If someone wants to have their cake and eat it often the crystal can be replaced by a RC oscillator . NOP/COG capacitors are often no worse on stability. The 74HC4060 divider sometimes had examples of that in data sheets. The typical crystal being 3.2768 MHz for 50 Hz. One can also start at a lower frequncy so as to use just one 4060. The HC version will drive the first passive element of a filter as it has at least 20 mA of current to offer. Ask it to do 1 mA and it will give the full 5 V peak. Most of these oscillators have a simple logic. Logic high / low. The 4060 is the most versitile, 4040 and 4020 can be used as stage two. 2 x 4060 can be banked. What the others offer is other divisions of frequncy. All seem to leave some ports unavailable.
PLL Synthesized vs DDS
It would be interesting to see how stable that design is. Any type of PLL is going to have some amount of phase noise above and beyond that of the crystal reference. VPI uses a PLL design (although quite different from Elvee's) and it's VCO has some pretty significant phase noise:
http://www.diyaudio.com/forums/vendors-bazaar/252742-digital-turntable-tachometer-dds-based-psu-16.html#post4217853
It would be interesting to see how stable that design is. Any type of PLL is going to have some amount of phase noise above and beyond that of the crystal reference. VPI uses a PLL design (although quite different from Elvee's) and it's VCO has some pretty significant phase noise:
http://www.diyaudio.com/forums/vendors-bazaar/252742-digital-turntable-tachometer-dds-based-psu-16.html#post4217853
I should say right away that I thoroughly respect Kevinr's project, and IMO such solutions are elegant for 50/60hz conversion, and might result in reduced motor vibration as a bonus which can only be a good thing.
However, when it comes to short term stability of platter rotation, IME many other factors dominate, rather than mains source variation. Such as eccentricity of centre spindle (inc hole tolerance), centre spindle axis error, warp, tonearm/headshell stability, pulley run out, pulley spindle vibration etc etc. So even with stable synth mains, there remain dominant sources of pitch variation which have nothing to do with mains generation and can't be fixed that way. Also with synth mains, there's the possibility of introducing accidental fresh short term variation in frequency too, though IME this is unlikely to be significant versus other (non-mains) sources, so I wouldn't sweat it.
Anyways, I've enjoyed following Kevinr's project, every international home should have one.
For curiosity, any chance of posting a short 10s recording of a c 3kHz test tone please, Kevinr ?
However, when it comes to short term stability of platter rotation, IME many other factors dominate, rather than mains source variation. Such as eccentricity of centre spindle (inc hole tolerance), centre spindle axis error, warp, tonearm/headshell stability, pulley run out, pulley spindle vibration etc etc. So even with stable synth mains, there remain dominant sources of pitch variation which have nothing to do with mains generation and can't be fixed that way. Also with synth mains, there's the possibility of introducing accidental fresh short term variation in frequency too, though IME this is unlikely to be significant versus other (non-mains) sources, so I wouldn't sweat it.
Anyways, I've enjoyed following Kevinr's project, every international home should have one.
For curiosity, any chance of posting a short 10s recording of a c 3kHz test tone please, Kevinr ?
You are right about all the mechanical aspects these have a huge impact on the speed but first of all it starts with the motor without it no speed !
A stable voltage for a TD124 motor is important I have done measurements on input mains voltage"s and speed some were in these postings.
I wonder how in the UK with voltages from 240-250 V it goes with the E50 motor ? When I measure the speed at these voltages it very difficult to compensate this with the brake .
It was actually originally built to power my 401 which has a 50Hz platter.. It will eventually run that table.
Any measurements done will be dominated by the physical health of the table it is powering.
The source is based on a stable crystal clock and I have not been able to measure any variation short or longer term in output frequency. (I have a decent 100MHz HP counter, but that's it.)
Any measurements done will be dominated by the physical health of the table it is powering.
The source is based on a stable crystal clock and I have not been able to measure any variation short or longer term in output frequency. (I have a decent 100MHz HP counter, but that's it.)
When designing the Garrard 501 we got rid of the pulley. The shaft taken up to 6 mm ( 4.91mm 401 ) which needs about 52 Hz to work. Oil getting on the wheel can result. A simple collar works well to stop that. We took the 301/401 to be a Bible of how it should be done. The 6 mm doesn't degrade anything as best we know.
The problem with the 52Hz is it beats with 50 Hz. 70 Hz is not so bad when setting time constants. It looks to me that a 401 shaft might work at circa 64 Hz if 33 1/3 and no pulley. A simple pulley of circa 5.7 mm for 45 RPM might be required which could serve as an oil deflector if using 64Hz. To see if you have a beat frequency problem look to see if voltage stability is better at 45 RPM. If under 1/2 volt at 230 V it should be OK. A light bulb of the older sort seems very good at showing this. As luck would have it the eye is more sensetive than the motor. Making the bulb seem constant will be good enough. I have used a 15 watt fridge bulb for this. A meter usually shows it well enough. You may find playing with chips in the oscillator helps. Sometimes even no-no devices like LM324 give the better results. Reason is they are working in an ideal way for once and minimising the nasty things very well. 70 Hz is not too taxing for the LM324 if giving > 1 V rms.
I see this thread has died. I will risk a new idea. One unseen oppertunity is to run a Garrard 301/401/TD124 as if in Swindon circa 1958. The 50 Hz pulley is the better option. 40 Hz to 60 Hz suits 78's. ESP Audio has an excellent section on State Variable Filters as oscillators. You need 30 very good watts to drive a 301. 100 watts amps serve well. 3.2768 Mhz and 2 x 74HC4060 and a 6 pole Chebishev filter via Texas Filter pro Freeware workes well. If using a SVF a 4060 can be used as a calibrator, 100 Hz if using LED. 300K + 18 K + 10 nF all at 1 % is about right for SVF at 50 Hz. A 50 K 20 turn pot in one arm should be enough for fine tuning. 10 nFCOG although 5 % typical are often 30 ppm temp co. A good cheap choice mostly in SMD.
Induction motors prefer lower frequencies within the laminations abilities ( 40 Hz ? ). The Garrard will tollerate 270 V. 253 V should be fine, run at 220 V in general or even 200 V if 1955 301 converted to oil. If the motor is left running there is a chance it will last longer if a good 1930's style SAE 20 oil is used. BBC did that. I serviced a 301 that had run without many stops since 1964 ( never stopped in principle ). It was perfect. A pressing plant in Hayes.
I have always thought a PP EL 34 amp should work. I suspect a resistive load needs to be in paralell to prevent flashover or whatever. I suspect it could be capacitor coupled if using the 115 V motor option. Never tried it nor even sketched it. Some feedback might be needed. It might even be cheaper. At face value the coils are 3K3 @ 230 V. That might mean a few 34's. I dare say many toroids would work if turns ratios were OK . Many I buy are tested at 1500 V for flashover. It is tempting to think when class A that would be forever. A simple fuse in the HT would be wise. PP 34's in pentode should work. 4 x 34 in triode might be OK without feedback.
Induction motors prefer lower frequencies within the laminations abilities ( 40 Hz ? ). The Garrard will tollerate 270 V. 253 V should be fine, run at 220 V in general or even 200 V if 1955 301 converted to oil. If the motor is left running there is a chance it will last longer if a good 1930's style SAE 20 oil is used. BBC did that. I serviced a 301 that had run without many stops since 1964 ( never stopped in principle ). It was perfect. A pressing plant in Hayes.
I have always thought a PP EL 34 amp should work. I suspect a resistive load needs to be in paralell to prevent flashover or whatever. I suspect it could be capacitor coupled if using the 115 V motor option. Never tried it nor even sketched it. Some feedback might be needed. It might even be cheaper. At face value the coils are 3K3 @ 230 V. That might mean a few 34's. I dare say many toroids would work if turns ratios were OK . Many I buy are tested at 1500 V for flashover. It is tempting to think when class A that would be forever. A simple fuse in the HT would be wise. PP 34's in pentode should work. 4 x 34 in triode might be OK without feedback.
We were given a BBC and Garrard unit that hint at that and have two Garrard 401 BBC units ( look like Doctor Who ). The Fairchild is the real deal. I will even more choose Fairchild when I can.
4 x EL 84 in pentode PP 320 V B+. +/- 12.6 V PSU to also drive op amps. 20 dB feedback. It could be cheap enough. Take the motor to be 3K3. EL84 I think in pentode can be driven by an op amp. Op amp inverter to get the PP. I would use simple cathode bias. PSU would be 56R 7W WO8 rectifier and 330 uF as a starting point. Most USA houses have 240 VAC somewhere.
4 x EL 84 in pentode PP 320 V B+. +/- 12.6 V PSU to also drive op amps. 20 dB feedback. It could be cheap enough. Take the motor to be 3K3. EL84 I think in pentode can be driven by an op amp. Op amp inverter to get the PP. I would use simple cathode bias. PSU would be 56R 7W WO8 rectifier and 330 uF as a starting point. Most USA houses have 240 VAC somewhere.
- Status
- Not open for further replies.