advertisement


Linn Axis PSU

Albert,
My methods for speed checking are unfortunately nowhere near sophisticated as stobes or tuning forks, as at present I have neither. I simply put a record on with a small blob of blue tack on it's rim and another on the deck as a reference. Then I counted 100 revolutions and timed it with a stop-watch. As per your suggestion I went for 33 1/3 first. 2mins 57.35 seconds was the best I could get, as the trim pot dialled out to it's max (needed to increase the resistance value to slow it down) before I got to the right speed. On 45 I got 1min 59.45 for 90 revolutions, so I'm happy enough with that, but again I needed to dial VR35 up to it's max to get there. Not sure why this should be. On 33 1/3 I think I can hear the difference (0.5rpm), so I may bump up the value of R38 to bring the pot back into it's operating range. This would also bring the 45 pot back into it's range. Then there's the bearing oil to think about!. Now there's a whole new can of worms !.
 
You can download a printable strobe disc from imhttp://www.extremephono.com/free_turntable_strobe_disk.htm IME they work well enough with even an incandescent lamp.

I may well be missing the point but it's not obvious to me that R35 and R37 are about frequency, they look more to do with quadrature. But I have to sit back and work it out.

Paul
 
Looking quickly at it again you could very well be right Paul. It is a 2 integrator quadrature oscillator of course, with frequency dependant on the values of C14/C15, (presumably equal), and the associated resistor values, R22,23,24 and R35,36,37,38, (which should of course also be equal). To keep these pairs of components precisely equal would need trimming capabilities exactly as we have there. This could have more effect on the phase difference than on the frequency, I don't have experience to know for sure. Although we have seen that there is a tolerance of phase difference in the original simple basik setup which would not be acceptable in terms of absolute speed.

It initially seemed to me that, as VR35 and VR37 offer the only adjustable area on the board, it must control speed. Surely something must? You must want to make the speed trimmable but design the circuit to keep quadrature locked in solid and not adjustable if that were possible with this sort of oscillator setup. Relying on 1% component tolerances to set the speed is not the way to go I would have thought.

The block around U4 in the bottom left has a strange effect too. The U5 block on the right seems to have an action which can pull either of the 33 or 45 lines down through D8 and D9 via SD, no matter which is selected. This opens both the switches in the bottom left making a fundamental change to the switching levels in that lower left area which could be the equivalent of a phase change in the feedback, or is it a pulse width thing? Component values would help here, but it seems that those U4 sections could feed a correcting signal into the oscillator via the -1XAmp section which is controlled by U5's action, (in turn significantly affected by the feedback from the motor resistor). U4 compares the level at one of the motor drive outputs with a preset level affected by the 33/45 line levels. And then there is the gating of the U5 section from the output of the lower left U4 switching too!

It's by no means trivial in its action, (at least to my limited capabilities). Any more thoughts?
 
Hi guys,

I've read with interest this thread. I too have a problem with my Linn Axis PSU, but it is somewhat different problem, hope you guys can help me out....

In my case the motor either does not turn, or if it does it either rotates correctly clockwise or in many cases anti-clockwise. Given a helping hand to rotate in a clockwise direction it will generally keep going okay running at 33rpm. However, most often when set to 45 rpm, the motor does not rotate and when it does, again this can be clockwise or anticlockwise.

I've taken a few voltage readings based on what has been said on the thread (many thanks guys). The following is what I've found:

C2 to 0V = 162V
C1 to 0V = 310V
C5 to 0V = 13.45V

The above I assume are okay....

At the motor as viewed from the front of the deck:

Terminal M1 (nearest the heatsink)
Terminal M2
Terminal M3

Voltage across M1 & M2 = 11.5VAC @ 33rpm rotating clockwise, 10.5VAC rotating anticlockwise, 0VAC when switched off.

Voltage across M1 & M2 = 15.2VAC @ 45rpm rotating clockwise/anticlockwise, 1.8VAC when motor does not rotate.

Voltage across M2 & M3 = 74.5VAC @ 33rpm rotating clockwise or anticlockwise, 10.5VAC when switched off.

Voltage across M2 & M3 = 83.5VAC @ 45rpm rotating clockwise/anticlockwise.

I assume M2 goes to 0V as I measure 0V between 0V rail and M2.

I also assume that C8 is connected to M1 and C9 is connected to M2 having measured the following voltages when switched to 33rpm:

C8 -ve contact = 10.2VAC, C8 +ve contact varies between approximately 1.0VAC - 7.3VAC.
C9 -ve contact = 74.5 VAC, C9 +ve contact = 73.5VAC

C8 & C9 are 350V 33mF. I note on visual inspection that at the positive (+ve) C8 connection the capacitor is showing signs of corrosion/leakage. Is this likely to be the cause of my problem? My Linn Axis is a UK spec 240V 50Hz PSU. PSU is dated 28/01/88 and board version is L4/R12.

Thanks, Steve
 
Hi Maybe.

Much of this is guess work without any data to substantiate it! So remember, in true Richard Bach fashion, "everything in this book may be wrong"!!!

The first set of readings look exactly as they should to me. Well done for getting that. The motor values seem a little strange though. You should have 2 separate motor coils, each with their own AC waveform supplied and as near as possible the same voltage but 90deg out of phase. The phase difference would not show at all in a simple meter measurement. They should share a common third connection.

This third connection is the one which runs into R14 and which you think is your M2 as that shows 0V. But I would have thought that you must have a small voltage on this terminal when the unit is in action because the motor current from that terminal passes through R14 to generate the control voltage. This will depend on the value of R14 of course and would not be expected to be very large but I would have expected it to be easily seen.

It sounds plausible that in fact the terminal reading 0V is the drive to one of the motor windings and is reading 0V because that side of the circuit is damaged and not passing any AC waveform via C8/R13 or C9/R15. Without any AC passing to it, the C/R combination will simply drop the voltage to 0V and stay there rock steady.

Assuming that only one side of the drive is working then you could get some effects like the ones you describe. It is the 90deg phase difference which makes the motor generate a semi-constant, (not meant to imply smooth), turning force. As the voltage driving one coil peaks and drops the other is rising and they pass the work from one to the other constantly. In the absence of one side of the drive to one of its coils there may be nothing to make the motor feel anything like a uni-directional turning force and it would certainly be weak and unsteady.

I think it is sensible to say that if you are getting something at the other terminal then both waveforms are being generated to O1 and O2. By the nature of the oscillator it must generate both phases of the drive during its action so it seems more likely that a problem is in the drive circuitry around the motor in the top right.

You will notice that the circuitry in that part is symmetrical. Voltages in the same positions on both sides should be round about the same value. You will have a mix of AC voltages and DC offsets if the unit is running, but you should still see roughly the same voltages on a meter. Of course if you have an oscilloscope....?

Without knowing the component values it is impossible to say what the voltages should be. If you can read resistor colour code you could let us know a few of the values? In the short term, look for badly scorched components and tracks. You may well find a couple of components are slightly scorched in normal use in this area of the board. Remember that when turned on all of the transistors should have about 0.5-0.8V across their base/emitter junctions. The base is the terminal which comes out of the middle of the vertical bar on the circuit schematic and the emitter is the one with the arrowhead on it.

....................I
....................I
.................../ collector
...............I /
...............I
-- base ---I
...............I
...............I \
...................V emitter (with arrow)
.....................I
.....................I

(Very crude but it should help.)

Resistor values are shown using the following:
Black - 0, Brown - 1, Red - 2, Orange - 3, Yellow - 4, Green - 5, Blue - 6, Purple - 7, Grey - 8, White - 9.
Normally the first 2 bands will represent the 2 figures of the resistance. The third ring will give you the number of zeroes to add after that. So Orange White Red is 3, 9, then 2 zeroes, or 3900 Ohms, 3.9 kOhms for example. This may help you to give us a few values of resistors in the drive circuitry.

(Sorry if I am teaching my granny to suck eggs there.)

And did I mention the fact that the schematic says 320V at the top of it?
 
Hi Albertb,

Thanks for the info, it's a very long time since I messed about with electronics.

I've listed the measured values of some of the resistors around the motor section of the PSU, they are so small I couldn't really make out the colours properly!

R12 & R18 = 43 Ohms
R16 & R17 = 44K Ohms
R14 = 10 Ohms
R13 & R15 = 2.3K or 2.4 K Ohms
R7 & R24 = 14.7K so I assume 15K Ohms
R11 & R19 = 22K Ohms
R8 & R22 = 6.2K Ohms*
R10 & R24 = 15K Ohms*

* Note: Very difficult to read whether these are the right Resistor numbers as the resistors are obscuring the marked resistor numbers.

I started measuring the voltage across the large transistors that are bolted to the heatsink, namely Q4, Q6, Q5 and Q8, but I was getting 150 - 160V not the 0.5-0.8V you mentioned. Then one of my probe slipped off one of the contacts of Q6 and I think touched one of the Q7 contacts and shorted something! Saw the spark, and smelt the smell! Then the motor stopped and of course will not start, 0V at the motor! Visually everything looks fine, no signs of any scorching. So it looks like I'm actually worse off then when I started!:)

Best Regards,
Steve
 
Hi Steve,

As Albert already knows I'm an electronics newbie, but I like to dabble. Just a thought that struck me when reading your post - the circuitry in the Axis is load sensing...the load in this case being the inertia of the platter. Your measurements would have to be made without the platter in place, and so no load on the motor. When I first got my Axis running again I noticed the motor going in reverse also, but once 'loaded' with the platter it was ok. It still occasionally needs a push to get it going....temperamental Scottish electronics I guess. I'll be opening mine up again today to make final speed adjustment on 33 (Albert - I made myself a strobe light....no more counting 100 revolutions!). Will post back some component values and readings if that helps.
Regards.
 
Steve....bad luck on the short. I did something similar early on and fried the rectifier. You may need to replace the transistor you shorted. As several people pointed out to me, this board is dangerous !!!. I found that clamping the -ve probe of the meter to the 0v helps avoid accidental contacts, as you only need to concentrate on placing one probe then.
 
Hi Redart,

I should note that you can adjust the two pots without opening up. There are two access holes (with removal covers) underneath the deck.

Best Regards,
Steve
PS I have also needed to give the platter a helping hand to get the thing moving!
 
Well I'll be damned !. Thanks for the info......Would certainly have saved me a bit of time earlier on. In the end I needed to open it up anyway as I'd dialled both pots up to their max values and 33 was still running fast. Added a 1K resistor in series with R38 and that did the trick :) .
Just a note on the power transistors you were measuring (Q4,5,6 & 8). Not sure if you're aware but the pinout on these goes Base Collector Emitter (left to right). May explain the readings you observed. Albert's diagram is correct for the schematic, but for some reason not all transistors have their base leg in the middle. Q7 does however, it's c-b-e.
Hope the short circuit doesn't cause you too many headaches and you get your Axis going again....it really is a great sounding turntable. Good luck!.
 
Hi Redart,

Yeah, I had assumed the Base was the middle contact! So very likely to explain the measured voltages. Hopefully, replacing Q7, will get me back to where I was! I can then restart trying to find out why the motor doesn't run properly at 33 and not at all at 45.....

Just need to find some where that stocks what are probably no longer produced transistors or at least find an equivalent!

Enjoy the music!
 
Bad luck with the short. Believe me we've all done it at some time or other. It is often safer to check on voltages around transistors by reading from the legs of other passive components like resistors. It is all too easy to slip on the legs of a transistor and bridge across its other pins.

Also don't automatically assume that the transistor which has blown is the one you were working on! Shorting out a junction may close off that transistor's action safely but cause damage to others around it. With Q6 and Q7 connected as they are, it would be very easy to damage one when working on the other. These problems can often have a knock on effect to other areas.

Something else to remember is that, when you are measuring resistance with your meter, you cannot reliably measure it with the test device in place on the PCB. There will very possibly be other components in place which will affect the readings. To be sure, you should carefully unsolder one leg of the device from the board and measure it then. That is why it may have been easier to read the values of resistors direct from their colour codes where possible.

When checking out suspect transistors, a simple first line of attack is to check the resistance between each of the pairs of pins. A transistor is basically 2 diodes connected together at the base like this:

( O----l>l----O represents each diode.)

collector / base / emitter
O----l<l----O----l>l----O

(Hope this is clear!)

You can get an idea of the serviceability of a transistor by checking each of the junctions BOTH WAYS as though they were simply diodes.

Start with the Base/Emitter junction. Measure its resistance with the red lead to the emitter and black lead to the base. Then reverse the leads and read it again. Do the same with the Base/Collector junction, then finally with the Collector/Emitter junction.

You should find that you get a very high reading in one direction and a low reading in the other for the first 2, but the Collector/Emitter readings should both be high as no matter which way round you measure it there is always a reverse biased diode in the path.

As a note here, NPN transistors will give their low readings with the leads the opposite way around to PNP ones. Don't worry too much about values, and beware of assuming a transistor is dead if it gives a suspicious low reading when it should be high if it is still in place in the PCB, remember that the other components around it will affect things. Problematically this can occasionally be due to elements like safety diodes or bias resistors built into the transistor itself.

Try here for some more useful information: http://www.elexp.com/t_test.htm

Do any of the transistors have their numeric codes written on? It may be easier and fairly cheap to just replace that whole section if they are not exotic. While it may only be possible to accurately translate the pace of the attack transients of the notes of a bird twittering in a tree in the background of the open air performance of Handel's Water Music on original instruments and in original barges when you have the correct Linn semiconductors blessed with holy water by a naked Count Ivor Tefenbraum on the night of the October new moon, or the anecdotally recommended improved substitue produced using traditional sword making techniques in quantities of 3 per month by warrior monks in the Toshiba sponsored Buddhist monastery in the mountains of Northern Honshu, to just get things up and running there WILL be a lot of possible alternatives which you could easily get and use. (Do I win the prize for the longest single sentence in board history?)

On the subject of transistor pins, if you don't know anything about case pinouts then one reliable way to identify them is to trace the tracks from each leg to their nearest named component and then see which one that would be from the schematic. In trying to get replacements you would obviously be aware of the pinout of what you are replacing and stick to that for the new component.

Also finally, be aware that this board is LIVE on the mains even when the unit is switched off, and there are components which will hold dangerous charges for some time after unplugging it!!! (Can't say those things often enough!)
 
Thanks Albert,

As helpful as ever. I probably won't be able to get any replacement components before the weekend at the earliest, but at least I can test out the existing ones.

Beautifully eloquent and yes probably the longest sentence I've ever read expressing the dreams at least, of all audiophile tweakers!! ;)

Steve
 
Hi

I Am also searching for info on the Linn Axis PSU and found this old post after a google. Can someone please send me the axis schematic/circuit diagram.

Thanks in advance
 
Hello
I know this is old thread,
Anyone has a spare heat sink, the solder in switch or bad whole board to sell?
I have Axis that’s missing heat sink and switch ( solder in version).
Would like to make it work again and linn does not carry them anymore.
 


advertisement


Back
Top