The oscilllator in this unit has always been unstable.
I've compared it with a better strobe tuner. It's good but not quite as good as my newer electronic strobe. When I tune with the ST-11 I can hear little beats when I play chords that I don't hear with the new strobe tuner.
I think the problem lies in the oscillator circuit because I'll calibrate it. Tune three or four strings. Check the calibrate mode again and it's a little out. I can't seem to keep it from drifting no matter what I do.
I think the ST-11 can be as accurate as the best new technology if the oscillator can be made more stable.
I'm wondering if anyone with some talent in digital design wants to help me try to improve this old workhorse.
I've compared it with a better strobe tuner. It's good but not quite as good as my newer electronic strobe. When I tune with the ST-11 I can hear little beats when I play chords that I don't hear with the new strobe tuner.
I think the problem lies in the oscillator circuit because I'll calibrate it. Tune three or four strings. Check the calibrate mode again and it's a little out. I can't seem to keep it from drifting no matter what I do.
I think the ST-11 can be as accurate as the best new technology if the oscillator can be made more stable.
I'm wondering if anyone with some talent in digital design wants to help me try to improve this old workhorse.
I know this sound corny but you could try leaving it on for a few hours before you use it. The drift could be caused by it just heating up. You could also try recapping the unit with higher tolerance caps.
I was wondering if it had a crystal oscillator but it does not. If it did you could have used a ocxo but it looks like it uses line frequency as a stable oscillator.
Nick
I was wondering if it had a crystal oscillator but it does not. If it did you could have used a ocxo but it looks like it uses line frequency as a stable oscillator.
Nick
nhuwar said:I know this sound corny but you could try leaving it on for a few hours before you use it. The drift could be caused by it just heating up. You could also try recapping the unit with higher tolerance caps.
I was wondering if it had a crystal oscillator but it does not. If it did you could have used a ocxo but it looks like it uses line frequency as a stable oscillator.
Nick
Hi Nick!
Thanks for the reply
I've actually worked on quite a few of these units. They all seem have the same problem. The oscillator circuit is TTL and is unstable no matter how long your warm it up. The oscillator circuit probably needs to be redesigned. It outputs 146.66666 khz when calibrated so it's not an off-the-shelf frequency.
Actually the line frequency is used just as a reference to calibrate the oscillator.
I'm hoping someone can help to devise a nifty circuit that retains the adjustability, yet is very stable. Maybe even with a center detent which locks the oscillator to dead-on 440 HZ concert pitch, since most people want to calibrate it for perfect pitch.
The good thing is that the oscillator circuit is all on one module which is removable. So a new board could be easily installed. Without making permanent modifications to the unit. Thus retaining the original value of this old classic.
There are a lot of ST-11's out there so I think it would be a good update. I just know know where to start. There are probably far more talented people than I, so I thought I would suggest the project.
Thanks
I got it ... I saw the oscillator ... I think is the section of 72121 and the NPN transistor ... well ... How can you pretend that this stage can be precise and stable ? Sorry but I saw: 10% resistors, instead of 1% or 0.5% .. no crystals ... just RC oscillator ... I ask myself: what should be the real frequency generated by this stage please ? And the motor: is a step motor ? Because I don't think it is ... please correct me if I wrong ... bu what I see it's just something was good 50 years ago ... today you wan have better circuitry ... with step motor and crystal oscillators ... so I have just the following questions:
- frequency of the oscillator ( what should be )
- is it a step motor already ?
Thank you very much
- frequency of the oscillator ( what should be )
- is it a step motor already ?
Thank you very much
It is not a stepping motor, no.
have you measured this drift in oscillator freq? In servicing these, especially as they age, I often have to replace the motor caps. Dry caps could make the motor drive unstable. And keeping fresh oil on the motor shaft also prevents it from dragging. And allow the wheel space back from the window to reduce air drag.
There may well be tons of these out there, but I doubt there is much of a market for a replacement board. COnsider that for the price of said board and the labor for a tech to install it, you could have a brand new tuner, more accurate and stable than the old Conn, easier to use, and with a warranty.
Yes, I realize this is an ancient thread.
have you measured this drift in oscillator freq? In servicing these, especially as they age, I often have to replace the motor caps. Dry caps could make the motor drive unstable. And keeping fresh oil on the motor shaft also prevents it from dragging. And allow the wheel space back from the window to reduce air drag.
There may well be tons of these out there, but I doubt there is much of a market for a replacement board. COnsider that for the price of said board and the labor for a tech to install it, you could have a brand new tuner, more accurate and stable than the old Conn, easier to use, and with a warranty.
Yes, I realize this is an ancient thread.
Hi Enzo indeed you're right ... but just wanting this instrument, for fashion, and just to have fun ... it's interesting to change this oscillator with another one more stable.
All the problematics you told are true ... but consider that Peterson still producing turning wheel tuners for professional use ... with a series of motors and so ... it means that some kind of precision is reached ... in order to compensate little changes in rotation, it's needed to have some kind of feedback that here in the ST-11 there is not ... as a turn counter made by otpical disk for instance. At that time, there were photocells and lamps .. so that it was possible since early '70s ... but I don't find this ... the engine rotates supposing that it's own speed stills constant all the time ...
Then I have even a question: what's the frequency of the oscillator ?
And also: what is the ratio of this synchro motor ? I mean: If I drive it with 50 Hz, how many turns per minute I will get ? 500 ? 5 000 ? 600 ? 6 000 ? Not clue ... those data will help to improve this circuitry maintaining the old fashion estetic ... what do you think ? Is it possible to get those data somewhere ?
here where to buy an high precision oscillator with the correct frequency: http://www.icmfg.com/oscillators.html
Last edited:
One could buy a higher freq oscillator and divide it down. But i think the first step is to accurately quantify the instability in the existing circuit. Just how much is there really? And is the power supply stable or does it drift?
Unless the existing dividers are also unstable, all you really need is a master oscillator to replace the RC one. Then use the existing divider matrix to get the individual speeds.
I brought up things like mechanicals and motor caps because I don;t automatically assume speed instability is caused by the oscillator. It may well be, but I don't assume that as a starting point.
And while we are at it, all those ICs are in sockets, are they not? And over the decades some of the socket contacts may be getting resistive.
If you want it as a project fine, then marketing is irrelevant.
I have several of the old 12-wheel COnn 6T5s here. A couple even work. Talk about projects. Tube driven tunig fork oscillator as the basis.
Peterson is shifting to LCD screens with a graphic representation of the wheel pattern. A moving pattern on the screen works the same way as the image on the spinning disc.
I have no idea what RPM the wheel spins on each note. Take an existimng unit and measure. it may be drifiting, but you should be able to empirically get the data you want.
Unless the existing dividers are also unstable, all you really need is a master oscillator to replace the RC one. Then use the existing divider matrix to get the individual speeds.
I brought up things like mechanicals and motor caps because I don;t automatically assume speed instability is caused by the oscillator. It may well be, but I don't assume that as a starting point.
And while we are at it, all those ICs are in sockets, are they not? And over the decades some of the socket contacts may be getting resistive.
If you want it as a project fine, then marketing is irrelevant.
I have several of the old 12-wheel COnn 6T5s here. A couple even work. Talk about projects. Tube driven tunig fork oscillator as the basis.
Peterson is shifting to LCD screens with a graphic representation of the wheel pattern. A moving pattern on the screen works the same way as the image on the spinning disc.
I have no idea what RPM the wheel spins on each note. Take an existimng unit and measure. it may be drifiting, but you should be able to empirically get the data you want.
You right but I cannot get an existing unit to make the misures I need ... you right for the new electronics devices also Turbo Tuner is a solution extremely valid but as I told: the fashon of this object is higher 
... the problem is to have more precision than what it has really and ... for me it's an hobby ... to get fun rebuilding oscillator ... but ok it looks alike a "industrial secret" so I will not do ... and I will simply merely buy a new generation one :-/ ... sometimes enthusiasms are frozen in this way
... the problem is to have more precision than what it has really and ... for me it's an hobby ... to get fun rebuilding oscillator ... but ok it looks alike a "industrial secret" so I will not do ... and I will simply merely buy a new generation one :-/ ... sometimes enthusiasms are frozen in this way Hard to say exactly what's causing the drift, but it seems like it's the crude oscillator circuit which is adjustable with a 100 ohm wire-wound pot. It just doesn't work well. It never did according to an old technician from Conn.
Changing the oscillator to a more stable one probably wouldn't be real difficult. It's doing it while retaining the adjustability without making major changes to the unit.
The motor in the ST-11 is a 24 V - 60 HZ - 1200 RPM synchronous motor. A while back I had measured some of the frequencies of the voltage applied to the motor while troubleshooting some of them, but I don't recall.
By the way Petersons virtual strobes are just that, virtual strobes. I have a Turbo-Tuner which is a real strobe. Great unit by the way. The best hand-held tuner I've ever used. Can't say enough about them.
Something about using an old Conn is cool though. Some professionals still use the ST-11.
Changing the oscillator to a more stable one probably wouldn't be real difficult. It's doing it while retaining the adjustability without making major changes to the unit.
The motor in the ST-11 is a 24 V - 60 HZ - 1200 RPM synchronous motor. A while back I had measured some of the frequencies of the voltage applied to the motor while troubleshooting some of them, but I don't recall.
By the way Petersons virtual strobes are just that, virtual strobes. I have a Turbo-Tuner which is a real strobe. Great unit by the way. The best hand-held tuner I've ever used. Can't say enough about them.
Something about using an old Conn is cool though. Some professionals still use the ST-11.
Attachments
The oscillator is the oscillator. The divider circuits don;t care what generated the signal. Changing the oscillator itself should have no effect on adjustability.
Put a freq meter on the old master oscillator and see what it does.
http://mmd.foxtail.com/Smythe/Conn_Strobotuner_ST-11_Manual.pdf
Put a freq meter on the old master oscillator and see what it does.
http://mmd.foxtail.com/Smythe/Conn_Strobotuner_ST-11_Manual.pdf
The ST-11 is adjustable plus or minus 50 cents using the "cents" pot. If you want to retain the ability to to calibrate to non standard (440) tuning, the oscillator needs to be adjustable. This is the problem with just changing to a crystal oscillator.
Searching for ST-11 info got me here. The Conn users forum that was on the Peterson Electro-Musical Site is defunct.
I suppose by now the OP has built a stable master oscillator for his unit, I doubt if my information will help him, but maybe it will help someone else.
I looked at the ST-11 schematic and made a few measurements on the circuit. The Model ST-11 uses two 74121 dual one-shots to generate a 6.825usec clock pulse for a chain of six 7476 dual J-K flip flops which are configured as a 12-bit binary presettable ripple counter. When the most significant bit of the counter toggles, the counter is preset (reloaded) with a number, the value of this number is determined by the CAL switch, or if not in CAL then by the note selection switch. The twelfth (most significant) bit of the counter is always reloaded to 1, so only the lower 11 bits change according to switch position. The output of the counter at the MONITOR pin (pin 10 of the last 7476) is a square wave. The frequency of the square wave determines the motor speed.
In a positive-logic implementation, the amount of clock pulses needed to count from the preloaded count to the terminal count is the difference between those numbers. In this negative-logic implementation, the number of counts needed is the difference between the terminal count and the one's complement of the preloaded number. Simplified, this is the same as loading a number into the counter and then counting from that number down to zero. Ignoring the 12th bit, the numbers loaded into the counter are as follows:
Note Hz Count tw,msec Motor Hz Motor RPM
C 261.6 5D6 20.39 49.036 980.724
C# 277.2 582 19.25 51.957 1039.15
D 293.7 533 18.17 55.041 1100.83
D# 311.1 4E8 17.14 58.328 1166.56
E 329.6 4A2 16.19 61.771 1235.41
F 349.2 45F 15.27 65.469 1309.38
F# 370 420 14.41 69.375 1387.50
G 392 3E5 13.61 73.481 1469.61
G# 415.3 3AD 12.84 77.853 1557.07
A 440 378 12.12 82.500 1650.00
A# 466.2 346 11.44 87.423 1748.45
B 493.9 317 10.80 92.617 1852.34
CAL 120 32E 11.11 90.000 1800.00
The Hurst Model CA hysteresis synchronous motor used in this instrument has a standard speed rating of 1200 rpm for a 60Hz drive signal. The pattern wheel mounted to the shaft has two fundamental patterns per turn. For example, on the CAL setting 1800 rpm is 30rps, so the 2X fundamental pattern is occurring 60 patterns per second. The CAL light is pulsed 120 times per second, The ratio of CAL light pulses to fundamental pattern is 2:1. For tones, the ratio of note frequency to pattern frequency is 8:1.
Edit: apparently this editor deleted all of the column-separating TABs in my data. Anybody know how to fix that?
I suppose by now the OP has built a stable master oscillator for his unit, I doubt if my information will help him, but maybe it will help someone else.
I looked at the ST-11 schematic and made a few measurements on the circuit. The Model ST-11 uses two 74121 dual one-shots to generate a 6.825usec clock pulse for a chain of six 7476 dual J-K flip flops which are configured as a 12-bit binary presettable ripple counter. When the most significant bit of the counter toggles, the counter is preset (reloaded) with a number, the value of this number is determined by the CAL switch, or if not in CAL then by the note selection switch. The twelfth (most significant) bit of the counter is always reloaded to 1, so only the lower 11 bits change according to switch position. The output of the counter at the MONITOR pin (pin 10 of the last 7476) is a square wave. The frequency of the square wave determines the motor speed.
In a positive-logic implementation, the amount of clock pulses needed to count from the preloaded count to the terminal count is the difference between those numbers. In this negative-logic implementation, the number of counts needed is the difference between the terminal count and the one's complement of the preloaded number. Simplified, this is the same as loading a number into the counter and then counting from that number down to zero. Ignoring the 12th bit, the numbers loaded into the counter are as follows:
Note Hz Count tw,msec Motor Hz Motor RPM
C 261.6 5D6 20.39 49.036 980.724
C# 277.2 582 19.25 51.957 1039.15
D 293.7 533 18.17 55.041 1100.83
D# 311.1 4E8 17.14 58.328 1166.56
E 329.6 4A2 16.19 61.771 1235.41
F 349.2 45F 15.27 65.469 1309.38
F# 370 420 14.41 69.375 1387.50
G 392 3E5 13.61 73.481 1469.61
G# 415.3 3AD 12.84 77.853 1557.07
A 440 378 12.12 82.500 1650.00
A# 466.2 346 11.44 87.423 1748.45
B 493.9 317 10.80 92.617 1852.34
CAL 120 32E 11.11 90.000 1800.00
The Hurst Model CA hysteresis synchronous motor used in this instrument has a standard speed rating of 1200 rpm for a 60Hz drive signal. The pattern wheel mounted to the shaft has two fundamental patterns per turn. For example, on the CAL setting 1800 rpm is 30rps, so the 2X fundamental pattern is occurring 60 patterns per second. The CAL light is pulsed 120 times per second, The ratio of CAL light pulses to fundamental pattern is 2:1. For tones, the ratio of note frequency to pattern frequency is 8:1.
Edit: apparently this editor deleted all of the column-separating TABs in my data. Anybody know how to fix that?
Last edited:
So the Conn uses an RC clock made from 74LS121 mono stables ?
Isn't this text book of how not to build a Clock ?
The unit is about 40 years old.
Its pretty well vintage - so I would be concerned about modding it.
I remember one of my teachers, explaining how he made a thermally stable RC circuit
using an R and C that had complementary temperature co-efficients.
I'm wondering if this technique could be used with the Conn.
BTW: if you use a dynamics compressor with the Conn unit, it will make the tuning much easier.
.
Scooperman,
Thanks for posting the technical info on the Conn ST-11.
Think I bought mine new circa 1979 ?
I guess they are not really collectable* - however, because of its vintage, I wouldn't want to mod the case.
Plus, from what you have described, the design is so archaic - I wouldn't try to mod the circuit.
If the RC clock uses a wire wound resistor - I'm wondering if it would be best to replace the clock's cap(s).
* I knew I should have bought a 1976 red wine Les Paul standard instead of the Conn Tuner.
There are a few decisions in my life I wish I could have back.
.
Thanks for posting the technical info on the Conn ST-11.
Think I bought mine new circa 1979 ?
I guess they are not really collectable* - however, because of its vintage, I wouldn't want to mod the case.
Plus, from what you have described, the design is so archaic - I wouldn't try to mod the circuit.
If the RC clock uses a wire wound resistor - I'm wondering if it would be best to replace the clock's cap(s).
* I knew I should have bought a 1976 red wine Les Paul standard instead of the Conn Tuner.
There are a few decisions in my life I wish I could have back.
.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.