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


It is currently Dec Sat 16, 2017 4:33 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 8 posts ] 
Author Message
 Post subject: Push pull 45 amp
PostPosted: Sep Mon 18, 2017 3:12 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 17981
Location: Warner Robins, GA
While doing audio frequency response tests on my R-390 to improve the bass response some I decided to test the push pull 45 amp I use with the receiver.

Found an issue with it so decided to start a topic about just the amp.

Attachment:
45 amp.jpg
45 amp.jpg [ 72.25 KiB | Viewed 817 times ]


I originally used a 1K resistor before deciding to go with a field coil speaker and decided to just leave the 1K chassis mount resistor in place for testing purposes as all I'd have to do is connect it to the banana jacks the field coil connects to.

Here's the datasheet for the 45 tube. The operating point is in red.

Attachment:
45 data.png
45 data.png [ 25.39 KiB | Viewed 817 times ]


Started testing the amp and realized I have no way to get the 46Vrms required to make the amp go to full output.

I then realized that I still had the solid state amp I built to drive high impedance 1920's speakers at work and it puts out 50Vrms with a 5K load undistorted.

Perfect I thought, although it won't go flat to 40Hz, but will be ok for testing the max amp output at 400Hz.

Connected it up and realized the output of the solid state amp was a bit distorted at way below 50Vrms.

I then tried the center tap of the transformer and one end terminal effecting halving both the voltage and output impedance of the amp. I got enough drive then.

That tells me the load on the Thordarson T-33A91 secondary is less than 90K which means the primary impedance is actually less than 5K. Don't know how that will affect the frequency response though.

The 45 amp originally came from an RCA Radiola 82 console that I had no cabinet to. The Thordarson was a replacement for the original interstage transformer. Seemed to work fine there and seems to work fine in the 45 amp which I rebuilt on a much smaller chassis.

Here's the measurements

Attachment:
Power measurement.jpg
Power measurement.jpg [ 13.75 KiB | Viewed 817 times ]


According to the data posted earlier I should get 7.4 watts output so that tells me the output transformer I am using (old power transformer HV winding to plates 120Vac winding to load) isn't 5K.

That leads me to another problem.

When I tested the amp connected to the R-390 driven by the local audio driver tube which is cap coupled to the local audio out terminal and had been modified to drive the transformer both the output of the receiver and amp started to distort at the same time.

Now when I get the proper output transformer for the amp I am hoping that the full 46Vrms required to make the amp reach full output will be what I really need which means I'm then getting full output power.

The driver stage will not be able to put out that much without distortion.

Thought I had a schematic of the driver stage here at work, but I do not so will post it when I get home.

The tube used is 1/2 of a 12AT7.

Now I can either install a triode on the amp with the transformer primary in the plate circuit and connect the local audio level control to the local audio out terminal or I can figure a way to make the local audio out driver produce more output voltage.

The easiest way would be to use a 27 tube, but that would require enlarging a hole meant for a 9 pin miniature tube for the larger tube socket.

That said can I get 46Vrms out of a 12AT7 stage at a B+ voltage of 174Vdc (actual plate voltage is 150 something I think) driving a load of a little less than 5K or will I have to use an external tube for that?

The turns ratio of the power transformer I am using as an output transformer is 3.878:1 which gives an impedance ratio of 15.04:1. With a secondary load of 600 ohms the primary is 9.023K. Doing calculations the output impedance of the amp is 333 ohms.

So I will need to get a proper transformer before doing anything else with the amp.

The 5K transformer I originally used with the amp just so happened to be the right transformer for a Magnavox CR-198C which had an intermittent output transformer short and at the time I didn't have the money to buy one so I made something work.

EDIT:

I took some measurements of the amp and here's what I got.

Attachment:
Measurement.jpg
Measurement.jpg [ 50.82 KiB | Viewed 792 times ]


The output of the signal generator does drop some and even under no load it drops some as well, but not quite as much so that will skew the results a bit.

It is clear that the output does drop a bit from 60-40Hz compared to the rest of the test frequency range. That drop should be ok provided the output of the R-390 is made fairly flat to 40Hz.


Top
 Profile  
 
 Post subject: Re: Push pull 45 amp
PostPosted: Sep Wed 20, 2017 5:43 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 17981
Location: Warner Robins, GA
Here's the updated amplifier with a 6S4A driver added.

Attachment:
45 amp driver.jpg
45 amp driver.jpg [ 89.29 KiB | Viewed 759 times ]


I do have a little hum not related to the driver or excessive ripple on the B+ line.

I think this is the problem.

Attachment:
20170920_002057.jpg
20170920_002057.jpg [ 230.52 KiB | Viewed 759 times ]


I seem to remember when first building the amp I had to locate the transformer farther away from the power transformers which was where the 6S4A tube is now and have it oriented the opposite direction from the power transformers.

Think I will mount the transformer on the side of the chassis and see if the hum goes away.

Tried driving the amp directly from the local volume control of the 390A, but didn't get enough gain so I had to use the 12AT7 driver stage. I lowered the plate resistor to 16K and removed the cathode bypass cap then had to put a 10K resistor to ground after the 4.7uF cap as there was excessive hum if I did not plus the output was too high.

With the power transformer I am using for the output transformer the amp is still good for only 4.2 watts of undistorted audio. Now the driver stage seems to want to distort just after the output starts to distort so when I get the proper 5K output transformer I will see if the driver stage then distorts first.

If it does then I will need a better driver tube perhaps the #27 will work or if I still have one the power triode section of a 6DE7 will work as it might be able to put out more voltage than the 6S4A can.

One thing I have noticed is that it seems like tubes with a lower amplification factor seem to be able to put out more AC voltage.

Now in order to use the 6DE7 I will need to drop the B+ to 150Vdc which means a large power resistor.

I may try the #27 tube as I know that tube will do it and it won't be too hard to make a hole for the tube socket. If the gain is too low I can just add the cathode bypass cap back to the 12AT7 stage and/or increase the plate resistor back to 30K.


Top
 Profile  
 
 Post subject: Re: Push pull 45 amp
PostPosted: Sep Thu 21, 2017 1:29 am 
Member
User avatar

Joined: Jun Wed 08, 2011 2:33 am
Posts: 6088
Location: Ohio 45177
A 27 might do it. I have a console with a 27 to 45 P-P. And they also run the electromagnetic phono pickup (RCA type) to the 27 and that works out. Not too loud, but that might be due to other factors. Of course radio section is louder than phono.


Top
 Profile  
 
 Post subject: Re: Push pull 45 amp
PostPosted: Sep Thu 21, 2017 3:26 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 17981
Location: Warner Robins, GA
I had looked at the data wrong and did wrong calculations.

Shows 140 volts peak grid to grid which is actually 280Vpp.

That divides by the transformer ratio of 3 gives me 93.333Vpp which is 32.993Vrms.

So the driver needs to be capable of at least 34Vrms.

The 6S4A is close at 26Vrms max before distortion but doesn't quite make it there.

The transformer primary is in the plate circuit feeding B+ to the tube which i think is 350Vdc best I can remember.

Cathode resistor is 10K. Any lowering of the resistor or adding a cathode bypass cap just increases gain without increasing the output voltage any.



EDIT:

Did some tests today at work.

I get 48.314 Vrms across a 598 ohm resistance for 3.9 watts.

The maximum grid drive to the 45 grids I can get before the driver starts to distort is 82.8 Vrms which is 234.192 Vpp.

Now I did notice a waveform across the 1K resistor between the 45 transformer CT and ground.

Attachment:
Waveform.jpg
Waveform.jpg [ 10.04 KiB | Viewed 706 times ]


If I remove the load on the output the waveform becomes a proper sinewave.

What exactly does that indicate?

Will be testing the 45 tubes this afternoon when I get home.

I also noticed a problem with the driver transformer.

With the primary in the plate circuit I noticed that as I go down to 40Hz the output drops to 4.06 Vrms and at 2KHz the output is 4.04 Vrms.

I need to cap couple the transformer I think to eliminate that issue.

EDIT 2:

Cap coupling the transformer did the trick.

Used a 1K cathode resistor and 20K plate resistor and get 35 Vrms with the transformer connected.

Gain of the driver is down slightly as it once took 9.8Vpp for full output of the driver, but now it takes 20Vpp.

Not a problem as i can connect the cathode bypass cap in the R-390 local audio stage and/or reduce the value of the 12AT7 plate resistor.

That's as far as I got before time to leave work.

The good news is that I can get 11 Vrms out of the R-390 12AT7 driver stage before distortion so it may have no problem driving the 6S4A stage as is.

Here's the latest schematic

Attachment:
45 amp driver 2.jpg
45 amp driver 2.jpg [ 92.43 KiB | Viewed 700 times ]


I used an 18uF 400V film cap to couple the driver to the transformer simply because I had it handy. Could have easily used a 4.7uF cap, but didn't have any of those.


Top
 Profile  
 
 Post subject: Re: Push pull 45 amp
PostPosted: Sep Fri 22, 2017 7:02 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 17981
Location: Warner Robins, GA
Looks like I got the amp as good as I am gonna get it.

Here's what should be the final schematic at least until I get the proper output transformer.

Attachment:
45 amp final.jpg
45 amp final.jpg [ 98.57 KiB | Viewed 658 times ]


Here's the various measurements.

Attachment:
Amp measurement.jpg
Amp measurement.jpg [ 24.28 KiB | Viewed 684 times ]


I did add a switch to the amp so that I could select the internal 10K resistor for use with a PM speaker or for testing the amp away from the radio or use a field coil speaker.

I do know the output transformer is one issue and I expected the driver transformer to roll off the bass some.

Will need to order a 5K to 4 ohm transformer and the only thing then reducing the bass some will be the driver transformer, but it should be ok and if it isn't I can have Edcor design one.

EDIT:

Can't remember exactly but when doing initial tests before adding the driver transformer I deduced the 45 tube grids were presenting a less than 90K impedance which makes the primary impedance of the driver transformer less than 10K.

Think I had figured the primary impedance was actually around 5K based on how it loaded the amp I was using the drive the transformer which can do 50 Vrms into a 5K load. Got about the same output voltage from the amp.

Also I noticed that one of the two filament sections in one 45 is glowing a good bit dimmer than the other section. I'm thinking perhaps I have a failing tube and maybe that's one reason why the amp cannot put out the full 7 watts it is capable of.


Top
 Profile  
 
 Post subject: Re: Push pull 45 amp
PostPosted: Sep Mon 25, 2017 4:46 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 17981
Location: Warner Robins, GA
Ok further testing of the amp reveals that it is now basically hum free.

I get 37mVrms on the 600 ohm output with the input jack shorted to ground.

I get 4.1mVrms across the speaker which is connected to the 4 ohm tap of the 600 to 4/8 ohm transformer.

With the input connected to the receiver I get 111.4mVrms and some hum in the speaker.

Tells me the rest of the hum is from the receiver somewhere.

I'm guessing the hum might be being picked up by the local audio out wiring from the audio chassis to the local audio out terminal.

I'll use my scope in the audio chassis tomorrow and see if the hum is B+ related or just being picked up by the wiring.

The hum is barely noticeable though unless in a completely silent room.

I think the issue is that the local audio output was meant to be low impedance which would have been 600 ohms and meant to drive a speaker so the hum would have been much lower or not present at all.

I currently have a 10K resistor connected between the local audio output and ground as that keeps the hum down to a reasonable level plus it reduces gain some.

Now if I could find a dual triode with a triode section the same as or very close to the 6S4A I could use the other triode as a voltage amp then the local audio level control could connect directly to the local audio out jack.

I would then put a pot between the first triode and the other triode so that I could have some control of the gain.

If I cannot reduce the hum from the receiver I am thinking the 6DE7 will be a good choice for the dual triode.

The power triode has an amplification factor of 6.0 which should allow it to put out a higher audio voltage.

The other triode section has an amplification factor of 17.5 which I don't know if it will be enough to fully drive the other section of the 6DE7 from the local audio level control given the 12AT7 has an amplification factor of 60.

That said with the 12AT7 I am using a 16K plate resistor an unbypassed 270 ohm cathode resistor and a 10K resistor to ground after the 4.7uF coupling cap which feeds the signal from the plate to the local audio output terminal.

Here's what I get.

Attachment:
Gain.jpg
Gain.jpg [ 7.93 KiB | Viewed 631 times ]


Looks like the 17.5 amplification factor of one section of the 6DE7 will work just fine for stepping up the voltage enough.

Here's what I get from the amp input to plate of the 6S4A

Attachment:
Gain 2.jpg
Gain 2.jpg [ 8.52 KiB | Viewed 631 times ]


Here's the total gain from the local audio level control to 6S4A plate.

Attachment:
Gain 3.jpg
Gain 3.jpg [ 9.06 KiB | Viewed 631 times ]



Now here's a thought.

Given the local audio 6AK6 was removed and its heater voltage dropped by a resistor due to the filament being at a little above 7Vac dropping the filament voltage to the other tubes I could use a 6C4 as a cathode follower which would allow for a low impedance output while allowing the 12AT7 stage to function normally.

That said I do feel that I want to go the route of using a 6DE7 though if it will eliminate the hum.

With the local audio out connected directly to the local audio level control I get 31.8mVrms of hum.

Now I don't know the actual amplification factor of both 6DE6 sections so I don't know how much the hum will be amplified.

I do know that when calculating the voltage gain based on the 6DE7 tube section amplification factors it would require 261.4mVrms at the amplifier input which is 10 more milivolts than the current configuration.

That might present a problem especially if I ever get the amp to put put the full 7 watts at which point I will need more input voltage.

The test I did was at full amp output right before the onset of distortion.



EDIT:

Didn't find a 6DE6 but I did find a 6BM8.

With the pentode wired as a triode perhaps that will work to get me enough output.

The triode has an amplification factor of 70 which I can always reduce by changing the value of cathode resistor and plate resistor.

I will install a 100K pot on the input so I can adjust the gain.


Top
 Profile  
 
 Post subject: Re: Push pull 45 amp
PostPosted: Sep Wed 27, 2017 3:01 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 17981
Location: Warner Robins, GA
Well the 6BM8 works great.

Here's the final schematic.

Attachment:
45 amp final 2.jpg
45 amp final 2.jpg [ 103.55 KiB | Viewed 599 times ]


I adjusted the pot to where the amp just does start to go into distortion when the local audio level control on the R-390 is set to 7.

Also I figured out a longstanding issue I had of a slight bit of audio being present in the speaker with the local level control set to 0. I shorted the center terminal to the grounded terminal of the volume control and the audio went away completely.

So I need to replace the volume control.




Now looking back at the measurements I took today when testing the amp, the triode connected pentode of the 6BM8 can produce a good bit over 35Vrms perhaps around 40Vrms if not a little more undistorted.

That said if so desired I could add a second 6BM8 and use its triode as a phase splitter to drive both triode connected pentodes and not even need the driver transformer. I would only do that if the driver transformer ever fails and I could not find a reasonably priced replacement.


Top
 Profile  
 
 Post subject: Re: Push pull 45 amp
PostPosted: Oct Wed 18, 2017 11:57 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 17981
Location: Warner Robins, GA
Been using the receiver and it sounds great with the 45 amp.

Arguably the best sounding receiver I've had and I suspect the 45 amp has something to do with it.

The only problem is that 45 tubes aren't exactly plentiful so it may be a use until the volume or sound becomes unsuitable then find more 45 tubes situation.


Top
 Profile  
 
Post New Topic Post Reply  [ 8 posts ] 

All times are UTC [ DST ]


Who is online

Users browsing this forum: fvan, wa7opy and 10 guests



Search for:
Jump to:  


















Privacy Policy :: Powered by phpBB