Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

donate to eham
   Home   Help Search  
Pages: [1] 2 Next   Go Down
  Print  
Author Topic: Tuned Grid Tuned Plate Oscillator question  (Read 6453 times)
NR5P
Member

Posts: 159




Ignore
« on: July 01, 2004, 09:46:02 PM »

I've been reading some older books lately and the tuned grid tuned plate oscillator circuit puzzles me as to how it works, the books I've read don't go in to great detail on this circuit.  I understand there is capacitance in between the grid and the plate but how does this have to do with feeding back to the grid?  It shows a tank circuit on the grid as normal and a tank ciruit on the plate tuned to the same frequency.

I've tried and tried to figure this out on my own but its bugging me.
Logged
WIRELESS
Member

Posts: 1




Ignore
« Reply #1 on: July 01, 2004, 10:02:03 PM »

You must have seen circuits that have caps between the output of one tube, transistor, ect. and the input of another device.  Right?  Caps pass rf current. If you connect the output of an active device (tube) to its own input, it will break into oscillation especially if the input/outputs have tuned circuits.
Logged
WIRELESS
Member

Posts: 1




Ignore
« Reply #2 on: July 01, 2004, 10:13:08 PM »

I forgot to add that an oscillator is really trying to amplify its own output (sometimes via interelectrode capacitance).  Stable oscillators correctly designed will use components connected so that rf out is fed back to the input at the correct phase.  But any active device can oscillate all by itself via interelectrode or stray coupling.
Logged
AA4PB
Member

Posts: 14360




Ignore
« Reply #3 on: July 01, 2004, 10:15:11 PM »

The capacitor couples a little of the signal from the plate circuit back to the grid so as to sustain oscillation. You will see some Tuned Grid Tuned Plate oscillators that do not have a capacitor. The interelectrode capacitance of the tube itself provides the feedback path.
Logged

Bob  AA4PB
Garrisonville, VA
W7DJM
Member

Posts: 1




Ignore
« Reply #4 on: July 01, 2004, 11:36:46 PM »

You have just discovered why "some" amplifiers need to be neutralized!!!


Type "tuned plate tuned grid" into Google, just like that, in quotes.


Also, type in       tptg, transmitter, and
                   
                    tptg, oscillator  
Logged
VE6XX
Member

Posts: 3




Ignore
« Reply #5 on: July 02, 2004, 03:41:29 AM »

Sir:  I am uncertain as to what exactly you fail to comprehend regarding the "tptg" osc.
Basically, the oscillator is a reliable starter with good freq control & a clean o/p because of the dual tuned ccts. I offer the following simplified explanation as I understand the circuit.
On power-up, plate current starts to flow & incites oscillation in the plate tank. An external cap or the interelectrode capacity feeds a sample of the circulating current from the plate tank back to the grid cct in phase,which boosts the oscillatory current in the grid tank, causing the voltage across the grid tank to exceed the bias & causing the tube to conduct.
Remember, this being an oscillator, the plate conduction angle is less than 90 degrees ( about 70 deg) & the plate current is in the form of short duration high current pulses which shock excite the plate tank at its tuned freq. The "flywheel" effect of the plate tank maintains oscillation until the next current pulse. The cap feeds back to the grid, the plate current pulse shock excites the plate tank, the "flywheel" effect sustains oscillation & on it goes. Please remember this is a "basic" explanation & fails to address a number of technical parameters.
"Basic Electronics" by Grobb  & "Electronics for Communications" by Kennedy are both older textbooks that are excellent references for tutorials.
I hope that my explanation, basic as it was, may have been of aid to you.
CHEERS!  Brian, VE6XX
Logged
W8JI
Member

Posts: 9749


WWW

Ignore
« Reply #6 on: July 02, 2004, 06:07:58 AM »

The TPTG oscillator, like ANY oscillator, depends on positive or regenerative feedback. It must be a LINEAR system, not a pulse amplifier with flywheel, in order to start oscillating without external stimulus and to maintain oscillation.

It is NOT a stable amplifier, because it employs TWO  tuned circuits that must be set at slightly different frequencies to maintain oscillation. Also the two circuits drift at different rates making temperature compensation virtually impossible. They are generally less stable and not any cleaner than other better designs. They also do not allow for much energy transfer to a load, and they have virtually no load isolation for the frequency determining components. They are bad news.  

That is why the TPTG oscillator quickly fell out of favor compared to single resonant circuit designs.  

The gain in the amplification must exceed the loss in the feedback, and the phase must be regenerative.

In a tube, when the grid pulls positive the anode drops in voltage (goes more negative). Normally feedback from anode to grid is negative, unless something adds phase shift. The tuned circuits and very tiny feedback capacitance (high reactance)does this.  

If you could visualize the operation starting at the anode, tube noise causes random variations in plate current. This translates to slight variations in voltage across the very lightly loaded anode (parallel tuned with very very light coupling to the load). This noise peak couples back to the input via anode to grid capacitance, which is generally small enough phase shifts someplace less than 90 degrees lead. To get a bit more phase shift, the grid circuit is normally tuned slightly higher in frequency than the anode. This makes the grid inductive at the peak gain frequency of the anode. The combination of series C and shunt L in the path to the grid results in increased feedback phase advance. It is like an "L" network with very high series C and modest shunt L.

At some frequency between where the anode is resonant (peak gain) and where the grid is resonant (another DIFFERENT frequency for peak gain) the circuit can have the correct phase advance and gain to result in an oscillation. Since the two circuits must ALWAYS be on different frequencies, the oscillator always wants to move around a good bit.

I recently built a TPTG oscillator, buffer, driver, and 100T PA, and despite using great care and excellent regulation still have constant problems with chirp, oscillator pulling, and unexpected frequency jumping.

The oscillator moves considerably even though back a few stages when the PA is tuned, and that is with EVERY voltage regulated on the oscillator and buffer!!

I did work a few VK's with it on 160 meters while using a regenerative receiver, but then my fascination went away. I think I'll change to a clapp oscillator. Hi hi.

73 Tom
Logged
NR5P
Member

Posts: 159




Ignore
« Reply #7 on: July 02, 2004, 10:53:23 AM »

Thanks,   What was confusing me was that I thought it would be negative feedback.

Thanks, Nathan
Logged
K6KV
Member

Posts: 5




Ignore
« Reply #8 on: October 26, 2017, 08:32:09 AM »

It's been thirteen years since any posts, however if anyone stumbles onto this, I'll mention that the TGTP oscillator was a much loved transmitter in the 1920's.  I built one recently and described it in detail in the October 2009 issue of the AWA Journal.  It's operation is ill understood because the plate-grid capacitance does not provide the 180 degree phase reversal needed to sustain oscillation.  Indeed the ARRL Handbook and experts like Terman just hand-waved how it worked, "as if by magic".  So my oscillator uses two type 10 tubes, works like a dream, develops 25W on 40M, and will be used in the upcoming 2017 AWA 1929 radio contest.  K6KV.
Logged
W1VT
Member

Posts: 2529




Ignore
« Reply #9 on: October 26, 2017, 09:21:55 AM »

Understanding how oscillators work is a challenging topic.
Here is one of the better books--used copies may be quite affordable.

Communication Circuits: Analysis and Design Hardcover – June, 1971
by Kenneth K. Clarke (Author), Donald T. Hess (Author)
Logged
KM1H
Member

Posts: 2644




Ignore
« Reply #10 on: October 26, 2017, 06:49:39 PM »

An amplifier was explained to me many decades ago as a failed oscillator.

Carl
Logged
G8HQP
Member

Posts: 612




Ignore
« Reply #11 on: October 27, 2017, 04:41:43 AM »

Take an inverting amplifier, say a BJT or valve. Put an impedance Zi across the input, Zo across the output, and Zf as feedback from output to input. Then it can be shown that to get oscillation you need one of the following situations:
1. Zi and Zo are capacitive impedances, Zf is inductive. (Colpitts)
2. Zi and Zo are inductive impedances, Zf is capacitive. (Hartley)
In addition, there must be enough gain.

The TPTG is a form of Hartley oscillator. Zf is stray and parasitic capacitance. Zi and Zo are inductive, because the two tuned circuits are both slightly off-tune - I am not certain that they have to be tuned to different frequencies, although in reality they usually will be. Any external load applied to either tuned circuit will modify the impedance and so cause frequency change. There are other oscillators which have much better isolation between the frequency determining elements and the external load.
Logged
VK6HP
Member

Posts: 186




Ignore
« Reply #12 on: October 27, 2017, 05:04:09 AM »

I happened to notice this simplified explanation of the TPTG oscillator the other day:

http://owenduffy.net/blog/?p=11594

Very useful, but perhaps a bit terse.
Logged
G3RZP
Member

Posts: 8149




Ignore
« Reply #13 on: October 27, 2017, 09:03:44 AM »

The feedback from the plate to the grid through internal capacitance leads to parameter changes, all caused by Miller effect. One change is that input resistance can become negative, and the other is that the input susceptance  can go either positive or negative. Which effects occur depend on the impedance presented to the plate circuit The input resistance is negative when the plate circuit is inductive i.e. tuned higher in frequency than the grid tuned circuit, and so provided that the dynamic resistance of the grid tuned circuit is such that the negative input resistance of the tube in parallel gives a net negative resistance, oscillation follows.

See 'Vacuum Tube Oscillators' by Edson.
Logged
K6KV
Member

Posts: 5




Ignore
« Reply #14 on: November 06, 2017, 07:07:06 PM »

I'm in agreement with the early Radio Amateur Handbooks, that the TPTG oscillator was very popular around 1930, and indeed works very well (mine certainly does).   I want however to respond to what I believe is some misinformation.   The TPTG circuit is not like a Hartley because there is no inductive coupling between the plate and grid tuned circuits.  It's also not necessary to tune the grid tank carefully; I speak from experience that it can be tuned a long ways to either side of resonance.  If anyone would like to see a thoughtful two-page analysis of the circuit I hope they will email me using my call letters at arrl.net.  Having built both push pull Hartley and TPTG transmitters, I would rate them as equal great performers.   My preference however is for the TPTG  because the grid tuning acts like a band-spread control, and because construction is simplified due to not needing to tap the plate coil.   
Logged
Pages: [1] 2 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!