I have acquired a 90s speed converter designed for the kind of AC motor used in a Linn TT - it adjusts speed by changing the frequency of the mains supply.
It consists of a PCB with half a dozen chips on it and 2 power transformers (and some caps!):
* the first transformer is 240/15v, feeding a rectifier to provide a DC rail for the chips, while
* the second is 240/12v, situated at the end of the circuit and is connected in reverse. IE. the secondaries are attached to the PCB and the primary winding feeds 240v power to the power socket on the back of the case (which the TT power lead, plugs into). The PS for the TT is the original, basic Linn PS consisting of a 10K resistor to drop the voltage getting to the motor and a 0.22uF X2-rated mains cap, to provide a phase difference.
My question is ... is it OK to "reverse" a power transformer in this way ... or is it better to use a transformer which has been specifically designed to have a primary of 12v and a secondary of 240v?
Any input greatly appreciated.
Andy
It consists of a PCB with half a dozen chips on it and 2 power transformers (and some caps!):
* the first transformer is 240/15v, feeding a rectifier to provide a DC rail for the chips, while
* the second is 240/12v, situated at the end of the circuit and is connected in reverse. IE. the secondaries are attached to the PCB and the primary winding feeds 240v power to the power socket on the back of the case (which the TT power lead, plugs into). The PS for the TT is the original, basic Linn PS consisting of a 10K resistor to drop the voltage getting to the motor and a 0.22uF X2-rated mains cap, to provide a phase difference.
My question is ... is it OK to "reverse" a power transformer in this way ... or is it better to use a transformer which has been specifically designed to have a primary of 12v and a secondary of 240v?
Any input greatly appreciated.
Andy
Last edited:
My Thorens motor needed at least 90Vac to start and run well.
So I aimed for 110Vac.
I built up a 50W audio Power Amplifier and fed it's output to a reversed transformer.
I could set the turntable to the 33&1/3 pulley and feed in 50Hz from an oscillator to run the TT motor.
Increase the oscillator to 67.5Hz for 45rpm records.
Both these frequencies can be generated from standard cheap crystals.
So I aimed for 110Vac.
I built up a 50W audio Power Amplifier and fed it's output to a reversed transformer.
I could set the turntable to the 33&1/3 pulley and feed in 50Hz from an oscillator to run the TT motor.
Increase the oscillator to 67.5Hz for 45rpm records.
Both these frequencies can be generated from standard cheap crystals.
OK - so it's efficiency that is the issue. Thanks.
Now I have heard that transformers designed for US mains (60Hz) don't work as well in Oz as designed-for-50Hz transformers.
So the fact that the Hz required for 45rpm is higher than for 33rpm (67.5Hz, according to AndrewT) ... does this mean that, again, I should use a transformer that has been built for higher-frequency usage?
Regards,
Andy
Last edited:
Andy
The efficiency thing is not really a problem. If you are worried about it, use one of the small Tamura transformers that use what they call "semi toroidal" contruction: these are effectively symmetrical.
I have about ten spare small 12V jobs, intended for a design that uses digital quadrature and one transformer per phase but they'll handle your application no problems. I've also had good success with the small toroidal transformers that Jaycar sells (They have a shop in A'Beckett St, IIRC that's just around the corner from you)
In general a transformer designed to work at 50 Hz will work better at 60 Hz and at 67.5 Hz. The opposite doesn't apply: a transformer designed for use at 60 Hz may overheat at 50 Hz (higher iron losses).
BTW if you use a variable frequency drive to a standard synch motor with a phase faking capacitor, the phase angle will be wrong at all but one frequency, since the phase angle of an LC filter varies with frequency.
The efficiency thing is not really a problem. If you are worried about it, use one of the small Tamura transformers that use what they call "semi toroidal" contruction: these are effectively symmetrical.
An externally hosted image should be here but it was not working when we last tested it.
I have about ten spare small 12V jobs, intended for a design that uses digital quadrature and one transformer per phase but they'll handle your application no problems. I've also had good success with the small toroidal transformers that Jaycar sells (They have a shop in A'Beckett St, IIRC that's just around the corner from you)
In general a transformer designed to work at 50 Hz will work better at 60 Hz and at 67.5 Hz. The opposite doesn't apply: a transformer designed for use at 60 Hz may overheat at 50 Hz (higher iron losses).
BTW if you use a variable frequency drive to a standard synch motor with a phase faking capacitor, the phase angle will be wrong at all but one frequency, since the phase angle of an LC filter varies with frequency.
Last edited:
Thank you for those 2 pieces of good news, Mark.
I am currently using an 80VA Harbuch 240v/12v transformer; after some conversation with them, they are suggesting I should've used a transformer which was specially designed for 12v/240v - rather than a normal 240v/12v one "in reverse". (But they haven't said why this would be a better thing to use.
)Bugger!
Can you tell me what the value of the phase-faking cap should be, for 67.5Hz?
Thank you for your kind offer, Mark, but unless there are other advantages in using it (in place of the 80VA toroid I have used) I'll pass, as efficiency is not really a concern.
Does this mean you are again making an AC motor speed controller?
Regards,
Andy
Thank you for those 2 pieces of good news, Mark.
I am currently using an 80VA Harbuch 240v/12v transformer; after some conversation with them, they are suggesting I should've used a transformer which was specially designed for 12v/240v - rather than a normal 240v/12v one "in reverse". (But they haven't said why this would be a better thing to use.
If it's a toroid, it's because the primary is wrapped on the core and the secondary over the top so the secondary captures flux leakage from the primary. BTW I get my trafos custom wound by Harbuch; Peter's really good but can be painfully slow.
EDIT: I just realised that I had Peter make up some transformers this way round for the first iteration of the Garrard controller. They are 50VA units but he wound them on a 120VA bonded core with epoxy centres to make them as quiet as possible. His part number is PTT5524 if you want to get one, or I could lend you one to try out. I'm in Williamstown, PM me.
ω = 1 / SQRT(LC)
ω1 / ω2 = SQRT(L2C2) / SQRT(L1C1) = SQRT (L2C2 / L1C1)
(ω1 / ω2)^2 = L2C2 / L1C1
If L1= L2 (the motor winding inductance is constant)
then (ω1 / ω2)^2 = C2 / C1
Since ω1 / ω2 = 1 / 1.35 , C2 / C1 = 1 / 1.35^2 ~ 0.55.
Halving it will be close enough, they're never accurate anyway.
Does this mean you are again making an AC motor speed controller?
The PC board layout for the new KT88 based Garrard controllers goes out to the fab house today. Once these are done and shipped I have to do some new three phase controllers using 6L6G outputs. Then there are half a dozen bike frames to make, then I'll think about it.
Last edited:
Yes, Peter provided me with toroids. Thanks for the explanation - that sounds like having the secondary being 240v would deliver an improvement.
Just as a matter of interest, why did you choose only a 50VA transformer? The original transformer in the speed converter was only 30VA but I decided to up it to 80VA on the basis that Linn use massive transformers in their Lingo PSes (at least 350VA and maybe 500VA).
Sorry, but you're way ahead of me, Mark.
Am I right in assuming you are saying:
* if 220nF is the phase cap value for 50Hz then 220*.55 = 120nF is the correct value for 67.5Hz?
So I assume the Garrard motor is different to the motor which Linn use (Premotec, in my case)?
Bike frames - wow! So all this has to be done before the new vintage is ready?
Regards,
Andy
The design load is 20 watts but I went to 50 to allow larger loads to be driven. Overrating the core reduces flux density and thus lamination noise.
Linn is Linn, 'nuff said.
Got it.
Yes, Garrard used a shaded pole induction motor, nominally 16 watts input but they often pull 20 or more. They are also a beast of a load, especially when starting up, unlike a synch motor which once locked on the line frequency presents a very benign load (no weird phase effects).
Bike frames - wow! So all this has to be done before the new vintage is ready?
Yeah, I'm a bit snowed at the moment. My construction method is new, based on acoustic principles using what luthiers call "tonewoods", so to get insurance I have to get some expensive testing done.
Last edited:
I've taken to driving my Linn Sondek LP12 off a reverse connected 12V mains transformer driven by one channel of a Dynaco ST70 from a sine wave function generator. That's got to be about the laziest possible way to do it, but still sounds better than house AC. I can get 45 rpm, too. Just dial in about 81 hz.
Last edited:
Can you please confirm that I've understood your setup correctly:
1. You have the "basic" power supply on your LP12 (resistor to reduce voltage getting to the motor plus a phase cap)?
2. Moving "backwards" from this power supply, you have:
* a reverse-connected 12v mains transformer
* one amplifier channel <-- NB1:
* a sine-wave generator running at 60Hz for 33rpm? <-- NB2:
NB1: I see the ST70 is rated at 35w into 8 ohms.
What load is the reverse-connected transformer supplying to the Dynaco?
NB2: I presume you can adjust the sine wave generator frequency slightly to get the correct speed?
Thanks,
Andy
Hi -
The load drawn by the Linn Sondek appears very mild (I have one of the original versions) - IAC, I am driving about 100Vrms into it.
The Dynaco ST70 is probably not an optimal driver for a reverse connected transformer but does well enough. I am using a single channel. The transformer I am using is a 12VA that I had laying around - it may be a bit 'small' for the task, FWIW. If I push it too hard, I begin to get harmonic distortion starting around 120VAC and the turntable performance doesn't seem to improve at 33RPM IAC at voltages much above 100 Vrms.
Starting torque is a bit low at 45 RPM, but plays fine.
My function generator, unfortunately, is a cheapie that takes about 20 minutes to stabilize its frequency. It'll usually drop about 7% after a cold start, forcing readjustment, which is the biggest real problem with this setup, besides its kludginess.
The load drawn by the Linn Sondek appears very mild (I have one of the original versions) - IAC, I am driving about 100Vrms into it.
The Dynaco ST70 is probably not an optimal driver for a reverse connected transformer but does well enough. I am using a single channel. The transformer I am using is a 12VA that I had laying around - it may be a bit 'small' for the task, FWIW. If I push it too hard, I begin to get harmonic distortion starting around 120VAC and the turntable performance doesn't seem to improve at 33RPM IAC at voltages much above 100 Vrms.
Starting torque is a bit low at 45 RPM, but plays fine.
My function generator, unfortunately, is a cheapie that takes about 20 minutes to stabilize its frequency. It'll usually drop about 7% after a cold start, forcing readjustment, which is the biggest real problem with this setup, besides its kludginess.
Last edited:
Record a 50 Hz sinewave on an Ipod for about one lp side length (say 25mins) using one of the easily available virtual generators. Nice stable frequency and auto shutoff.
If you want to get fancy, record a second track time delayed 5ms from the first and use a stereo amplifier. If the speed's not correct, change the frequency of the track, don't forget to adjust the time delay.
If you want to get really fancy, mix in some third harmonic antiphase with the peak.
Last edited:
The load represented by the motor is not simple. It has components due to the winding resistance, the winding inductance and the back EMF from the motor acting as a generator. Fortunately with a synch motor these are all fairly minor and a 5 watt amplifier will handle them easily.
One thing you do have to watch with transformers is the magnetising current for the core. With some IE core transformer this is enough to let the smoke out of some amplifiers.
You might like to see Mark Kelly's post in this thread "Effect of the Phasing Cap at different TT speeds?"
It would seem you have low torque at 45rpm bcoz you need to reduce the value of the phasing cap, compared to the value used for 33rpm?
I am now going to implement a capacitor that can be switched in or out, to go with changing from 45rpm 33rpm.
Regards,
Andy
this is very interesting
http://www.norre.dk/TTPSU/TTPSU.html
if anyone is interester it might be a good idea to get the pcbs knocked up as a group?
i was wondering about the phasing capacitor etc, any thoughts
http://www.norre.dk/TTPSU/TTPSU.html
if anyone is interester it might be a good idea to get the pcbs knocked up as a group?
i was wondering about the phasing capacitor etc, any thoughts
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.