Turntable speed control

[morricab]

Yes, it would help to know what numbers the SpeedNic is using. I would doubt the duty cycle is as short as you proposed, but it is technically feasible and possible. On the RR tach, there is a built in software triggered stobe that I used during development to check for trigger ambiguity, and never removed in production. The pulse width of my strobe was 300uS and required a very bright white LED run at higher current than normal to be visible. Even at 300uS, the flash does not appear very bright. The brightness of the LED is directly proportional to the current. For a single ended pulse (PWM) the RMS value of current will be sqrt(duty cycle), so in your example the brightness would be sqrt(1/10,000) or 1% or normal.

My purpose for crunching the numbers in the previous post was to illuminate the fact that strobe disks can have limited resolution, especially when using an ambient light source, and the difference between the Timeline and a normal strobe disk, even if using a crystal controlled source.

In theory I could measure it in my lab with photomultiplier tube and oscilloscope. Just need some time to try it out with the LeCroy Waverunner scope. We have the system setup to measure laser pulses, light scattering and mass spectra (coming from a micro channel plate detector).

I was basing my pulse width on some literature and experience with laser pump strobes. Could be I am off by a fair bit. But, is it the full pulse we see or the leading edge? Then rise time would matter from zero to full output and not how long it is sustained.

 

[morricab]

Yes, it would help to know what numbers the SpeedNic is using. I would doubt the duty cycle is as short as you proposed, but it is technically feasible and possible. On the RR tach, there is a built in software triggered stobe that I used during development to check for trigger ambiguity, and never removed in production. The pulse width of my strobe was 300uS and required a very bright white LED run at higher current than normal to be visible. Even at 300uS, the flash does not appear very bright. The brightness of the LED is directly proportional to the current. For a single ended pulse (PWM) the RMS value of current will be sqrt(duty cycle), so in your example the brightness would be sqrt(1/10,000) or 1% or normal.

My purpose for crunching the numbers in the previous post was to illuminate the fact that strobe disks can have limited resolution, especially when using an ambient light source, and the difference between the Timeline and a normal strobe disk, even if using a crystal controlled source.

Let me ask you a question; have you seen some TTs speed fluctuating during playback? I noticed this with a few TTs.

 

[Phoenix Engineering]

Let me ask you a question; have you seen some TTs speed fluctuating during playback? I noticed this with a few TTs.

Using the tachometer, I do see speed changes during playback. The changes fall into 3 broad categories:

1. Slow drift over time as the table/bearing/belt warm up. The speed always increases over time and can range from 0.2~0.3 RPM increase. Even after stabilizing (30-40 minutes), if I switch to 45 RPM for a record side, then switch back to 33, the speed will be higher (this is without the corrective feedback from the tach).

2. Stylus drag. Clearly visible, usually 0.015~.025 RPM drop in speed as the stylus drops. The decrease in speed is greater at the beginning of a record vs the end. In theory, stylus drag should increase and decrease with changes in groove modulation, but if it exists, it is below the threshold of the tach (3 decimal places).

3. Random fluctuations. This can vary greatly between tables and in severity. Even on the same table, if the belt/pulley/platter is contaminated by oils from my hand, the speed can vary +0.01 RPM. Belt drift (up and down the platter) or twists in the belt can also create this type of variation. I run 3 belts with a non-tapered pulley and the short term stability seems much better. I had a Lenco idler table and the native stability was fairly poor (±0.030~0.050 RPM), but the idler tire and bearing had seen better days. I have seen Lenco tables using the Eagle PSU and tach, and the stability was quite good.