My statement of series resonance oscillators seems to be forgotten. Why is that nobody uses it? It's rather easy to order crystals for that and as I understand the crystal is more controlled.
http://www.diyaudio.com/forums/showthread.php?postid=281098#post281098
Picture here, simplified of course.
http://www.diyaudio.com/forums/attachment.php?s=&postid=281098
http://www.diyaudio.com/forums/showthread.php?postid=281098#post281098
Picture here, simplified of course.
http://www.diyaudio.com/forums/attachment.php?s=&postid=281098
P-A
Just for my understanding a couple of questions.
1: Where do you tap the clock signal in the image shown?
2: Are there any spice models of crystals?
Just for my understanding a couple of questions.
1: Where do you tap the clock signal in the image shown?
2: Are there any spice models of crystals?
1 You can take the signal at the collector of the main transistor or from te JFET but you must have an emitter follower if you don't want to load too much.
2 Yes, but I think someone here can give you that information. Netlist, where are you? 🙂
2 Yes, but I think someone here can give you that information. Netlist, where are you? 🙂
Re: Re: Re: The best(?) oscillation mode is series resonance
? 
peranders said:
You will get them...
? 
Here we go again...........
Herr Per on a fishing expedition......"just collecting the facts"........probably so he can start building clock kits. Right???
You really don't expect me to help you, do you???
Hey, Lars, don't look behind you............someone else wants to make clock kits.
Only it won't be me. But you will be able to say that I was the first to warn you.
Jocko
Herr Per on a fishing expedition......"just collecting the facts"........probably so he can start building clock kits. Right???
You really don't expect me to help you, do you???
Hey, Lars, don't look behind you............someone else wants to make clock kits.
Only it won't be me. But you will be able to say that I was the first to warn you.
Jocko
Jocko, in a positive mood these days. Suppose I plan to make an oscillator and also plan to make it in SMD, how many would be interested do you think? Just compare the ALW group buy and my SMD regulator. I think Lars can be quite calm about the competition if there was any. SMD is not for the average DIY'er, that much do I know.
Futhermore I think SMD is the way to go when it comes to MHz circuits and fast rise times.
This thread did I start only because this type of oscillator seems to be rather unknown. Maybe Elso can improve his clock although his is the best already.
Futhermore I think SMD is the way to go when it comes to MHz circuits and fast rise times.
This thread did I start only because this type of oscillator seems to be rather unknown. Maybe Elso can improve his clock although his is the best already.
Geared toward the average rather than the exceptional
"SMD is not for the average DIY'er, that much do I know."
WHAT!!!!?????? I thought it was no sweat to work with.........
If you stay with 1206 and larger passives it really isn't that bad. Maybe it wasn't the fact that your regulator is SMT that is the reason for the greater popularity of the AWL regulators. Maybe it was the fact that he spent a very large amount of time and effort in designing and optimizing the circuit and even listening to changes in the design. Or the fact that he has been answering questions on regulator design instead of asking them on the forum. Na......... that couldn't be it.
Maybe someone will come along with a design based on the work that others have done, but with some unique design approaches to further develop the regulators performance. Maybe they will use SMT where there is an advantage in the circuit and not just for the exercise of doing the whole thing in SMT just to be different.
Maybe someone will design a regulator with the option of series prereg (something better than that dinosaur the 317) followed by the option for a shunt, series, or dual tracking regulator with an active ground derived from a single input voltage. Maybe someone will design a supply with small dip breadboard section of the PCB for configuring the reference voltage circuit of their choice: voltage regulator, series voltage reference, CCS and shunt, TL 491.... ect. Maybe the provision to run the op amp rails greater than the regulated output voltage for greater gain bandwidth or with smaller rails than the regulators output voltage to allow regulated voltages greater than the supply limits of the op amp. If only someone who knew enough to go beyond fairly close copies of what's already been done would come along. I wonder where we could find someone like that? If we could just find a guy smart enough to do a new design (for free of course) and yet stupid enough not to know that someone on the forum (or even another DIY forum) will copy it and sell it as their own work. For instance a forum member who would put patented by (their name) on a copy of someone else's board layout, or would use the name of an another designer to promote their effort without his OK and anger him enough not to participate in the forum anymore.
"SMD is not for the average DIY'er, that much do I know."
WHAT!!!!?????? I thought it was no sweat to work with.........
If you stay with 1206 and larger passives it really isn't that bad. Maybe it wasn't the fact that your regulator is SMT that is the reason for the greater popularity of the AWL regulators. Maybe it was the fact that he spent a very large amount of time and effort in designing and optimizing the circuit and even listening to changes in the design. Or the fact that he has been answering questions on regulator design instead of asking them on the forum. Na......... that couldn't be it.
Maybe someone will come along with a design based on the work that others have done, but with some unique design approaches to further develop the regulators performance. Maybe they will use SMT where there is an advantage in the circuit and not just for the exercise of doing the whole thing in SMT just to be different.
Maybe someone will design a regulator with the option of series prereg (something better than that dinosaur the 317) followed by the option for a shunt, series, or dual tracking regulator with an active ground derived from a single input voltage. Maybe someone will design a supply with small dip breadboard section of the PCB for configuring the reference voltage circuit of their choice: voltage regulator, series voltage reference, CCS and shunt, TL 491.... ect. Maybe the provision to run the op amp rails greater than the regulated output voltage for greater gain bandwidth or with smaller rails than the regulators output voltage to allow regulated voltages greater than the supply limits of the op amp. If only someone who knew enough to go beyond fairly close copies of what's already been done would come along. I wonder where we could find someone like that? If we could just find a guy smart enough to do a new design (for free of course) and yet stupid enough not to know that someone on the forum (or even another DIY forum) will copy it and sell it as their own work. For instance a forum member who would put patented by (their name) on a copy of someone else's board layout, or would use the name of an another designer to promote their effort without his OK and anger him enough not to participate in the forum anymore.
Fred, I suppose you have nothing to say about series resonance oscillators?
The hard part of SMD is that you
* MUST be able to solder
* keep the parts in ORDER, because they are ummarked in many cases and when the do have text, the text is too small if you are a bit older.
Those two aspects are the hardest to live up to, not the small parts itself. 0406 is quite easy to solder but IC with 0.63 or 0.5 mm is very much harder.
About patent: The patented design I have made pcb's for is made by permission from the patent holder and you can also read about and also the conditions for the usage of the pcb's.
The regulator you talk about is not patented as far as I know.
The hard part of SMD is that you
* MUST be able to solder
* keep the parts in ORDER, because they are ummarked in many cases and when the do have text, the text is too small if you are a bit older.
Those two aspects are the hardest to live up to, not the small parts itself. 0406 is quite easy to solder but IC with 0.63 or 0.5 mm is very much harder.
About patent: The patented design I have made pcb's for is made by permission from the patent holder and you can also read about and also the conditions for the usage of the pcb's.
The regulator you talk about is not patented as far as I know.
Re: Geared toward the average rather than the exceptional
Seriously, it would be interesting to see what you could come up with. I say this not to tease go, I mean it.
Ok, back to series resonance oscillators. That much do I know that it's very unsuitable to tune crystals towards series resonance if they where made for parallel resonance. The Q factor of the crystal will be really bad and the stability very poor. I took a 4.194 MHz parallel and placed it in a series resonance oscillator. The result was a very temperature sensitive oscillator with much drift.
And now... I have got the paper saying how much you decrease the performance.
Fred, seems like a good idea. Maybe you should start to design the ultimate regulator. Maybe it could be great, even greater than Elso'sFred Dieckmann said:
Seriously, it would be interesting to see what you could come up with. I say this not to tease go, I mean it.
Ok, back to series resonance oscillators. That much do I know that it's very unsuitable to tune crystals towards series resonance if they where made for parallel resonance. The Q factor of the crystal will be really bad and the stability very poor. I took a 4.194 MHz parallel and placed it in a series resonance oscillator. The result was a very temperature sensitive oscillator with much drift.
And now... I have got the paper saying how much you decrease the performance.
Codswallop.......
Only if you try to operate on the wrong frequency, which you obvioulsy tried.
It is the same crystal for either circuit. Only the frequency will be different.
No more design help from me.........
Jocko
Only if you try to operate on the wrong frequency, which you obvioulsy tried.
It is the same crystal for either circuit. Only the frequency will be different.
No more design help from me.........
Jocko
Indeed, I wanted to have 4194304 Hz = 2*2*2*2*2*2*2*2*2*2*2*2*2*2*2*2*2*2*2*2*2*2 => 1 Hz from a 22 bit counter
My main concern at that time was accurancy because it was for a ordinary clock, big one!
My main concern at that time was accurancy because it was for a ordinary clock, big one!
The guys that make the crystals don't care what you do with them.......it is all the same to them. Just order the right one for the circuit.......and specify the load. It isn't that hard.
Jocko
Jocko
At that time I was a poor student and I had a junk box... and I have one rarity.. a glass tube crystal, same shape a a regular vacuum tube. This crystal is really nice looking with a little of gold inside. Maybe I'll take a photo of it? The crystal has been inside a thermos, OCXO from the 60's.
"a glass tube crystal, same shape a a regular vacuum tube."
I've seen many of those crystals. They were used in Broadcast transmitters. Vacuum-sealed for better frequency stability.
I've seen many of those crystals. They were used in Broadcast transmitters. Vacuum-sealed for better frequency stability.
The series resonance was discussed here two years ago in great detail:
http://www.diyaudio.com/forums/showthread.php?postid=19523#post19523
Basic but important info can be found here:
http://www.mwrf.com/Articles/Print.cfm?ArticleID=5510
http://www.euroquartz.co.uk/pdf/application-notes.pdf
http://e-www.motorola.com/files/netcomm/doc/app_note/AN2049.pdf
http://ww1.microchip.com/downloads/en/AppNotes/00849a.pdf
http://www.emi.dtu.dk/education/master_courses/31415/Ch6Oscil03.pdf
Aha, and if someone just looking for a smart guy who will do a new free design I feel that Phred is the right person. 😀
http://www.diyaudio.com/forums/showthread.php?postid=19523#post19523
Basic but important info can be found here:
http://www.mwrf.com/Articles/Print.cfm?ArticleID=5510
http://www.euroquartz.co.uk/pdf/application-notes.pdf
http://e-www.motorola.com/files/netcomm/doc/app_note/AN2049.pdf
http://ww1.microchip.com/downloads/en/AppNotes/00849a.pdf
http://www.emi.dtu.dk/education/master_courses/31415/Ch6Oscil03.pdf
Aha, and if someone just looking for a smart guy who will do a new free design I feel that Phred is the right person. 😀
You expect everything to be handed out for free?????
You need to tear the crystal out of a junk ISA card, and measure the Z vs frequency of a typical crystal.
Jocko
You need to tear the crystal out of a junk ISA card, and measure the Z vs frequency of a typical crystal.
Jocko
- Status
- Not open for further replies.