Speed Controller Status

[willbewill]

The complete unit, with top cover fitted, from the back -



There you go. Didn't leave anything out, did I?  

Oh yeah .. I didn't show you the front, did I? Why so? Well ... it's .... functional. Of course, if you take a lot of time and care creating the front panel, then it will look really professional. If, on the other hand, your facilities are limited to a shed and some blunt drill bits and you do it as quickly as possible, then maybe it won't look so great.

Uh-oh ... Nigel has created something BUTT UGLY again, hasn't he?

YES HE HAS  



Let's move on quickly.

Does it work? Yes, thankfully it does. I need to do a bit of testing and tweaking now and then it will get passed on to a couple of carefully selected members, for them to try it out and give their verdict. Keep watching

I should just add - don't worry, if you don't like the look of this thing (I don't  ). There's a lot of different ways to build the controller and to use different pots, switches, etc. Actually, I wouldn't particularly recommend the case that I used - it's not really a very good shape, but it serves the purpose of trying it out, so it's OK, for now.
I'm sure we're going to see some much better looking builds than mine - I only said I could do electronics, not metalwork, OK? 

[willbewill]

After a couple of days of using the controller, some thoughts ....

I've used it for two listening sessions of around 4 hours each, the first one with the output set at 220V and the second with the output set to 180V.

I honestly could not detect any difference in sound at 220V, from running on the normal AC supply. This seems to contradict what some others have reported, but this MAY have something to do with how clean (or not) the AC supply is to start with. I live in a rural area, no industrial or commercial premises nearby and I was listening late at night, so the load on the supply generally would be quite low, I would think. I suspect the supply in my house is quite "clean" to start with - I've certainly never had any problems whatsoever with mains-borne interference.

Running on 180V, however, does seem to make a difference. As we would expect, there is less vibration from the motor. The motor in my L75 is pretty quiet anyway, but it's quieter still on 180V. I'm trying to think of a way of describing the effect on the sound, without resorting to clichés .... but I can't, so - "it's as if a veil has been lifted" 

Is it more dynamic?  .... No.
Tighter bass? ....  No.
Sweeter highs? Yes   

It just seems to give greater clarity, more precision, we could say. Does it make the music more enjoyable? YES - and that's the important thing, after all! What I'm hearing is more music and fewer "artifacts", I think.
Is this purely down to the reduction in motor vibration?  Well, I don't know the answer to that and I don't much care 

At the moment, I'm not hearing any downside to running at reduced voltage, but a few LPs isn't enough to say for sure, so I need to listen some more....

On the technical front, there are a few issues to be sorted out -

1. After about an hour or so, the supply transformer starts to get VERY hot. This is obviously undesirable from the point of view of the transformer lasting a long time, but also because it is heating up the oscillator board, which I separated from the amplifier board in order to prevent it getting heated up! Doh !!
Looking back through my notes, I realise that I measured the total power drain of the controller to be about 40W at full output. Somehow, I had forgotten this at some point and ordered a supply transformer rated at ... 40VA.  Oops! So, it's running right on its limit, which is a BAD idea. I have ordered a 63VA transformer to replace it, which should run a good deal cooler. 

2. I also found that the power supply oscillates a little, at very low frequency. The reason for this is pretty obvious, if we think about it. The ripple frequency of the supply will be 100Hz (from the 50Hz AC line frequency) and we are trying to amplify a sine wave of ... 50Hz or thereabouts. So, what I'm seeing is a "beat" frequency. I think the supply is "sagging" too much under load and I thought this might be related to the under-rated tansformer. I tried replacing it temporarily with a 100VA frame-type transformer, but that actually doesn't make any difference, so instead I'm going to try uprating the supply electrolytic caps a little, from 4700uF to 6800uF (which is what I started with, actually ). We'll see if that makes a difference.
It's unlikely that I can completely eliminate this effect, just reduce it and it's also questionnable whether it matters anyway, but I "don't like it".

3. I need to alter a couple of resistor values around the amplitude control. At the moment, the output is about 180V at minimum setting and is already up to 270V at halfway  Clearly I miscalculated something there! I had intended it to be no more than 250V at maximum setting, so I need to fix that ..

[willbewill]

Are these findings different from your breadboard version?

Well -

(1) overheating transformer - no, I hadn't noticed this before, but I only fitted the toroidal supply transformer at a fairly late stage. Before that, I was using much larger frame-type transformers, with a much higher power rating. The toroidal was only fitted because the frame-type transformers had an audible buzz, which was annoying me. I probably didn't ever run the unit with the toroidal fitted, for very long listening sessions. It takes quite a while, before the high temperature is noticeable. The real problem is that I have developed it in so many short bursts over such a long period of time - I tend to forget stuff. It's a "schoolboy error", to be honest, I am suitably

(2) oscillating power supply - no, I hadn't seen this before - I think the reason here may be that the last time I measured the supply current, the electrolytic capacitors in the supply were 6800uF each, not 4700uF. I changed to the lower value because it is more common and more easily available. I've probably gone too low now - I could calculate this, but I'm too lazy 

(3) range of attenuator pot - the pot on the breadboard version was unrestricted i.e. I had not included any resistors to restrict the voltage to within a certain range. Obviously I slipped somewhere, whilst I was calculating the resistor values. Easy to fix!

[willbewill]

OK, I fixed it all now

I have replaced the supply transformer with a 63VA version. After about 3 hours' use, it is only slightly warm, so that's fine. I changed the two main power supply electrolytics from 4700uF to 6800uF and this has significantly reduced the ripple, to a level where I am now happy that it's OK. I also tweaked a few resistor values, to give a more sensible range on the amplitude pot.

It looks like this now - note the larger supply transformer & electrolytics -



(I forgot to clear away all the objects behind the controller, before I took the photo, so there's a free cigar to the first person who can identify the LP cover in the background  Just to make it difficult, there's a receipt from a veterinary surgeon obscuring most of it!)

I'll just run it for a few more listening sessions now and if all goes well, I'll be posting it off for some "field trials" shortly

[willbewill]

Now, a couple more observations, after using the controller for some more listening sessions ....

1) 45 RPM - I built my prototype controller for just 33 and 45, because I have a separate TT for 78s and also because this is a very simple configuration to build, as all the parts can be PCB-mounted. I hadn't really tried out the 45RPM setting very much, but I thought I should, so I tried a few 12" singles. An immediate problem was apparent - although the platter gets up to speed easily enough when changing from 33 to 45 (i.e. with the platter already running), it will NOT get up to 45 from rest. In fact, it didn't run at all , the motor stalled!
Remember, however, that I had the output turned down to 180V. Increasing the frequency of the motor supply will increase its speed, of course, but at the expense of less torque. It seems that on my Lenco, at least, there isn't enough torque to get the motor moving, if I select 45, with only 180V output.
So, I increased the voltage to 200V and now it starts with no trouble at 45RPM (phew!).
This may not be an issue with every Lenco, but it's something to bear in mind. I would think that anybody who wants to achieve 78RPM, with the idler at the "33" position, will almost certainly have to run at full voltage.
Of course, it would be easy enough to arrange for a different output voltage at each speed, but that's adding complexity again, so I'm not going to detail how to do it. 

2) I realised something interesting, whilst listening last night. For the previous few days, I had not given any thought to the sound. Nothing had troubled or disturbed my listening - distortion, mistracking, surface noise, etc, etc, etc. I was just completely lost in the music.
Significant? I don't know, but it was  .... interesting 

OK, I think I'm happy for this thing to go out on "field trial" now - I'll send it off next week

[willbewill]

Some facts & figures for you ...

One purpose of the speed controller is to hold a steady output, in the face of fluctuations in the AC line voltage. How good a job does it make of that?

Well, in the EU, the AC line voltage is supposed to be 230V +/- 10%. i.e. from 207 to 253V
So, using a variac and setting the output to 200V, I measured as follows ..

Line voltage       Output voltage        Power supply voltage

207                  200.0                     +/- 29.5
230                  200.0                     +/- 32.8
253                  200.0                     +/- 36.4

... so, no problem there, then! 


Next, I decided to look at amplitude stability - i.e. does the output voltage change over time, as the unit warms up? The simple answer is yes, it does, but I don't think it's a big deal. Take a look at the following graph, measured over about 5 hours or so, with the turntable connected and running -



Looks horrific, doesn't it?  Look a the Y-axis carefully, though. What we actually have is a drop of around 8V in 5 hours, which stabilises to within little over a volt within 2 hours. The most drastic change is right at the beginning - looks roughly exponential, doesn't it? So, what are we seeing here?
Well, two separate effects, I think. I suspect the intial rapid change is caused by the chip amps heating up and changing their gain very slightly, this will happen pretty quickly, as it doesn't take them long to warm up. There will also be a much slower underlying trend, caused by the XR2206 function generator IC, being heated up by the other parts in the case (i.e. the chip amps, LM317 regulator and supply transformer).

So, 8V in 200V over equates to a 4% drop in voltage. This beggars the question - can you actually HEAR a 8V voltage drop??

Well, if I tweak the amplitude pot from 200 to 190V in a few seconds, I can't hear any difference at all, so I don't think we have a problem

I suppopse if you want the ultimate precision, then the thing to do, is ensure everything is well warmed up, before making final adjustments.

Frequency stability is not so easy to measure - I don't have any way of measuring 50Hz with great precision, so here I will have to rely on the Lenco strobe. Of course, we have to allow for the fact that the Lenco itself drifts quite noticeably from cold, so the answer here is to start with a hot Lenco and a cold speed controller! I'll let you know what I find 

[willbewill]

Use of the speed controller for 78 RPM

Here's something I wanted to remind everybody about. As you know, the controller can be built for 1, 2 or 3 speeds (33/45/78).
At 33RPM, the motor will be driven by a 50Hz (or 60Hz) sinewave. Increasing the frequency will increase the speed of the motor, hence allowing us to obtain 45 or 78RPM, with the idler at the "33" position. The downside of this, is that the motor produces less torque, as the frequency is increased. If, therefore, you are trying to run your Lenco at reduced voltage (say 180V), you may find that the motor has insufficient torque to get the platter moving, at the higher speeds, or else the idler may slip on the motor shaft and not "grip" it properly.

My L75 is not able to start at 45RPM on 180V. However, Jean has tried this with Yoda and, despite Yoda's heavy stacked platters, it works just fine. Why the difference? Well, I suspect that it may be due to how well set up Yoda is and the cleanliness and condition of the idler wheel.

It certainly is possible to get the platter moving at 78 in the same way and people have done it, but the setup and general condition of the deck will become even more critical in this case.

So - my point is - in order for the controller to work at its best, your Lenco must be well set-up and with everything clean and in good order, otherwise it quite possibly won't work properly at the higher speeds. Just something to keep in mind ...