UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > General Vintage Technology > Components and Circuits

Notices

Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets.

Reply
 
Thread Tools
Old 9th Jan 2019, 7:33 pm   #61
Argus25
Nonode
 
Argus25's Avatar
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,115
Default Re: Puzzling audio circuitry

Quote:
Originally Posted by G8HQP Dave View Post
You are confusing two things:
1. a simple model to calculate approximately what the circuit does
2. an explanation of how the circuit works
Only to the extent that I was trying to point out that, despite the lower device having no voltage gain (as measured across its input and output terminals) its transconductance is responsible for the conversion of signal voltage input into anode signal current (and the current of the series circuit that comprises cascode) and therefore the signal current in the load, regardless of the presence of the upper device (for low frequencies at least). But I agree with all your points too.
Argus25 is online now   Reply With Quote
Old 9th Jan 2019, 9:19 pm   #62
Argus25
Nonode
 
Argus25's Avatar
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,115
Default Re: Puzzling audio circuitry

Quote:
Originally Posted by G8HQP Dave View Post
No, the transconductance gain of the circuit is attributable to the lower transistor/valve/FET. The voltage gain comes from both, except perhaps for a BJT cascode where all the voltage gain comes from the upper device.
Regarding the upper device as a common base or common grid amplifier, it is true that this configuration is said to have a high voltage gain, low current gain (usually near unity). These remarks though relate to driving it with a voltage source and comparing the input & output voltages across the load.

In Cascode of course, the drive is primarily a current source (or sink) from the lower valve (or transistor) and the dynamic current signal (derived from the lower device's transconductance) is in no way "amplified" by the upper device.

The dynamic current signal is converted to a dynamic voltage in the load resistance in the usual way. Thinking about it this way, it is hard to attribute any "gain" to the upper device (in the application) even though measuring across its 3 terminals (as RW pointed out) it is behaving in the usual way as a transconductance amplifier, but its current is being controlled by the lower device.
Argus25 is online now   Reply With Quote
Old 9th Jan 2019, 9:31 pm   #63
G8HQP Dave
Dekatron
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,737
Default Re: Puzzling audio circuitry

As I keep saying, the lower device has voltage gain if you use a big enough anode resistor. Just to pick some numbers at random, assume gm of 5mA/V and mu of 20. An anode resistor of 20k means a voltage gain of around 100. 20 of that comes from the upper valve, and 5 of that comes from the lower valve, approximately. 10mV in would become 50mV at the 'join' and then 1V at the output. What would you call an amplifying stage with 10mV in and 50mV out (or 50mV in and 1V out)? I would call it a voltage amplifier.

I have never said that the current is amplified by the upper device. I, and others, have said repeatedly that the voltage is amplified by the upper device.

You seem fixed on the idea that a triode is a transconductance device. It is also a voltage amplifier. You need to use the most appropriate model, not just stick to just one all the time. Your simplified model of the cascode will not enable you to derive the full equations you quoted; to do that you need to take account of everything which is going on.

Last edited by G8HQP Dave; 9th Jan 2019 at 9:32 pm. Reason: typo
G8HQP Dave is offline   Reply With Quote
Old 9th Jan 2019, 9:49 pm   #64
Argus25
Nonode
 
Argus25's Avatar
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,115
Default Re: Puzzling audio circuitry

Quote:
Originally Posted by G8HQP Dave View Post
Your simplified model of the cascode will not enable you to derive the full equations you quoted; to do that you need to take account of everything which is going on.
Yes I agree and the valves are not ideal transconductance amplifiers either and that is indicated by the presence of the plate resistances in the equations (and of the valves) which alter the the voltages seen on the actual anodes and affect the gain calculations accordingly.
Argus25 is online now   Reply With Quote
Old 9th Jan 2019, 11:04 pm   #65
Argus25
Nonode
 
Argus25's Avatar
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,115
Default Re: Puzzling audio circuitry

Quote:
Originally Posted by G8HQP Dave View Post
You seem fixed on the idea that a triode is a transconductance device. It is also a voltage amplifier.
I think you are right about this too. I have always been stuck on this notion, primarily because the conversion of anode current to a voltage (except for that lost in plate resistance) occurs outside the valve in the load.

Also I have been pretty well stuck on the idea of a transistor as a current amplifier all my life too, though I have a lot less difficulty thinking of them as transconductance devices by programming their base-emitter voltage than I do thinking of a valve as a voltage amplifier. So maybe I'm 3/4 of the way there!
Argus25 is online now   Reply With Quote
Old 10th Jan 2019, 11:21 am   #66
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 4,912
Default Re: Puzzling audio circuitry

Quote:
Originally Posted by Argus25 View Post
Yes. I guess I was just struggling with the notion that a voltage gain figure for the lower valve on its own (its dynamic plate to dynamic grid voltage) could even be specified or have any meaning in the Cascode configuration, since the plate was held to a near fixed potential by the upper valve. Of course in transistor Cascode circuits the collector of the lower transistor has almost perfect clamping to a fixed potential and better than in the valve case.
Hopefully we've resolved this - and it could be that Horowitz and Hill has misled you a bit Hugo - but the transistor cascode has the 'join' no more fixed than the valve cascode.

The lower device does not have zero voltage gain, it has unity gain, assuming the gm's of the two devices are equal.

Quote:
Originally Posted by G8HQP Dave View Post
As I keep saying, the lower device has voltage gain if you use a big enough anode resistor. Just to pick some numbers at random, assume gm of 5mA/V and mu of 20. An anode resistor of 20k means a voltage gain of around 100. 20 of that comes from the upper valve, and 5 of that comes from the lower valve, approximately. 10mV in would become 50mV at the 'join' and then 1V at the output. What would you call an amplifying stage with 10mV in and 50mV out (or 50mV in and 1V out)? I would call it a voltage amplifier.
That's a particularly useful illustration, Dave!

Quote:
Originally Posted by Argus25 View Post
Yes I agree and the valves are not ideal transconductance amplifiers either and that is indicated by the presence of the plate resistances in the equations (and of the valves) which alter the the voltages seen on the actual anodes and affect the gain calculations accordingly.
Valves (triodes) aren't particularly ideal as transconductance devices, no. BJT's are better, from the point of view of the output. But don't forget that BJT's have a rather lousy input characteristic, as well as the necessary voltage drive they suck significant base current as a by-product. Triodes don't. If you look as a figure of merit, Rout/Rin, valves win hands down!
kalee20 is offline   Reply With Quote
Old 10th Jan 2019, 12:07 pm   #67
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 4,912
Default Re: Puzzling audio circuitry

Sorry, too late to edit - my figure of merit is rubbish, I posted without thinking it through! good transconductance device with voltage in, current out, has a high resistance at at the input, and output current hardly varies with voltage at output node, so output resistance is high too. And it's helpful to factor in the transfer parameter too. So figure of merit is Rin x Rout x gm. (The triode beats the bipolar junction transistor).
kalee20 is offline   Reply With Quote
Old 10th Jan 2019, 1:12 pm   #68
Argus25
Nonode
 
Argus25's Avatar
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,115
Default Re: Puzzling audio circuitry

Quote:
Originally Posted by G8HQP Dave View Post
Your simplified model of the cascode will not enable you to derive the full equations you quoted; to do that you need to take account of everything which is going on.
This remark made me look at how to derive the equation for the gain of Cascode quoted by Millman & Halkias. Actually I can derive it quite easily except for one aspect that was initially difficult (for me).

To be able to calculate the voltage gain for cascode it is necessary to know what the impedance is looking into the cathode of the upper device (once this is known its dead easy). The best way to do it is to use the principle that the impedance between two nodes in a circuit equals the open circuit voltage divided by the short circuit current.

Doing that, the upper device has an impedance of (rp+R)/(u+1) looking into the cathode. Which also shows how u effectively lowers the input impedance of the upper device.

Once this is known it is all downhill easy, because in the cascode circuit you can then replace the upper device by an impedance of (rp +R)/(u+1) and replace the lower device by a voltage generator u(Vin) in series with an impedance rp.

Hence the generator is u(Vin) and the total load is rp + (rp + R)/(u+1).

Then to calculate the output current it is uVin/(rp + (rp+R)/(u+1) ). Then, multiplying this current by the load resistance R to get Vout, yields their exact equation for Vout/Vin.
Argus25 is online now   Reply With Quote
Old 10th Jan 2019, 3:47 pm   #69
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 4,912
Default Re: Puzzling audio circuitry

Hugo,

That's brilliant. Deriving the formula in the book yourself from first principles gives an insight, and also verifies what is written! I have seen the occasional statement in a book which is wrong. Not often, but it happens. And then, the reader is misled - especially if he just takes it as gospel.
kalee20 is offline   Reply With Quote
Old 10th Jan 2019, 4:08 pm   #70
G8HQP Dave
Dekatron
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,737
Default Re: Puzzling audio circuitry

Quote:
Originally Posted by Argus25
I think you are right about this too. I have always been stuck on this notion, primarily because the conversion of anode current to a voltage (except for that lost in plate resistance) occurs outside the valve in the load.
What if the load is infinite impedance i.e. not actually present? We are talking AC, so no need to worry about DC bias. Where does the 'conversion' of current to voltage take place then? The only impedance there is the anode impedance itself. Is that inside our outside?

Quote:
Originally Posted by Argus25
Doing that, the upper device has an impedance of (rp+R)/(u+1) looking into the cathode. Which also shows how u effectively lowers the input impedance of the upper device.
But u is the raw voltage gain of the valve! How can a device which is not a voltage amplifier have a voltage gain defined?

Quote:
Once this is known it is all downhill easy, because in the cascode circuit you can then replace the upper device by an impedance of (rp +R)/(u+1) and replace the lower device by a voltage generator u(Vin) in series with an impedance rp.
Same comment as above.

So to derive the equations you quoted you assume that the valves are voltage amplifiers with finite impedances.
G8HQP Dave is offline   Reply With Quote
Old 10th Jan 2019, 9:59 pm   #71
Argus25
Nonode
 
Argus25's Avatar
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,115
Default Re: Puzzling audio circuitry

Quote:
Originally Posted by G8HQP Dave View Post
So to derive the equations you quoted you assume that the valves are voltage amplifiers with finite impedances.
We are in agreement on this. The thing though that perhaps I was a bit obsessed with, shown in the final step that is required to complete the task of the equation solution for the overall gain, it is required that the signal current in the series circuit of cascode arrangement, is multiplied by the external load resistance to calculate Vout. I was more concerned about the way the signal current came about by the lower device (focusing on transconductance) and the role of the lower device generating it, vs the upper device. But now I know better, thanks !
Argus25 is online now   Reply With Quote
Old 11th Jan 2019, 1:27 pm   #72
G8HQP Dave
Dekatron
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,737
Default Re: Puzzling audio circuitry

Yes, you can analyse it either way - voltage amp or transconductance. You can even use a mixture of the two. For some reason it took me a long long time to realise that
mu = Ra x gm
applies just as much to a circuit as it does to a device.
G8HQP Dave is offline   Reply With Quote
Old 11th Jan 2019, 6:21 pm   #73
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 4,912
Default Re: Puzzling audio circuitry

Quote:
Originally Posted by Argus25 View Post
I was more concerned about the way the signal current came about by the lower device (focusing on transconductance) and the role of the lower device generating it, vs the upper device. But now I know better, thanks !
Yes. Intuitively the signal current is dependent on the gm of the lower device. The output voltage (and hence the gain) is dependent on the load on the upper device.

The gm of the upper device does have a major impact on how 'still' the join is. And the signal at the join makes a big difference to the Miller fed-back capacitance.
kalee20 is offline   Reply With Quote
Old 14th Jan 2019, 12:53 pm   #74
trobbins
Hexode
 
trobbins's Avatar
 
Join Date: Oct 2009
Location: Melbourne Australia
Posts: 341
Default Re: Puzzling audio circuitry

The lower triode cathode node also has complexity from the feedback voltage and current. It may be remiss to just assess that as a voltage or current feedback node.

Morgan Jones (3rd Ed) had a go at assessing that node in the similar, but probably simpler Williamson input stage - its not a simple set of calculations.
trobbins is offline   Reply With Quote
Old 15th Feb 2019, 6:29 pm   #75
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 4,912
Default Re: Puzzling audio circuitry

I came across the circuit looking through an old Radio Constructor magazine! See attached.

There's no other explanation other that what is shown in the photo, but hopefully all clear. Interesting to see is used in a power output stage!
Attached Thumbnails
Click image for larger version

Name:	image.jpg
Views:	74
Size:	92.4 KB
ID:	178444  
kalee20 is offline   Reply With Quote
Old 18th Feb 2019, 2:40 am   #76
Maarten
Nonode
 
Join Date: Dec 2007
Location: Haarlem, Netherlands
Posts: 2,315
Default Re: Puzzling audio circuitry

It looks like a somewhat more complicated version of the familiar Philips totem pole output stage.
Maarten is offline   Reply With Quote
Old 18th Feb 2019, 3:46 pm   #77
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 4,912
Default Re: Puzzling audio circuitry

It is complicated, but that is largely because it uses pentodes with the complication of a screen-grid supply, suitable decoupled, AC isolated, etc.

If you replace the pentodes with triodes, it simplifies. Then consider removing the upper triode, and have just the lower one. For ease, think of the load as connected to HT+. Then load current flows through the cathode resistor of the (removed) upper valve, creating a voltage drop across it as in the sketch:
Attached Thumbnails
Click image for larger version

Name:	image.jpg
Views:	19
Size:	40.9 KB
ID:	178620  
kalee20 is offline   Reply With Quote
Old 18th Feb 2019, 4:17 pm   #78
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 4,912
Default Re: Puzzling audio circuitry

And then, plug in the upper valve, noting that it gets exactly the drive it needs to deliver an equal current into the load.

It looks as though the lower valve operates into a load RL+Rk and the upper valve RL, but this is not actually the case.
Attached Thumbnails
Click image for larger version

Name:	image.jpg
Views:	15
Size:	43.1 KB
ID:	178621  
kalee20 is offline   Reply With Quote
Reply

Thread Tools



All times are GMT +1. The time now is 6:49 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2019, vBulletin Solutions, Inc.
Copyright ©2002 - 2019, Paul Stenning.