[RodB]

Mike, you can see and hear the careful work you put into that radio. Great job.

[Ron Ramirez]

Took me a couple of months, yeah.....but I was working very, very slow.

In the last few years, it takes me longer to fix a radio than it used to. But I'm also working slower, taking my time on each one. As Gary said, it isn't a race.

Looks good. I can go to the Facebook link you posted, but since I'm not a Facebook member, I see three photos of your AK but no video.

[Eliot Ness]

Looks good Mike! The AK 80 series are one of my favorite line of radios with the 82 at the top of my list.

[GarySP]

Mike, I have never rushed working on radios. It's a hobby ... something to be done for recreational purpose, and in my case, education on the fly! I may spend hours on a project one week, then walk away for a month. If you throw a timeline on it, it becomes work. My intent is to one day understand the "why it works" theories as well as most of you all do. To that end I will always pester everyone with loads of questions! Take care and BE HEALTHY! Gary

[morzh]

Ron

It is under one of the replies to that post with the three photos. As you open them, the vide will be there.

[MrFixr55]

Hi Mike,

I know that you don't wanna disassemble your 82 again, but my thoughts:

Most of these early radios have rather small (8-10 uF filter caps), because I don't think that they made much bigger. By the 1940s, Filter Caps increased to 20 - 30 uF or higher., and higher still when they went to permanent magnet speakers

The RCA Manual states that the maximum input filter (between cathode and B-) for an '80 or 5Y3 is 40 uF. You state that you bridged C2 with a 20 uF Cap to no avail. That would increase the current through the field coil and may add to the hum. Try bridging the input filter cap "Filter Condenser 1" on the Rider's Diagram).

The other contributor is the '47 Output tube being directly heated (The fiament is the cathode). Try subbing (by making an adaptor) a 2A5 which was the successor to the '47. It has the same filament voltage, plate load impedance and grid bias as the 47 but has the "separate heater and cathode" design instead of the direct heated cathode. The only addition is the cathode terminal which should be grounded. If this greatly reduces hum, this will prove that the output tube is the major contributor.

The hum level is probably a lot lower than the AK36, 40, RCA Radiolas, etc. which used 71As and worse yet, 26s. 26s were the first "AC" Tubes, used in in the Radiola 17, 18, AK36, 40, etc., and were a compromise by using a heavy filament and 1.5V. They still hummed. The Philco 20 had all tubes as being true AC Tubes (24A, 27), except the 71As but these were push-pull, and the pie plate speaker had very poor frequency response due to the very hard suspension.

I think that the AK 82 had a much better suspension, similar to the Rolas of the time, and agree that very likely did not have a hum bucking coil.

I have not worked on an 82 but did restore a 165, which except for the choke / cap IF coupling to the detector, was a thoroughly modern transformer powered 2.5V version of the AA5. The field was in the B+ side of the power supply, instead of the B- leg like the 82 did, and does have the hum bucking coil. If I remember, it had a Rola speaker. It performed very well with negligible hum. However, I used a 33 uF Caps for the input filter (C13).

BTW, a lovely job on a true icon of the "cathedral" style radio!

[Ron Ramirez]

Mike

Regarding the video on Facebook - By golly, you're right. But no wonder it sounds bad, with that crappy "music" playing in the first part of the video.

Seriously, though, if you're satisfied with how it turned out, keeping its limitations in mind, then that's all that matters.

[morzh]

MrFixr55


Valid suggestions.

Let's see.

1. I did not increase the output cap C2. I tried increasing it, with absolutely no effect, so I took it off, and it is 10uF again.
Speaking of icreasing the current - no, it would not. The load after LC filter draws current which is constant.
Also, in general, the ripple current from the inductance with a constant value of the output Volatge really depends on the inductance, not capacitance. And the inductance is a current source, so there will not be "inrush" type charge, as the one that occurs with the first cap where the current is only limited by the plate resistance of the 80 rectifier. However the output ripple voltage will decrease with increasing the C2 cap.
Also the ripple voltage will decreas with using a low impeadance cap, as the output ripple voltage has a Ir*ESR component (in DCDC concverters this is the only significant component).

Now as for increasing the input (C1) cap - yes it could be safely (for the 80 rectifier - up to 32uF) raised to that value. However the value of the B+ then will aso increase, as the input ripple before the inductance will be much less. So your tubes will run hotter. The difference could be quite significant. COnsidering the radio is already plugged into 120V AC, it is something to consider.


2. Yes, many folks agree the direct heated tubes with large heater current are spurces of extra hum. The technique goes from replacing them with different tubes to using DC for heating.
Thing is, I try to be as original as possible, so if a capacitance increase was a consideration for the C2 (if it worked, which it dod not), a schematic change for me is not an option.

3. One way they used to compabt extra ripple (which is not necessarily what causing it all in my case) - putting 0.15uF or so cap across the choke (field coil) to compensate the ripple through the inductance. But many think it does not work well, and even remove igt altogether from places where it i used. Like Philco 20 and such.

4. I have a Philco 70. It uses a very similar tube line-up. 24As and 47 at the Audio. (I specifically do not mention the 90 I have as it is a P-P output one, and the 20 as it is also P-P, both are direct heated tubes, 45 and 71A). It does not have as strong a hum. However, it has a different scheme for powering the tubes.
They use one winding to power the 4x 24A, dial light and the 27, and another separate winding to power th 47 tube. And it is a separate centertapped winding with the tap Grounded.


So to summarize, were it some very common tube radio that I wanted to make a musical exibition piece, I might rework the schematic, bring it to a near hifi level (would probably also require changing the speaker) and then marvel at its acoustical performancel however the AK82 is rare and beautiful enough radio that to me is worth keeping as original as possible.
So are my Philcos 20 (have three of them) and the 90 (the more rare version with 2x45 output).