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


It is currently Jun Tue 27, 2017 2:57 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 68 posts ]  Go to page 1, 2, 3, 4  Next
Author Message
 Post subject: Ballast tube discussion
PostPosted: May Tue 05, 2009 4:55 am 
Member

Joined: Feb Thu 26, 2009 6:00 pm
Posts: 150
Since this discussion was off topic I thought I should create a seperate thread for it.

Here is an excerpt from Coyne, Applied Practical Radio-Television concerning ballast tubes.

https://www.msu.edu/~yurkon/Documents/ballasts.pdf

It explains how they regulate, and points out that not all are intended to provide regulation.

Earlier I said that a high TCR (Temperature Coefficient of Resistance) was important for regulation. That's not quite true. It's more important to have a critical temperature. The resistance vs temperature for iron can be approximated by two straight lines that intersect at the critical temperature. For iron this is 430C+/- 30C.

Maybe only TahoeTV and myself care. :wink:

John


Top
 Profile  
 
 Post subject:
PostPosted: May Tue 05, 2009 6:13 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 5054
Location: Woodinville, WA USA
Dumb question: the article says that iron provides regulation while other resistance elements may not. I made a ballast replacement using modern power resistors for my Motorola VT-73 and it seemed to work fine.

Is current regulation not such a big deal nowadays as in the past, when line current may have varied more? My fanciest boatanchor radio has a current regulator ballast as well as a voltage regulator tube, but I don't recall seeing anything like that in the few old TVs that I've worked on. This radio also works fine with a 6V6 tube in place of the original unobtanium ballast (i.e., no current regulation).

Phil Nelson

Image


Top
 Profile  
 
 Post subject:
PostPosted: May Wed 06, 2009 4:40 am 
Member

Joined: Feb Thu 26, 2009 6:00 pm
Posts: 150
Thanks for the photo. I wasn't sure if the VT-73 used the same ballast as the 9T1. It's very clearly labeled on your chassis.

The 17A485459 is the ballast that I did the I/V measurements on. Over the range of currents that I made the measurements for, it acted no differently than a purely resistive ballast. I measured the gauge of the wire, from a open air metal ballast replacement, and the resistance per unit length and it matched that of standard nichrome. I couldn't find any alloy with a closer match. I also measured the I/V for a length of it and it was quite linear up to the point at which it burned up.

I think your resistor ballast is a perfectly good solution. I must admit I started researching this because I originally thought the Motorola sets needed something more than dropping resistors. I was wrong.

I don't have direct experience, but I've read posts of others doing similar tests on the ballasts used in various boat anchor radios and they do provide regulation as you state.

It's difficult to know what was in the minds of engineers at the time. My 9T1 has no transformer and I think that is the sole reason they used a ballast tube, to adjust the voltages.

My Sylvania 1-108 chassis was made just a year later. It has a transformer, no active regulation, and no ballast tubes. I think the AGC circuits were adequate to handle the line voltage variations present during those years.

Some of the oldest radios without AGC control the gain by adjusting the filament current. Those would be incredibly sensitive to line voltage variations. The brochures for my Majestic 71 boast that is uses a pot for the volume control rather than adjusting the filament current. It's still pretty sensitive to the line variations in my house since it's lacking AGC.

John


Top
 Profile  
 
 Post subject:
PostPosted: May Wed 06, 2009 6:00 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 644
Location: Crystal Bay, NV
Thanks, John, for finding that interesting write up in Coyne. I have their set of books, but I didn't look there. That critical temperature wasn't described in the other stuff I looked at, but it makes a lot of sense.

Your measurements of resistance vs temperature didn't show the critical part of the curve I would expect. What was the temperature (or color) of the wire when it burned up?

I still think that Motorola intended the ballast to perform a voltage regulation duty. Line voltages around the US varied from 105 to 120v, depending on local and loads on the power grid. On the other hand, transformer sets did not seem to worry about voltage regulation. Motorola never used "left over" technology. I can't think of any other manufacturer which used the ballast tubes, and I don't think Motorola would have any difficulty eliminating them if they wanted to.

Phil --- that resistor network looks very well made, but rather incongruent with the rest of the chassis. I have been thinking that your resistor solution might actually prolong the life of the tubes by reducing the turn-on current surge. When I checked the cold resistance of a 6AU6 I found that it was less than a third of the operational resistance. Thus, a resistor acts a current limiter (or reducer) as the tubes warm up.

Just maybe, Motrorola thought about this too. The schematic of the TS-4J show a ballast with five sections (part #17A470303). Two the the sections are used in the twin 300ma fillament strings. Subtracting 6v for the 7JP4 leaves 111 volts for each string (117v line assumed). The voltages for the tubes in each string add up to 73 volts. That leaves 44 volts for the ballast. 44 volts at 300ma comes to 145 ohms. Yet the cold resistance of the ballast is only 105 ohms. (Did you use 145 ohms, Phil?)

If the ballast operates near the critical temperature of iron wire (430 C), that is a rise of 400C from room temperature. Let's assume that the regulation effect should begin at 110 volts. The desired drop is 37 volts and the desired resistance would then be 37/.3 or 123 ohms. The temperature coefficient of the wire should be (123-105)/(105 *400) = .00043. Iron is way too high (.006), but nichrome is too low (.00017). Of course, these calculations were based on the critical temperature being 430 degrees celcius. (based on Iron).

The three other sections of the ballast might be made of a different wire, but that doesn't seem likely. The section connected between pins 6 and 8 is not used in the TS-4J chassis, but it is used in some earlier versions with a different tube lineup. It has a cold resistance of 22 ohms.

If the ballast works the way I hope it does, it provides protection against the initial current surge (when the tube resitances are very low), and it regulates against high line voltages.

John, you did not say what the resistance coefficient was, just that it was linear. Someday, not soon, I will put a TS-4J on the bench and try to measure the ballast temperature with an infrared gun. Also, I'll look at the voltage drop at turn-on vs. warmed up.

As far as I'm concerned the ballast becomes more interesting all the time. I am thinking of holding a national convention for ballast enthusiasts, and I'm calling for papers on topics such as: The history of the ballast and its effect on global warming; the ballast and the decline of Motorola; recent advances in ballast technology, etc. I'm open to suggestions.

====
Ron


Top
 Profile  
 
 Post subject:
PostPosted: May Wed 06, 2009 6:44 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 404
Location: Greenville, SC USA
One thing to think about is that in 1949 there were still a few places in the US that supplied 25 Hz AC or even DC from Wall Outlets! (Mostly in New York City, but there were a few other locations as well.)

I don't know if this adds another "wrinkle" to the discussion (I wouldn't really *think* it would), but you never know.

Any thoughts?

Bill


Top
 Profile  
 
 Post subject:
PostPosted: May Wed 06, 2009 7:06 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 644
Location: Crystal Bay, NV
The Motorola would not work on DC. That's because it has a voltage doubler circuit for the B+ supply. If it weren't for the doubler circuit, it could work on DC like all the little 5 tube radios.

The 25 Hz line current, I didn't know about. That's the standard for Europe and at least was for some parts of Canada. This set would work on 25 Hz, although there may be some audio hum or minor picture degradation.

The ballast, itself, will work on AC (any line frequency) and DC. Very flexible little sucker - with no moving parts!
=====
Ron


Top
 Profile  
 
 Post subject:
PostPosted: May Wed 06, 2009 9:25 pm 
Moderator

Joined: Jan Thu 01, 1970 1:00 am
Posts: 25299
Location: Detroit, MI USA
There apparently were other TV's that used plug-in ballasts besides the Motorolas we are so familiar with.

I remember having a catalog many years ago, perhaps in the late 50's-early 60's which had a section for TV replacement ballasts. Can't remember if it was JFD, Clarostat, or which manufacturer. But there were several different TV brands listed, maybe 6 or 8 different ones.

I have since seen those ballasts turn up at radio shows or on the shelves of old-time parts distributors. Doesn't look like the others were ever popular, but only rarely does a replacement ballast for the Motorola turn up.

_________________
Dennis

Experience is what you gain when the results aren't what you were expecting.


Top
 Profile  
 
 Post subject:
PostPosted: May Wed 06, 2009 9:37 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 558
Location: Kalamazoo, MI
I got a few of these Motorola TS-4Js sitting up in the attic awaiting restoration, and this is what I had been considering as an option for replacing the ballast tube.

First of all I was planning on running the 7JP4 off a filament transformer. And I was planning on replacing it with a resistor in the filament string. But eventually I noticed that without the crt, if I were to tie the 2 separate 73v filament strings together then substitute the 2 25L6s with 12L6s, it would add up to exactly 120v. Then I wouldn't need any resistors at all.

But I would still need 2 of the resistors from the ballast tube, not in the filament string, I think one goes before the voltage doubler, and one inbetween 2 of the filter caps. But I thought with just 2, I could empty out an old ballast tube, have enough room to stuff 2 of those power resistors in. Then I could mount the filament transformer for the crt under the chassis, and have a more original look to the top of the chassis.

I believe that a 12L6 is exactly the same as a 25L6 except for the filament voltage, but I'm not 100% sure about that. Plus there might be some other unforeseen consequences of messing with the filament string in that way. Anyway, it'll probably be months before I work on these sets, I have plenty of time to revise my plans before then.


Top
 Profile  
 
 Post subject:
PostPosted: May Wed 06, 2009 10:19 pm 
Moderator

Joined: Jan Thu 01, 1970 1:00 am
Posts: 25299
Location: Detroit, MI USA
12L6 filament current is 600 ma, compared to 300 ma for the 25L6.

_________________
Dennis

Experience is what you gain when the results aren't what you were expecting.


Top
 Profile  
 
 Post subject:
PostPosted: May Wed 06, 2009 10:31 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 558
Location: Kalamazoo, MI
Mr. Detrola wrote:
12L6 filament current is 600 ma, compared to 300 ma for the 25L6.



My plan seemed almost too easy, I knew there had to be some reason why it wouldn't work...


Top
 Profile  
 
 Post subject:
PostPosted: May Thu 07, 2009 12:02 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 644
Location: Crystal Bay, NV
Dennis, right again! There were several ballast sets. Here are some I found:

Stewart Warner AVC1, AVC2, AVC3 and AVT1

Belmont (Raytheon) A-10DX24, 10DX21, 10DX24, Coronet, Observer, A-7DX22, 7DX21 and 18DX21.

One reference shows Teletone 220 (also Firestone 13-G-3 and 13-G-33), but I think these use ordinary power resistors.

The Belmont-Raytheons are not very rare. Don't hear much about their ballast tubes though. Both Stewart Warner and Raytheon were well established with good engineering departments.
====
Ron


Top
 Profile  
 
 Post subject:
PostPosted: May Thu 07, 2009 2:25 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 404
Location: Greenville, SC USA
Regarding the different Line Voltages and frequencies, I *think* the 25 cycle AC was mainly adjacent to the New York Central Railroad lines. I can remember seeing incandescent bulbs flicker at some of the local train stations back in the 1950's.

The DC line voltage was a holdover from the old DC generating stations that were built by Tom Edison's "Edison Electric Company." Of course, this later became "Consolidated Edison" or "Con-Ed". There were still a few "neighborhood" DC generating plants operating as late as the late 1950's!

"Now back to your regularly scheduled thread."

Bill


Top
 Profile  
 
 Post subject:
PostPosted: May Thu 07, 2009 3:11 am 
Member
User avatar

Joined: Mar Sun 01, 2009 10:27 pm
Posts: 4123
Location: Ann Arbor, MI
jeyurkon wrote:
Some of the oldest radios without AGC control the gain by adjusting the filament current. Those would be incredibly sensitive to line voltage variations. The brochures for my Majestic 71 boast that is uses a pot for the volume control rather than adjusting the filament current. It's still pretty sensitive to the line variations in my house since it's lacking AGC.
John


If I understand correctly, when the filament current was used to control the volume
the current would be lowered way below what would normally be considered the
correct current. It would be lowered to the point the tube would almost quit working.

If a tube that is run at anywhere near the normal current is sensitive to the exact
current, you may have a weak tube.


Top
 Profile  
 
 Post subject:
PostPosted: May Thu 07, 2009 4:50 am 
Member

Joined: Feb Thu 26, 2009 6:00 pm
Posts: 150
Ron, the ballast for the TS-18 was different. If you need the 17A470303, vacuumtubesinc.com has two in stock. Unfortunately the one I needed isn't.

A ballast that really does regulate current would help with the heaters as you suggest. When you have a series string, even when they are rated at the same current, they don't all have the same thermal mass. The CRT cathode heater can have a smaller mass and then it heats up more quickly and it's resistance then is higher than the rest. So it dissipates much more power than intended and looks like a light bulb for a few moments.

I haven't found all the tests I did. I was rather disgusted with all the time, effort, and money I expended on this project and them reaching the conclusion that it was a wasted effort. (I may still rebuild the ballast, but with nichrome)

The wire that I heated in air burned up when it started to get yellow.

The ballast that I tested was a 17A485459. The 37 ohm segment that fed the voltage doubler was burned out. The 111 ohm and 132 ohm segments both tested out at 100 ohms. These were for the heaters and so should operate at 0.3 amps. The 200 ohm section tested out to 200 ohms within 1%.

The first test run that I did was with one of the filaments that tested at 100 ohms. My supply only went up to 30 V so I didn't feel that this was a complete test. The resistance varied from 100 ohms up to maybe 130 ohms over this range. The plot can be seen here.
https://www.msu.edu/~yurkon/Documents/ballast.pdf
The kink is a measurement error. I think I had a bad contact on the meter lead at this point.
The ballast dissipated a maximum of 8 watts during this test.

The later test is the one that I haven't found the data for yet. I can only report what I sent in an email to the person that gave me the ballast.

I used a variac to control the voltage. My old Majestic 71 was rated to operate over a range of 90-130 VAC so I thought this was a reasonable range to expect the line voltage to vary and adjusted my test accordingly. I ended up dumping 50 watts into the ballast and it was getting extremely hot. I didn't see any glow yet. The resistance never went through a critical point. The plot was linear enough that I didn't feel it provided much regulation.

A ballast provides current regulation, but what you want to regulate is voltage. For that to happen you need a constant load impedance. I'm not sure a TV set provides that.

Good luck with your tests. I'm not sure that you'll be able to use an infrared thermometer though. Especially if it's a glass tube ballast. Most thermometers require a couple of cm wide area to look at. The transmission of glass in the infrared changes the reading completely.

I purchased a Thermo Hunter VF-3000 for work just to be able to do that and image a 1mm wide spot on a target that was being heated by beam. It can do this only in two color mode. In that mode it can only read down to 600C so it wouldn't work for the ballast test. The one color mode will go lower, but not work through glass.

John


Top
 Profile  
 
 Post subject:
PostPosted: May Thu 07, 2009 6:33 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 5054
Location: Woodinville, WA USA
Quote:
Phil --- that resistor network looks very well made, but rather incongruent with the rest of the chassis. <snip> Yet the cold resistance of the ballast is only 105 ohms. (Did you use 145 ohms, Phil?)

Yah, I made no attempt at building an authentic-looking replacement, just something that worked. And because of the heat, you couldn't hide them under the chassis, as some people do with replacement electrolytics.

The original ballast in mine had four elements, not five. Two of them were 105 ohms, which I made up by connecting 100-ohm and 5-ohm resistors in series.

Regards,

Phil Nelson


Top
 Profile  
 
 Post subject:
PostPosted: May Thu 07, 2009 3:54 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 644
Location: Crystal Bay, NV
John, I have not come across any glass ballasts. Perhaps these are replacements. I have seven or eight of the TS-4 and TS-18 sets. I don't need any ballasts at the moment, but, since there are so many to these great sets around, it would be nice to have a way to repair the ballasts.

The plot you show (voltage vs. current) does show a linear resistance. Without the temperature, we cannot tell that the resistance/temperature coefficient would be. Also, the ballast is designed to drop more than 30 volts, so we should not expect to see a "resistance knee" yet.

Phil, The reason I asked about the 105 ohm resistance is because I think that the correct value is 145 ohms. If you measure the heater voltages in your set, I expect they are a little high. With your obvious attention to detail, you will want to check this.

Rebuilding the ballast with nichrome wire wire sounds like a good idea to me. Even if it only acts like a big resistor, it is an excellent solution. There are still many details to consider: the length of the original wires; the size of the nichrome wire which matches the hot resistance and has about the same length and the method of attaching the wire. I'm out of town until June 1st so I expect that all these issues will be solved by the time I get back!?
====
Ron


Top
 Profile  
 
 Post subject:
PostPosted: May Fri 08, 2009 5:20 am 
Member

Joined: Feb Thu 26, 2009 6:00 pm
Posts: 150
Ron, I had measured the diameter (gauge) of the wire. Also, the diameter of the coils. I'll have to check the price of the nichrome though.

I found the data for the variac test. This was with the 200 ohm segment of the 17A485459.

https://www.msu.edu/~yurkon/Documents/200_ohm_ballast.pdf

I did the test over the range of 5 VAC to 100 VAC. The cold resistance was 200 ohms, at 100VAC it was 213 ohms. The dissipation at 100 VAC was 47 watts.

John


Top
 Profile  
 
 Post subject:
PostPosted: May Sat 09, 2009 7:01 pm 
Member

Joined: Feb Thu 26, 2009 6:00 pm
Posts: 150
Here's an interesting related link.

http://www.bunkerofdoom.com/xfm/amperite/AMPR_AB51.html

John


Top
 Profile  
 
 Post subject:
PostPosted: May Sun 10, 2009 1:29 am 
Moderator

Joined: Jan Thu 01, 1970 1:00 am
Posts: 2302
Location: Seattle, WA, USA
Here is my solution to the VT-71 ballast problem. I settled on this solution after trying three other designs, one of which all but electrocuted me. I started working on this problem when I was a sophomore in high school and I finally built this last one after my sophomore year in college. The tube is about 1" longer than the original and otherwise looks similar. All of the copper parts are cut by hand with nothing but a hack saw an electric drill and a file, from copper plumbing pipe. The socket is from AES, the resistors are Ohmite Brown Devil Vitreous Enamel Power Resistors bought from Mouser, and the hardware was bought from McMaster Carr. The wiring is tinned bronze (to stand up to the heat) and it is insulated with two layers of fiberglass bought by the roll from Mouser. Because the tube operates so hot, all of the connections are made with high temperature crimp connectors bought from my local ACE Hardware and intended for stove repair. In fact, when it is fully warmed up I can easily melt solder on the copper tube, however, all of the components I used can take the heat.

As for the resistor values, I used a custom-built variable power resistor array which I dubbed the "Lot-A-Watt 325" to adjust the values of the individual resistors until the B+ and heater voltages agreed with those printed in the original Motorola service manual.
Click Here for a thread about my 'Lot-of-Watt'
I did this calibration at my line voltage (~120v) so the resistor values are likely higher than they would have been originally. The two resistors in the heater string are each 150 Ohm @ 20 Watt, while the two resistors in the B+ circuit are 200 Ohms and 50 Ohms each at 10W. This design will only work in the TS-E and later chassis because the TS-B,C, and D require another resistor nominally 22 Ohm resistor for which there is no room in the tube. If anyone would like the templates I used to cut out the pieces along with all the specifics needed to put one of these together, please PM me.

Image

-Matthew D'Asaro

_________________
If it ain't broken, you’re not trying hard enough...


Top
 Profile  
 
 Post subject:
PostPosted: May Sun 10, 2009 2:11 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 9321
Location: Redlands CA
Nice work Matthew, the Motos originally had a small square of Asbestos above the Ballast to protect the cabinet from the heat.

This seems to run hotter than the original ballast and it's taller so I was wondering if you are using anything to keep the heat away from the cabinet?


Top
 Profile  
 
Post New Topic Post Reply  [ 68 posts ]  Moderators: 7jp4-guy, Mr. Detrola Go to page 1, 2, 3, 4  Next

All times are UTC [ DST ]


Who is online

Users browsing this forum: carlpete50 and 10 guests



Search for:
Jump to:  




















Privacy Policy :: Powered by phpBB