Anyone built a VPI-SDS clone?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
VPI's SDS turntable motor controller is more than just a voltage controller. It's a frequency controller that stabilizes what is fed into VPI's synchronous motor. It's $1000 new (about $650-800 used), which is a lot of dough to spend (for me).

Anyone have plans for something like this?

Thanks,
Bob
 
YES!

Thought no one would ask! :)

Bob, I'm designing one right now. Its based arround a pic with the output passing a butterworth filter to turn it into a sine wave. Has the same resolution as SDS (.01Hz) can store multiple speed configurations and accuracy is pretty much dependent on the clock feeding the pic (which can easly go down to 1ppm if you think it's worth it). Has 2 outputs so you can digitally control phase of your motor. Also has voltage ramping and fold-back, cartridge mileage counter.
Heck, come to think of it, its better than SDS. And yes, for the DIYer it should be waaay cheaper that SDS since most of its functions are done in the software domain (err.. I'm a programmer).

Right now I'm aiming at low voltage output since that's what I need for my TT, but if others get interested in it I may extend it for 110/220V output.

At this moment I have a prototype on breadboards with most part of the digital section as well as the software, done. Filtering and output amp is not implemented yet, but I'm getting to it (very little time :( )


Stay tuned. I'll start a thread on this forum once I have someting worth looking at.

Guilherme.
 
Hi Peter.

Because of resolution. You'll want your 100 slices to have the same duration otherwise you'll introduce distortion. So, when you want to change frequence you have to sum/subtract the same amount to all of them and so the total delta is multiplyed (in your case) by 100.

For example, .01Hz change is a 0.9999 uS change in period. You would have to divide that by 100 to get the right timing for each slice. That would be 9.9 nS. So your counter would have to run at 9.9 nS steps, not an easy thing for a pic or any other mcu for that matter.

Doing it the way i'm doing it I just have two counters with a step resolution of 200 nS which is still almost 5 times lower than what I need.

Guilherme.
 
Yes Frank, please do go take a class of physics (and take another one on fairplay while you're at it - your comment was rude to say the least).

Anyway, back on the subject, if you're going to use an AC motor (either because that's what is on your TT and you dont want to canibalize it or because you feel they're superior to DC motors), me thinks that then it is a good idea to use a device similiar to SDS, because:

1) You can get a more precise RPM that is independent of what is going on at mains and to compensate for the tolerances in the TT. Most TT given exactly 50.00Hz/60.00Hz will not spin at exactly 33.33 RPM, so you need a device to adjust the freq.

2) Phase control is important because it can minimize motor vibration. REGA uses such a system in their upstream line, dont know why SDS has not addressed this.

3) Voltage foldback is important again because of motor vibration. The torque needed to win the inertia of a still platter is not the same as the one needed to keep it running, so its a good ideia to reduce voltage (therefor reducing torque) once the right speed is achieved because the motor will run more quietly.

4) Other goodies are not as important, but since I have 4K of program memory to burn...

To every one else that replyed: as I said, I'll start a thread once I have it more completed. The pic can be programed using a PC printer port so thats not a problem and PCB (the layout at least) will be available.


Guilherme.
 
diyAudio Senior Member
Joined 2002
Hi,

your comment was rude to say the least).

Nope, it was out of respect for your project I refrained from commenting.

A courtesy you apparently don't deserve but I'll let you find out the hard way.

Brett,

If you want me to explain why this isn't such a good idea, send me an e-mail...we have some catching up to to anyway.;)

Cheers,;)
 
diyAudio Senior Member
Joined 2002
Hi,

Guilherme,

Pursue whatever you feel happy with, I didn't mean any offense to you as a person.

As a long term commisioned designer of TTs and associated problems I don't think your approach is correct.

If memory serves me correctly I do recall you had trouble adapting a US Thorens to European voltage/cycles and are pursuing ultimate speed accuracy from your TT.

The 50Hz spec of the mains is something that usually doesn't deviate significantly, voltage may vary and may affect some crappy TT.

Solving that problem however will be far more expensive and complex than opting for a TT with sufficient inertia to maintain correct speed provided someone did do the maths.

Once you have a platter spinning with a 30 kgs mass at the correct speed, and this only requires some math and a decent motor, nothing is going to upset it.

Lower the mass of the spinning platter and ask yourself how on earth anything except a low noise, low torque AC motor running from the mains is going to do anything for a badly designed TT, it won't and nothing ever will.

If people want to engage in tailchasing, than that's fine by me...it's just not something I would embark on.

Cheers,;)
 
Frank

Do you really think suggesting everyone builds/buys a 30kg platter is a good advice? And cheaper than to design a simple regenerator? An AC motor fed by a clean sinewave seems to produce significantly less vibrations and everyone who has listened to various Linn supplies, per example, would attest to this. The huge number of Garrard/Thorens/Linn/etc owners can experience a very audible improvement from a single/dual phase regenerator without having to add 30kg to the platter. It may not yield the same sound as a truly top flight table but the expense/effort is incomparable.

cheers
peter
 
Hello Frank.

Thank you for your explained answer. Yes we did talk in the past about this subject: it was not about Thorens, but rather me trying to use a test LP to autoset the platter speed. I agree with you that a heavy plate will not be so afected by the motor.

My point is that there are a lot of TTs out there that dont have 30Kg platters (mine included) and are being driven by AC synchronous motors. For these, I still think that units like SDS, Walker Motor Controller, Lingo and so on will help.

All these units cost a lot and probably some would cost more than the TT itself (a bit ridiculous). Since I'm not aware of any DIY project that can compete with something like the SDS, I think that a new project would be wellcome. I can tell you that so far my project is quite cheap, and looking at what still needs to be done I see no reason why it will not be so in the end.


Think I'll keep chasing my tail for just a bit longer :)

Best Regards,

Guilherme.
 
diyAudio Senior Member
Joined 2002
Hi,

All these units cost a lot and probably some would cost more than the TT itself (a bit ridiculous).

True and I've yet to hear a single one that actually does what they're supposed to do.

For the usually high asking price I'd rather buy a better TT or tonearm or cart.

Another approach would be to find out what is upsetting these A.C. motors in the first place.

From what I gather it's HD and spikes riding on the mains which can be filtered out quite effectively with a limited BW powerxformer and a few snubbers.

A typical TT load is roughly only 5 VA yet I find bigger xformers better sounding. ( I often go for 1KVA EI core)

While on the topic of xfmers, you could split the secundary into two and ground the CT so you can now feed the TT from a balanced supply.
This in turn will reduce any tendency to pick up mains hum by some MM cartridges.

Spec the xformer with a static shield that you connect to Earth as well and voila, you can now also shield the secundary side of the xformer by connecting the CT to Earth too and enjoy a clean 50/60Hz mains cycle.

You can make the linefilter as sophisticated as you deem necessary but I'm pretty confident that this is pretty adequate already.
It won't break the bank and heck, even a CDP will benefit as well.

If you do have the tools necessary to analyse the mains supply before and after the xformer you should be able to view the difference it can make.

Make sure to take all necessary safety precautions when working with mains voltage and make sure you filter/xformer is boxed and has no voltage carrying connection(s) touching the chassis.
In other words if you don't feel up to it let a trained technician do it for you.:att'n:

No offense to anyone in favour of electronically regenarated devices, so if you feel it's worth going for please do.:)

Cheers,;)
 
Although I don't want to get into a flame war, I tend to side with fdegrove.

It's my understanding that a turntable's job is simple. It should be a flat table which turns at a very constant rate and does not vibrate at all. It should also be able to absorb vibrations into it and not store and release that energy later. All these properties can be easily realized with mass. There are ways to make a light table spin evenly, but the stored energy problem has to be dealt with as well. Mass takes care of it all and simply too. Mass also helps to absorb any bearing noise as well. moment of inertia is your friend with a turntable:

I=Mr^2

I always thought it was odd that linn would build a $1500 power supply to power a $10 tape deck motor.

All that said, there is a lot to be said for trying to improve existing designs, and to me this is where this project fits in. It's not like you can drop a 30 kg platter on a linn and call it a day. You'd end up building a table from scratch.


Sheldon
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.