Magnavox AMP 142 rebuild

[Tube Radio]

I just purchased a Magnavox AMP 142 which was stripped of most components except tube sockets and the power transformer. The 12AX7 socket is gone as well.

I found the stock output transformer.

I plan on rebuilding it but I may make a change or two.

1. Moving the rectifiers to where the speaker output jack was as there's an identical hole next to it and I think they are octal socket sized. May use a 5U4 versus dual 5Y3 which will reduce the 5 volt current by an amp.

2. Replace the phase splitter with the type of phase splitter using one triode to produce both the normal and inverted signals (forget what its called) which will most likely be a 6SN7.

I then have an extra octal socket.

What I'm wanting to do is at minimim add a low pass filter to drive a 15" Jensen speaker to add some bass to an antique radio and also I possibly will add a second HI-FI input given the Jensen and it's three tweeters sound quite good.

How would I do the filter and second HI-FI input? The filter needs to be a low pass of 150Hz and at least two pole.

 

[Tube Radio]

That said would it be posible to add a single 6V6 to power the tweeters?

Then add a two way two pole filter with the crossover frequency being 6KHz along with a separate 150Hz low pass filter on a separate input.

 

[BinaryMike]

This schematic at 4tubes.com shows something similar to what you're looking for. It uses a line-level L/C crossover with inductor values not given, but your crossover frequency is probably different anyway. This doc gives design info for sharp R/C crossovers.

 

[Tube Radio]

I'll look at it when I get home.

The idea is to make the amp HI-FI capable when fed a HI-FI source and able to reproduce low frequencies from 40-150Hz when using the antique radio with it.

Will use separate inputs level controls and filters for both.

Not sure if the amp can handle a single 6V6 given I'll have a quad of 6V6 already running from the B+ supply.

EDIT:

That link you posted about filter design does help a great deal.

Now my next question is this.

Is it possible to do an active filter using a single tube in much the same way an active filter is done with an OP-AMP with the filter being in a feedback loop?

Also versus a filter in between stages given the amp uses feedback can the filter be done in the feedback loop?

Another option is to use a cathode follower as the first stage, use it to drive a car audio subwoofer crossover which then drives the amp.

I would then use a SPDT switch to select between the two inputs. The bass channel and the HI-FI channel.

The only issue I can see with a triode as a gain stage on the HI-FI input is Miller capacitance which will affect the upper treble unless the input impedance is low.

I do think that I will go with the phase splitter design I have as it is self balancing.

Now the input was done like that because I used it with a radio that the first audio tube also had the detector diodes in it.

I can use a 12AU7 or 6SN7.

 

[BinaryMike]

Is it possible to do an active filter using a single tube in much the same way an active filter is done with an OP-AMP with the filter being in a feedback loop?

Also versus a filter in between stages given the amp uses feedback can the filter be done in the feedback loop?

Tube cathode followers can be used very effectively in active filter circuits of the type with unity-gain op-amp buffers, i.e. unity-gain Sallen-Key type.

You could easily de-stabilize the amplifier if you start fiddling with its feedback loop or internal gain parameters. Don't do this unless you have audio test gear and a good understanding of feedback vs stability issues.

 

[Tube Radio]

Agreed about the feedback loop.

I'll just do the passive filter.

 

[gadget73]

The AMP-150 is pretty much what you're talking about. I believe the bass section is identical to the AMP-142, and it has a single ended 6v6 for highs, along with a passive crossover in front of the whole thing to make it work.

 

[Tube Radio]

Thanks I'll look it up.

Can the single 6V6 I used with the AMP-142 as in will the power transformer have enough B+ current capability and enough on the heater winding along with a 5U4 being enough of a rectifier?

Now concerning the input for the radio.
It is a Philco 118 cathedral.

I replaced the 100K plate resistor of the first audio tube with a 90.9K and 10K resistor with the 10K resistor connecting to the filter cap. That way I would have a 10K output impedance and a reduction in voltage level. The plate is fed to the driver tube via a .002uF cap to form a high pass.

I originally used a solid state amp with an OP-AMP two pole low pass filter, but due to how the amp connects when the radio would be turned on the amp would momentarily go to the rail voltage putting that across the speaker.

To fix that would have been more trouble than it was worth.

So I decided to go tube.

Here's how the amp will connect to the radio

 

[gadget73]

The 150 uses twin 5U4GB. Not sure what it's power transformer is, but if its the same as the 142, you're good. Even so, a single 5U4 wants 3 amps of current, two 5Y3 want 4 amp, so swapping rectifiers frees up some current capability. I expect there is an extra heater winding in there to power the tuner that isn't in place anymore so you can run the extra amp circuit from that if you want.

 

[Tube Radio]

For now I think that I will stick with the stock amp as to BI-AMP will require a bit more work since I'll have two different low pass filters that will not be used at the same time.

If I do decide to BI-AMP at a later date I'll just get an AMP-150 and use that.

I did find a passive low pass two pole filter calculator online which makes it easy to calculate the components needed for the filter.

I referenced an old schematic I had of something I never built which had a two pole passive low pass filter so I plugged those values into the calculator so I could see the crossover frequency.

It is something like 150Hz.

I can post an exact value along with a schematic when I get home as at work I cannot log into the forums.

 

[Tube Radio]

Here's the schematic.







Now does that schematic look ok or are there things I can do to improve it?

I am not sure how much gain the voltage stages have so the plate resistors are subject to change in value.

 

[Tube Radio]

Now if I wanted to BI-AMP here's the circuit I would need.



The crossover frequency was chosen as it is close to the stock crossover for the Jensen speakers of 6.284KHz.

If I do BI-AMP at a later date the treble amp will be a totally separate amplifier. Thinking it may be push pull perhaps a couple 6V6 or 6BQ5 tubes.

Given the 470K plate resistor of the phase splitter driver (circuit uses a 47K resistor, but stuck with the Magnavox's 470K given the 220pF cap to ground afterwards) there may be enough gain to where the one 6SN7 voltage amp is not needed and if so I can use that as the second cathode follower.

Now I'm pretty sure given the different phase splitter circuit I will have to adjust the 220 ohm resistor.

Some caps are larger in value than what one may typically use. That's to keep the phase shift at a minimum.

I used this calculator for the coupling caps and cathode bypass caps.

http://sim.okawa-denshi.jp/en/CRhikeisan.htm

I used this calculator for the filters.

http://sim.okawa-denshi.jp/en/CRCRkeisan.htm

 

[gadget73]

On the thought of bi-amping, I know GordonW has converted a few AMP-150's to stereo 6L6 amps and removed the crossover in the process. Possible he's got one of those crossovers laying around. The Magnavox crossover is somewhere around 1100hz, not sure if thats of any use to you or not.

 

[Tube Radio]

No real use to me as the crossover frequency is too low unless the chokes can be varied enough to adjust the crossover that high.

I would like to stick to the original crossover frequency as when I tested the speaker on a 14 watt solid-state amp it sounded real good and normally I'm no fan of cone tweeters.

 

[BinaryMike]

Now if I wanted to BI-AMP here's the circuit I would need.

I believe that circuit yields about 3dB suckout at crossover frequency. If you're cascading two RC sections for each filter, you want each one to be down 1.5dB at crossover frequency, not 3dB. See the Harrison article.

 

[Tube Radio]

From what I understand what I made is basically a 12 dB/octave high pass and low pass filter. At least that's what I saw when reading up on passive RC filters.

I got the idea for that filter circuit from a Motorola three channel console which is two stereo channels and one bass channel.


Here's the data from the calculator I used to figure the component values.

 

[BinaryMike]

From what I understand what I made is basically a 12 dB/octave high pass and low pass filter.

That's correct, but you end up listening to the sum of the outputs of these two filters, which will be down 3dB at crossover frequency (ignoring phase). Harrison applies a correction for this error. You don't have to understand all of his math --- just use the simple formulas given in the figures.

 

[Tube Radio]

Now if I use the Linkwitz Riley arrangement the combined output from the woofer and tweeter will remain flat at the crossover frequency.



https://www.rane.com/note160.html


Now if the 95K resistor is made 950K and the 270pF cap is made 27pF I get
Q = 0.49751243781095
Damping ratio ζ
ζ = 1.005

 

[Tube Radio]

As much as I want the whole signal path to be all tube I think that will be more trouble than it's worth.

So my thought is to use an active solid state crossover for the HI-FI input.

That would be simplest as I can feed a source to the input of the crossover then feed the output to the bass amp and treble amp.

The built in level controls on the crossover can be used to adjust the levels of the amps, although each amp will have a level control.

A volume control will be used before the crossover or maybe I will try a passive preamp.

For the radio input I will have a cathode follower and two pole low pass filter.

A switch will be used to select whether the bass amp gets its input from the crossover or low pass filter.

The HI-FI and radio inputs will have separate level controls.

Now I do know Miller capacitance may be an issue with the first amplification stage of the treble amp unless the source impedance is low and the level control is of lower resistance.

In addition to making it easier to build it also opens up the possibility of swapping crossovers at a later date to something better as I get the money to do so perhaps eventually getting one of those vacuum tube crossovers.

I currently have a DBX 234XS crossover that is not being used so maybe that will work for the time being.

https://dbxpro.com/en/products/234xs

Is this a good idea?

Also for the treble amp will a single 6V6 be enough to drive three 5" tweeters in series loud enough to match the output of the 15" Jensen running on a max of 20 watts?

Now input sensitivity of the bass and treble amps isn't super critical as the pro audio crossover I have can put out pro audio levels and if necessary I can build an active preamp to go before the crossover that will output a higher level.

Given where the speaker is located the source will most likely be a Sylvania RM-300X radio from its speaker outout when said radio is not being used to drive a full range plate amp which drives a 150 watt speaker. So I should have as much drive voltage as I could ever need.

I know I'm jumping around with ideas, but this is all sort of new to me far as building passive crossovers when said crossover does not feed a speaker directly is concerned and I want something that I know will work without problems.

 

[CopperWizard]

Is this a 142? I have a 142bc.