates the fact that there are only 6 conductors in the cable to the speaker (and 6 pins on the special, expensive Amphenol connectors). The "W" actually stands for Wurlitzer, in order to differentiate from another contemporary six-pole Leslie interface, "6H", in which the "H" meant Hammond. The 6H interface was used for - you guessed it - Hammonds, as exemplified by the Leslie 122. The 6W interface, developed from a Wurlitzer wiring standard, is also called the "Universal" interface and is used for other applications, as in the Leslie 147. These two interface standards are dangerously INCOMPATIBLE, yet they share the same physical layer in the special "6 pin" connector cable, which is used in both standards.

Here, for my own records as well as your edification, are the pin assignments in the 6W standard...

Pin	Colour	Function
1	Black	Signal Ground
2	Yellow	Motor Relay
3	Gray	AC Power In
4	Blue	AC Power In
5	Brown	Motor Relay
6	Red	Signal Input

The colours refer to the cores in a real Leslie cable.

Note the apparently inoccuous pins 2 & 5, connected to the Motor Relay. In order to switch between the two available speeds ("chorale or tremolo"), one has to apply 240V AC between pins 2 and 5 (in the UK at least). Speed control certainly isn't a "simple, low voltage switching scheme" and it implies switching mains electricity. No voltage gives tremolo (the fast rotor speed) and 240V gives chorale (the slow speed).

Call me a wuss if you like, but I have a pretty well-developed fear of high voltage electricity. It all started when my childhood friend, Paul Sibley, tried to stop the smell caused by a budgerigar feather burning on the element of a small electric fan heater. Paul used a pair of scissors to remove the feather with predictable results. He flew across the kitchen at altitude of about six inches and slammed into the opposite wall. Mr Sibley was a successful builder who had built his family an impressively proportioned home, so Paul's trans-kitchen flight was a long one. He survived and I learned what has proven to be a useful life lesson.

With all the baggage of burning budgie feathers, I wasn't about to go switching mains voltages on my virtual organ console. I don't think the guitar pickup selector switch in my homebrew half-moon switch would have handled 240v AC - much less, into slightly inductive loads. So - a safer switching scheme was indicated...

Of course, I could have simply used a relay to switch the voltage to the motor relay (big fleas have little fleas upon their backs...) and used the Telecaster switch to energize the local relay coil at low voltage. That, however, wasn't entirely true to the spirit of the ideas brewing in the back of my mind for more flexible speed switching options, including 9-pin, 6-pin and maybe even 11-pin interface compatability, so I added a little electronics.

Here's the low-voltage aspects of my speed control circuit - it is based on a somewhat over-specified relay from Maplin (which required me to learn again for the umpteenth time how to create a new part for Eagle). The design includes a stabilized 12V supply (not strictly needed for the switching operation, but important for the rest of my design, as you'll see in a moment), which is at the top of the schematic below.


I've edited out all the high voltage parts of the circuit. I don't want to be held responsible for people who haven't learned the consequences of burning budgie feathers.

Here's the switching circuit in the flesh...


Still glowing with the success of my pre-amp for the 825, with its HF shelving filter, I made another identical circuit for use with the 6W interface...


All these printed circuit boards are all very well, but they only become a practically useful system when placed in an enclosure with the appropriate connectors. In sourcing a suitable box I learned that some of Maplin's enclosure range is to be discontinued - I hope this isn't yet another stage in the long, drawn-out demise of this once-useful source of components. (For overseas readers, Maplin is a high street retail chain filling the spot once occupied by Tandy - the UK version of Radio Shack. Until five or so years back, Maplin rivalled the big "trade" players in component supply with on-line and high street options. The demise of hobby electronics (indeed, of any practical pursuit) and the nonsense of trying to provide it on the high street has driven Maplin's metamorphosis into yet another consumer electronics / tech store. I doubt they will survive much longer.)

Anyway, here are some shots of my finished 6W interface, allowing the 125 to be connected to my virtual organ (i.e. a computer soundcard) or anything else which can produce close-to line level audio into high impedance.


The big red switch turns on the whole shooting match (including the Leslie) and the two controls on the front are for volume and HF boost.There are internal jumpers selecting 0, 10 or 20 dB overall gain. On the rear are mains inlet via an IEC socket, the 6-pin chassis socket (I splashed out and purchased a 32.5mm hole punch to add to the collection), audio input via 1/4 inch jack and speed control via a 3-pin DIN (as used on my other "pre-amp"). It works perfectly and makes hooking up the 125 (or any other Leslie using the 6W interface) a walk in the park.

Here is the finished 6W interface unit, sitting atop the 825 (the 125 is on the other side of the room).


Now to return to the commercial world...

I was chatting with Hammond and Leslie expert George Benton, after reading of his adventures using a Hammond Super B console as controller for a virtual organ. George had seen the games I was playing with PICs in an organ context and asked if I'd like to join in developing a speed controller for Leslie speakers. I accepted, thinking I'd learn from the experience and from friendship with George - both have turned out to be the case.

First, a little explanation.

Early Leslie speakers (the 21H, 22, 22H, 22R, 31H, 44W, 45, 46, 47and 51) were single speed (or, at least, single speed plus "stationary"). Later models evolved to the more familiar two-speed operation mentioned above. It will come as no surprise to hear that the old, single speed models are coveted and valued for their special, desirable properties, real and imagined. Prices soar accordingly. Despite this, owners of single speed units also like the two-speed feature, so opportunity exists for electronic speed controllers able to convert a single-speed unit to two-speed operation. A number of such controllers are available.

George had the bare bones of such a system but needed to develop software - enter yours truly.

Our Leslie speed controller is a fairly standard motor control application, using Pulse Width Modulation techniques. The motor is placed in an H-bridge, formed of IGBTs, under the supervision of a PIC microcontroller...


Although Motor Control is a standard PIC application, we chose (for legacy reasons) to use a very simple 8-pin PIC with no hardware Capture/Compare/PWM module. This made the entire project a matter of trying to squeeze a quart into a pint pot. Not in terms of memory capacity or computational load - rather in handling the time constraints imposed by the interrupt-based software PWM modulator I developed.

The single-speed motors are built to run at "tremolo" speed when driven by 110V, 60Hz, so the controller generates this output signal when "tremolo" is requested. The motors can also be persuaded to run at slower speed, appropriate to "chorale", by lowering the frequency. However, this drop in frequency is accompanied by a drop in the inductive impedance offered by the motor, such that the motor passes more current and runs hot. This can be countered by dropping the voltage magnitude and frequency in "chorale". However, the motor speed is actually a complicated, non-linear function of both voltage magnitude and frequency, all of the consequences of which are anticipated and handled in the control algorithm.

Transitions between each speed (the controller also has a "stop" mode, giving stop, chorale and tremolo as the three available operating speeds) are handled with various boost or brake options, some of which can be selected by the installer.

What with the adaptive interrupt service routine and several other "smarts", there's quite a lot of innmovation and intellectual property in the controller. George and I are real proud of it!

Here is my development platform for the code - I implemented the IGBT H-Bridge on the white PCB for development purposes, but never connected it to a motor...


Here's the final product, which replaces the relay in a single speed Leslie...


It was rather time consuming developing real-time motor control software on one side of the Atlantic and testing it on harware on the other - but we got there! Learning the obvious lesson from that geographical separation between development system and target, I now have assembled all the bits required to add a horn, under the control of our speed controller, to my 125. I also have the single speed motor for the LF unit, so it will go "all electronic".


This will make development of the next versions of the controller a whole lot easier - plus I get to have a great organ speaker!

Over the past few days there has been some disquiet as the Bank of England engages in the Carollesque delusion called "quantitative easing". I can't see any real economic benefit in that. Instead, why don't you make a real difference by heading off to the Benton Electronics on-line store and getting yourself an MTCLogic 2-speed Leslie controller.

Every home should have one. At least, every home with a single speed Leslie.

...-.- de m0xpd