Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives :: Books
Support This Site: Contributors :: Advertise


It is currently Sep Sun 24, 2017 11:32 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 20 posts ] 
Author Message
 Post subject: Cathode follower question
PostPosted: Sep Mon 11, 2017 5:01 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 17506
Location: Warner Robins, GA
While discussing mods to the R-390 in the communications portion of the forum the discussion of audio quality came up and it was mentioned that I could get better quality if I replaced the first audio transformer with the cathode follower circuit the R-390A uses.

Here's the original circuit.

Attachment:
Original circuit.jpg
Original circuit.jpg [ 51.13 KiB | Viewed 319 times ]



Here's the cathode follower.

Attachment:
Cathode follower.jpg
Cathode follower.jpg [ 55.15 KiB | Viewed 319 times ]


What I need is for this circuit to have the same output level as the first circuit and it has to have a 600 ohm impedance for driving the audio filters.

Will the circuit work as I have drawn it or do I need to change anything?

OR

Is it best to directly couple the cathode follower grid to the voltage amp's plate as seen in this circuit

Attachment:
vox-ac30-voltage-amp.jpg
vox-ac30-voltage-amp.jpg [ 30.16 KiB | Viewed 315 times ]




I'm using the local audio driver stage as the cathode follower. The local audio level control will go directly to the local audio out terminal on the receiver.

Was using the local audio driver to drive an external driver transformer for a push pull 45 amp, but I will add a driver stage to the amp.

The mod being it requires only some moving around of wires and components will be easy to put back stock if so desired by me or a future owner at a later time.

It sounds decent as is, but the bass under 100Hz isn't very good.

I want to get the audio to at least be flat to 60Hz which is where the driver transformer starts rolling off and if done right the audio will sound a lot like a good console radio of the day.

I do know it's easier to take the audio from the diode load, but I then lose the use of the audio filters.


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Mon 11, 2017 6:50 pm 
Moderator

Joined: Jan Thu 01, 1970 1:00 am
Posts: 35920
Location: Livermore, CA
A cathode follower doesn't give voltage gain (1:1) but supplies increased current. The R390 circuit shows impedance of 10K to 600 ohms. This would be a gain reduction but current increase. Eliminating an audio transformer would give better frequency response but you are going through filters anyway.

I am not one to make a lot of changes to Collins R390A design. It is/was one of the best receivers in the world.

_________________
Norm


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Mon 11, 2017 7:25 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 17506
Location: Warner Robins, GA
The R-390 has two filters and one selection on the audio response switch that is wide which means no filter at all.

Basically what I'm wanting is to make the output voltage of the cathode follower equal the output of the transformer for the same input voltage to the first audio stage.

I too normally don't like to make mods unless they are beneficial and can be easily put back stock without any indication a mod was done.

If not for the compactness of the R-390 I could get an Edcor 10K to 600 ohm transformer which has a frequency response to 20Hz and just use that in place of the audio transformer, but I know the Edcor transformer is way too large and will not fit at all.



EDIT:

Tried the mod and got some hum which I cured by moving the cathode follower B+ to the same feed as the first audio tube.

Got distortion which I didn't like so I decided to rewire it as original.

My next question is this.

With a B+ of 174Vdc what plate resistor and cathode resistor values would I need to properly drive an external 10K to 90K driver transformer for a pair of 45 tubes?

Originally used a 30K plate resistor and 270 ohm cathode resistor bypassed by a capacitor which seemed to work well, but I want to be sure I am getting enough distortion free voltage out of the driver stage to cause the 45 stage to go into distortion first which then means I'm getting the maximum clean output power the amp can deliver.

Concerning the low frequency response I either will not worry about it as it does sound decent as is or I will get an Edcor 10K to 600 ohm transformer and somehow shoehorn it in there in such a way that no permanent mods are done.


When I had the radio in my 12' X 12' building the bass seemed to sound a bit better, but then again the building itself is resonant at 40Hz so perhaps that helped the bass sound better.


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Mon 11, 2017 10:58 pm 
Member

Joined: Apr Thu 14, 2016 8:25 pm
Posts: 252
Location: pensacola fl
Hi all that distortion you have is probably caused by in part the wrong bias on the 12AU7. If you look in the rca tube manual in the back in resistance coupled amplifier charts there is one for that tube. At any of the voltages from 90 to 300 B+ the cathode resistor is always larger than 680 ohms. at the 180 volt section and with a 47k plate resistor the cathode resistor goes from 920 to 1400 ohms depending on voltage output and load presented by following stage. A cathode follower will present a very light load to the 12AU7 so with 47k plate resistor and 220k load and bypassed cathode resistor of 1.4k will give an output voltage of 29 volts peak and a voltage gain of 12. So input grid drive should not exceed 2.4 volts peak approx. The cathode follower will present a load of more than 220k depending on it's grid resistor(s). A light load and distortion should be low feeding the 12AU7 from 174 volts B+. I i would increase the cathode resistor just a bit from 1.4k to 1.5k. The reason is because the plate current will be just a little less and the bias will need a tiny boost so that should put you in to the ball park. The cathode follower can be done direct coupled or cap coupled depending on the tube used and how you bias it to meet it's requirements. What load impedance you are trying to drive and at what level.


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Mon 11, 2017 11:12 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 17506
Location: Warner Robins, GA
I was trying to drive a 600 ohm impedance to eliminate the 10K to 600 ohm transformer which was causing a loss in bass response.

I'm thinking the distortion came from the cathode follower. Perhaps it wasn't biased right. Did seem to improve when I increased the cathode follower cathode resistor to 56K.

That said I wasn't much into dicking around with resistor values as I'm getting over a cold so I just put the circuit back to the 12AU7 driving the transformer like it originally was.

All I now need is to know the correct values of cathode resistor and plate resistor for the 12AT7 section I am using to drive the external transformer with a 10K input impedance.

If I do decide to improve the bass response an Edcor WSM series 10K to 600 ohm transformer just will fit if I remove two audio transformers one being the unused local audio output transformer and the other being the transformer driving the filters. The mod will of course be reversible.


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Mon 11, 2017 11:44 pm 
Member

Joined: Apr Thu 14, 2016 8:25 pm
Posts: 252
Location: pensacola fl
Well you can also get if you wish an A24 transformer made by UTC. They are pro audio high fidelity and are made to drive 600 ohm from tube plate. Beware they are hard to find these days and HIGH priced. However you may find a cross reference from it to an edcor or hamond at a better price same basic specs.


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Mon 11, 2017 11:59 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 17506
Location: Warner Robins, GA
I would need a UTC A-25 in order to have DC current through the primary.

The problem is the primary is 15K not 10K.

Not sure how that will affect the overall signal levels, but I'm sure I could increase the gain of the local and line audio driver stages to compensate.

EDIT:

Back to the 12AT7 driver stage.

Here's the operating point for the 45 tubes. It is shown in red on the photo.

Attachment:
45 operating point.png
45 operating point.png [ 67.31 KiB | Viewed 247 times ]


Requires 140 peak volts. Am assuming that because it is grid to grid it would be 140Vpp which is 49.497Vrms.

Now the transformer I am using is a 1:3 so I would need a drive voltage of 46.667Vpp which is 16.499Vrms.

Will use my scope to set the output of the 12AT7 driver to the maximum it can produce before visible distortion and see what voltage I measure with my DMM. If it is greater than 16.5Vrms then I'm good with the values of resistors I have chosen. If it is less than 16.5Vrms then I need to alter the plate and cathode resistor values to set the proper operating point that will get me a little above 16.5Vrms to give some headroom.

I am also still debating on if I really want a better transformer for the one driving the filters or if I just want to leave well enough alone.

My concern would be the line audio output. Given it is only 1.4 watts , better bass response would not allow that output to go as loud without distortion. I only use it to drive a 32 ohm stereo set of headphones through a 560 ohm resistor with the left and right drivers in parallel plus I am not sure how better bass response would affect the intelligibility of SSB reception.


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Tue 12, 2017 9:29 pm 
Moderator
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 21928
Location: Somers, CT
A cathode follower provides power gain.

I was a bit confused why you used DC coupling plate to grid, how is the grid biasing achieved with a positive voltage applied on the grid?

Also, in the original circuit, the cathode of the amplifier is not bypassed. You could use that point to capacitor couple a low-impedance audio source from the radio, with little modification. You might need to use a higher gain input, say for a magnetic cartridge as an example, but I am not sure what levels you would see. But, it should work.

Pete

_________________
Just because it can be done doesn't mean it should be done.


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Wed 13, 2017 12:16 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 17506
Location: Warner Robins, GA
I'd have to see how much gain that stage has by measuring the input voltage and the output from the transformer secondary.

If the voltage is similar I might be able to increase the gain of the local and line audio driver tubes a little to compensate.

That said for a 12AU7 what cathode resistor would I need to get a 600 ohm impedance?

I would then have the issue of figuring out what cap to ground I would need to provide the same function as the cap across the transformer primary provides.

EDIT:

Well I tested the amp and the driver stage starts to distort just as the output stage starts to distort, but there is a drop in signal level when I connect the amp to the R-390 output so perhaps the driver stage needs something other than a 15K plate resistor and 270 ohm cathode resistor.


EDIT 2:

Did some measurements and here's what I got.

Overall audio response. The amp output was taken with the speaker connected so I'm sure it varies with frequency versus just using a resistor.

Attachment:
Response.jpg
Response.jpg [ 46.72 KiB | Viewed 229 times ]


Measured the stage driving the transformer driving the filters and here's what I got.

Attachment:
Audio stage.jpg
Audio stage.jpg [ 9.3 KiB | Viewed 229 times ]


What I don't understand is this.

There's a net loss in gain by using the transformer so is there some reason why they couldn't just use a cathode follower?

Seems like I should be able to make a cathode follower produce that output voltage for the same input voltage and have a 600 ohm impedance, but I don't deal much with cathode followers so am not sure. Also I would then need to figure out the right value of cap to cut the highs by the same amount as the cap from the plate to ground of the 12AU7 stage.



EDIT 3:

If y'all could hear this R-390 in person you would understand why I am just tempted to leave well enough alone.

Sounds so good as is. As I reduced the audio frequency from 200-40Hz I could hear some output from the speaker to at least 50Hz and it fell off to near nothing at 40Hz.

If I can find a UTC A-25 for a reasonable price provided it is the same size as the original audio transformer I may install it and just play around with the cathode resistor and bias to get the output signal back to what it should be given the A-25 has a 15K impedance.

That said the 15K impedance being a lighter load on the 12AU7 may not need anything at all done to the gain as it may increase some due to the higher load impedance.

If I do anything to the receiver I believe that using the A-25 is what I will do. I may still need to alter the value of the cap to ground at the primary of the transformer though to keep the same HF roloff.

That transformer having a LF response down to 40Hz is just right for my needs.

Probably gonna be hard to find one because I feel the going rate for the transformers is a bit much.


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Wed 13, 2017 3:09 pm 
Moderator
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 21928
Location: Somers, CT
The cathode of V601 is a 680 ohm resistor, and will give you an approximate 600 ohm source to drive whatever external amplifier you are using. The only question is whether the signal level is high enough.

You could try a .22 cap from the cathode to tap off the signal--that will work as a cathode follower since the
cathode is not bypassed. If it works, it would be a very simple, and easily reversed modification.

Pete

_________________
Just because it can be done doesn't mean it should be done.


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Wed 13, 2017 3:41 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 17506
Location: Warner Robins, GA
If I use a .22uF cap for coupling here's what I get.

Attachment:
Cathode follower data 1.jpg
Cathode follower data 1.jpg [ 113.32 KiB | Viewed 203 times ]


Now if I use a 10uF cap here's what I get.

Attachment:
Cathode follower data 2.jpg
Cathode follower data 2.jpg [ 109.49 KiB | Viewed 203 times ]


Looks like I can get by with the 10uF cap. I have a film cap that I can try if it is small enough.

Will first measure the output of the transformer then connect a 600 ohm resistor cap coupled to the cathode resistor and see if I get the same voltage.

Now if I take the signal off the cathode will the .0015uF cap from plate to ground will the .0015uF cap to ground have any effect on the audio at all? Also what value of capacitor would I need to cause the same HF roloff?

My next question is this. Do I need to bypass the plate to ground with a capacitor?

Reason I ask is as the frequency decreases below 200Hz wouldn't that affect the output on the cathode?

To test this mod is real easy. All I have to do is remove the link between both halves of the secondary then I can connect the 10uF cap and see if it works right.


Now Av=Vout/Vin
Av= 1.667 / 2.195
Av= .76

So when I try the cathode feeding the filters I need to get the same Av from the circuit.

The only reason I can think of for them using a transformer is a 10uF cap back then would have been larger and possibly not as reliable as a transformer.


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Wed 13, 2017 3:51 pm 
Moderator
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 21928
Location: Somers, CT
I don't think you need to bypass the plate. The audio response curve will depend on the load being driven; is this driving a grid or a transformer?

Pete

_________________
Just because it can be done doesn't mean it should be done.


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Wed 13, 2017 4:29 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 17506
Location: Warner Robins, GA
This will be driving one of two filters with a 600 ohm impedance. Will also directly drive the line and local audio level pots which has a resistor in parallel with them to equal 600 ohms.

Either way in all three positions of the audio response switch (sharp narrow wide) the load will be 600 ohms.

EDIT:

I did a test and at the cathode the audio is flat from 200-40Hz with 40Hz being the lowest I tested it at.

I also did another test comparing the output voltage from the cathode and transformer secondary from 1KHz-5KHz so I could have some idea as to the HF roloff, but I goofed and left the audio response switch in MED which affected the output from the transformer.

Will re-run the measurements when I get home today.

Figured that would tell me the amount of roloff the .0015uF cap provides, but I just thought that the transformer impedance is 10K so that cap possibly isn't going to do much.

I might play around with the cathode resistor and see if I can get the output up to where it needs to be.


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Thu 14, 2017 6:48 pm 
Member

Joined: Apr Thu 14, 2016 8:25 pm
Posts: 252
Location: pensacola fl
Ok but the A25 is a 30k ct push pull type and is with out a gap. The A24 is designed for single plate and has a gap so it won't saturate as easily and the difference of 15k to 10k is not much as you can load the secondary to reflect a lower load to the tube if I am correct as we did in broadcast consoles and mic preamps we terminated them into 510 ohm resistors when the true load was much higher than the transformer was designed to be loaded by such as an active balanced input device.


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Thu 14, 2017 7:21 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 17506
Location: Warner Robins, GA
Here's the data on the A-24 and A-25 transformers.

Attachment:
Transformer data.jpg
Transformer data.jpg [ 33.19 KiB | Viewed 158 times ]


I can tell you this. With the audio response switch in the MED position the audio starts to hit a brick wall around 4.95KHz and by 5KHz is barely audible.


EDIT:

I tried the cathode follower and after trying several resistors up to 22K I couldn't get over 100mV out of the stage with it connected to the rest of the audio circuitry minus the transformer that normally drives the filters.

Looks like I either leave it as is or find a transformer with better LF response that will fit.

Here's a thought.

If I cannot find a proper transformer with the correct frequency response maybe I can find a PC board of the same size as the can, install an OP-AMP on it wired as a buffer, use a 10K trimpot to set the audio level to match the output of the transformer and then rectify the heater voltage to supply the B+ for the OP-AMP. I would do this in such a way so that I can just disconnect the wires from the PC board, remove it, install the transformer and connect the wires back up to put it back original again. I doubt the current of the OP-AMP and power supply would cause any problems. I could connect the diode to pin 3 of V603 which will give an unloaded B+ of 16.968Vdc. May do that as a last resort if nothing else works.



EDIT 2:

What I would like to do is find a spare audio chassis and:

1. Strip it of all unnecessary parts.
2. Make voltage regulator totally solid state.

Doing the first two allows me room to do the third.

3. Redesign the audio section.
A. Line audio output remains the same only using an Edcor 10K to 600 ohm transformer for better LF response.
B. Local audio driver remains the same.
C. Driver for the audio filters stays the same only using an Edcor 10K to 600 ohm transformer for better LF response.
4. Return the current audio chassis to original spec.

Doing that I can return the receiver to original audio spec by removing the modded chassis and installing the original.

That may ultimately be what I do provided I can find another audio chassis.


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Sat 16, 2017 4:49 am 
Member

Joined: Apr Thu 14, 2016 8:25 pm
Posts: 252
Location: pensacola fl
Yup I pulled up my utc sheets and sure enough i read the wrong line for the higher impedance version it should be A26 at 30k primary. I also pulled up the schematic for the preamps and line amps using that as the output transformer and 12AU7 output tube and we capacitor coupled to the transformer. We did use the A24 because it had a better low frequency end down to 20 HZ. We also had a stereo phone preamp with A26 transformers and push pull output for noise and hum concerns not power level as the levels we ran did not exceed +8 dbm and we ran preamp outputs a 0 dbm to console inputs. You might try the A24 since you want that good low end. The A25 goes to 40 HZ. Both transformers have a +/- 2db response curve as does the A26. A26 and A24 have wider response at both ends but 20 KHZ is of course all we need at the top end. I think I got myself corrected here after reading the sheets and have added more useful information here as well.


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Sat 16, 2017 5:09 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 17506
Location: Warner Robins, GA
Going down to only 40Hz is plenty enough for a communications receiver.

Plus I won't have to figure out the capacitor coupling to the transformer and wonder if it will reduce the audio level any.

Really looking for a spare audio chassis so I can permanently mod it with solid state regulation and after removing some things not necessary I'll have room for two of the edcor 10K to 600 ohm transformers.

Now if I use the edcors they are good down to 20Hz so I'd need to change the value of the coupling cap to the line audio out power tube so that the bass rolls off at 40Hz. I only use that for headphones.


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Sat 16, 2017 10:10 pm 
Member

Joined: Apr Thu 14, 2016 8:25 pm
Posts: 252
Location: pensacola fl
That is not that critical except if it is too small will roll off low end. Here you go at 40 HZ 10uF.
397.88735772973837 ohms.

You can use electrolytic since the plate voltage is high dc or aa mylar job as well less leakage to transformer but larger.
Considering the impedance of the transformer is 15k the loss in the cap is small.


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Sat 16, 2017 10:29 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 17506
Location: Warner Robins, GA
Yes the 10uF would work for capacitor coupling. Not sure what plate resistor I would need.


EDIT:

Before I try a better transformer I need to take a look at this and see if it gets me what I need.

I put my scope on the 8uF filter cap feeding the filter driver transformer stage and here's what I get with the 8uF cap and 22uF cap.

Attachment:
Filter caps.jpg
Filter caps.jpg [ 26.05 KiB | Viewed 66 times ]


There is some improvement which may help the bass response some.

Now I took the measurements previously made on the output of the R-390 local audio driver and the current measurements and here's what I get.

Attachment:
Local audio driver output.jpg
Local audio driver output.jpg [ 16.69 KiB | Viewed 66 times ]


So it looks like the bass response should be a little better.

Some time tomorrow I'll take a look at the amp and see what its frequency response is.

Might do it at work so I can get more accurate results.

EDIT 2:

Just played some music through my AM transmitter and the bass response is a little better.

Long as I can get decent output at 40Hz more along the lines of a decent antique console radio I'll be good with that.

I'm almost there now and if the amp is reducing the output at 40Hz any then I will further test the amp to see if it is the driver transformer or the output transformer. If either I will get a replacement and try things again and if it sounds good enough I will consider this finished.



EDIT 3:

I started to test the amp and realized that I have no way of generating the 46Vrms the amp input requires to drive the amp to full output.

I'll start a topic about the amp.


Top
 Profile  
 
 Post subject: Re: Cathode follower question
PostPosted: Sep Mon 18, 2017 11:05 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 17506
Location: Warner Robins, GA
I think one problem is that I am so used to hearing the R-390 in my 12' X 12' building which is resonant at 40Hz which tends to boost the lower bass some.

Figured it was the room making the bass sound so good as my RCA K-80 seems a little light in the bass as well and it is in the room my R-390 is in.

That said the issue is the transformer itself as I can feed an audio signal to my part 15 AM transmitter and as I go lower in frequency somewhere around 60Hz the sinewave starts to look a little distorted and it is a little more distorted at 40Hz which I think means the transformer cannot handle that low of a frequency.


Top
 Profile  
 
Post New Topic Post Reply  [ 20 posts ] 

All times are UTC [ DST ]


Who is online

Users browsing this forum: Mikeinkcmo and 8 guests



Search for:
Jump to:  




















Privacy Policy :: Powered by phpBB