Hi Pyramid,
0.1Hz change in frequency is indeed a change of 0,057 in RPM.
The difference between 33.476 and 33.204 is like you said almost 5 steps from 0.1Hz on the generator.
But when measuring RPM with 3 digits behind the decimal point, each 0.1Hz step from the generator should already be clearly visible instead of the 3 steps suggested by you.
0.057 RPM on 33 1/3 RPM means an accuracy of 0.17%, shouldn't that be good enough. I googled a bit but could not find what deviation is still allowable.
And in case of a VPI TT, there are several ridges on the pulley for further fine tuning.
Hans
Hi Tom,
Reading your posting, it seems to be the generator that sucks.
Either you have a unit that is malfunctioning, or it is by design inadequate for the job.
Hans
Hans-
The tachometer should most certainly show a 0.055 RPM speed change with a change of 0.1Hz on the generator. That's why I suspect the generator does not change output frequency with each step.
Regarding speed accuracy: Define "good enough". I think the RIAA suggested LIMIT is 0.3%, but that was made at a time before variable frequency drives made it possible to adjust more precisely. It may also have to do with maintaining the inflection points in the RIAA curve and not any other metric of audibility.
The tachometer should most certainly show a 0.055 RPM speed change with a change of 0.1Hz on the generator. That's why I suspect the generator does not change output frequency with each step.
Regarding speed accuracy: Define "good enough". I think the RIAA suggested LIMIT is 0.3%, but that was made at a time before variable frequency drives made it possible to adjust more precisely. It may also have to do with maintaining the inflection points in the RIAA curve and not any other metric of audibility.
Hi Guys,
After reading about the problems you are having I have borrowed an HP frequency meter and will test my generators ( I bought 2), but I rather suspect they will exhibit the same problem.
I had a good look at the HS380 board and whatever the chip used is, the top has been sanded off.
Sadly I suspect the device uses a PWM generator, not DDS
I am now looking into using AD9833 dds generators,possibly in conjunction with an arduino, but this will not be a quick solution.
In the meantime I will continue with the conversion of my TD124 to using the Papst motor and the 'less than perfect' 3 phase generator, I don't have perfect pitch, so if my TT runs at 33.30 instead of 33.33 I really won't worry, as long as it's quiet.
After reading about the problems you are having I have borrowed an HP frequency meter and will test my generators ( I bought 2), but I rather suspect they will exhibit the same problem.
I had a good look at the HS380 board and whatever the chip used is, the top has been sanded off.
Sadly I suspect the device uses a PWM generator, not DDS
I am now looking into using AD9833 dds generators,possibly in conjunction with an arduino, but this will not be a quick solution.
In the meantime I will continue with the conversion of my TD124 to using the Papst motor and the 'less than perfect' 3 phase generator, I don't have perfect pitch, so if my TT runs at 33.30 instead of 33.33 I really won't worry, as long as it's quiet.
Hey Ralph
The good news is it seems to warm up quickly and the rpm remains stable. I played some albums last night and it sounded good at 33.444. Sounds like there is no easy fix to fine tune the frequency so out to the lathe I go. You have the machines so it will not be a problem for you either.
Maybe I just got a bad one. Will see when you test yours.
Tom
The good news is it seems to warm up quickly and the rpm remains stable. I played some albums last night and it sounded good at 33.444. Sounds like there is no easy fix to fine tune the frequency so out to the lathe I go. You have the machines so it will not be a problem for you either.
Maybe I just got a bad one. Will see when you test yours.
Tom
Hi folks, I have exactly the same problem that Tom has with mine. I was not sure if it was real, or not because of the possible inaccuracies of my tach, and the fact the motor was unloaded. I suspect the problem is with the generator. I'm re-engineering my project along the lines that Ralph has done, because it makes more sense. I'm going to sell my bigger amps on ebay along with the supplies to recoup some of the costs. I was also suspecting my transformers at 10va weren't enough, so am going to 30va transformers to make sure. I wish we could find a better source to generate a very accurate sine wave and frequency. Ralph please keep us informed if you find a better source, and also about the results of the frequency meter test.
BTW being a semi-pro harmonica player, I have pretty close to perfect pitch, as I have to bend the reeds to get a lot of the notes, almost like playing a fretless instrument. This makes me very conscious of variations of pitch, especially if playing along with the music, so pitch control is very important to me.
twystd
BTW being a semi-pro harmonica player, I have pretty close to perfect pitch, as I have to bend the reeds to get a lot of the notes, almost like playing a fretless instrument. This makes me very conscious of variations of pitch, especially if playing along with the music, so pitch control is very important to me.
twystd
Hey Twystd
All is not lost just put it together and find the closet interval with current pulley. Then have a pulley made for the right rpm. Pitch issues are more related to variation of speed not to a specific exact rpm. The rpm is holding steady just not easily adjusted like they advertised.
Well I will machine that pulley to get it close and do some comparisons. Also going to try a brushless DC motor with 3 phases.
Enjoy Tom
All is not lost just put it together and find the closet interval with current pulley. Then have a pulley made for the right rpm. Pitch issues are more related to variation of speed not to a specific exact rpm. The rpm is holding steady just not easily adjusted like they advertised.
Well I will machine that pulley to get it close and do some comparisons. Also going to try a brushless DC motor with 3 phases.
Enjoy Tom
Tom, you're right about that if you are just listening, but when I play harp along with a record, my harps are tuned to concert tuning. When the pitch of the music is just right, I only have to bend the reeds for the missing notes. Diatonic harmonicas only have 20 reeds for a 3.5 octave scale, which has 42 notes (12 per octave). It makes it very difficult to play along with something, or someone for that matter, that is not tuned to the correct pitch. Don't mean to get of topic, but my needs are a little more demanding than most.
twystd
twystd
I have been following this thread with mych interest, unfortunately most of it is beyond me. Sounds like an obstacle at this pount is producing a clean sine wave. Would this work?
Take an AC motor and use it to generate the wave. It could be driven by a stable dc motor or even a stepper motor. It will undoubtedly produce 60hz plus a lot of junk. So pass the output through a notch filter, and use the filter's output in place of the function generator.
Take an AC motor and use it to generate the wave. It could be driven by a stable dc motor or even a stepper motor. It will undoubtedly produce 60hz plus a lot of junk. So pass the output through a notch filter, and use the filter's output in place of the function generator.
The real problem seems to be getting accurate .1hz steps from the generator. I think with the filtering that's being done by the low pass filter on the amp inputs, and the snubber caps across the output caps from the transformers, most of the noise is removed.
As far as a motor generator, since one of the problems we have is trying to control the variables of the 60/50 hz mains power, and it's effect on stability, we would have the same problem with a motor generator. The 3 phase Paspt motors are about as stable as they get, but still subject to variables in the frequency of the incoming line power, voltage doesn't really affect them. I suppose you could change speeds with such a rig, but I think it would still be subject to variables of line voltage and frequency. Again, I'm certainly no expert, could be wrong on all that, lord knows I've barked up the wrong tree many times on this project.
twystd
The accuracy necessary for the capstan is very critical. For example a Papst motor running at 1800rpm (which it should at 60 hz) driving a 12" platter to 33.33 rpm would require a capstan of 1800/33.33 which is a ratio of 54.054:1, so if our platter is 12" in diameter we would need a capstan of 12/54.054 = .222".
If we changed the diameter by .001" to .221" that would be a ratio of 12/.221 = 54.298:1. At 1800 rpm motor speed, would give us a platter speed of 1800/54.298 = 33.15 rpm. That .001 change in diameter gives about .18 rpm of platter speed.
That is more than 3 times the speed change of .1hz @60hz, .1/60 =.00166 or a change of .16%, which should yield a change of 33.33x.00166 = .0553 rpm on the platter speed. That is IF the signal generator is really capable of .1hz accuracy, which seems in doubt at this point. :-(
twystd
Thanks Pyramid,
The figure of 0.3% seems to justify the 0.25% drift of the VPI ADS after one hour playing.
VPI
Hans
It is a pitty that the 3 phase generator does not deliver what was to be expected.
In my case I am very happy with the super stable Project Speed Box plus phase splitter, in combination with a motor pulley with several sleeves for fine tuning.
Once set, you will never have to touch it again, even after 10 years of use.
Hans
In my case I am very happy with the super stable Project Speed Box plus phase splitter, in combination with a motor pulley with several sleeves for fine tuning.
Once set, you will never have to touch it again, even after 10 years of use.
Hans
Track down the answer but be careful not to fix the problem too welll. Often things that need understanding work best. Always chech resistors and capacitors are best they can be if they set the speed. Ceramic NPO/GOG are 30 ppm and good price. I use 10 nF and 318 K to get 50 Hz.
3 phase must work.
3 phase must work.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.