Welcome to Australia's only Vintage Radio and Television discussion forums. You are not logged in. Please log in below, apply for an account or retrieve your password.
Australian Vintage Radio Forums
  Home  ·  About Us  ·  Discussion Forums  ·  Glossary  ·  Outside Links  ·  Policies  ·  Services Directory  ·  Safety Warnings  ·  Tutorials

Tech Talk

Forum home - Go back to Tech talk

 32V Power Supply
« Back · 1 · 2 · Next »
 Return to top of page · Post #: 1 · Written at 8:07:06 PM on 2 May 2017.
Flakes's avatar
 Location: Adelaide, SA
 Member since 27 February 2010
 Member #: 630
 Postcount: 392

Hi All

Looking to build a power supply for powering some 32V radios.

Just starting to design it, wanting input from anyone who has built one and any pitfalls with there design.

I am thinking of doing a fully regulated supply so when the switch on the radio is turned off the Terminal voltage doesn't rise too much and then cook the heaters on turn on.

Also I was thinking of making it 4A or do I need more / less for most of the 32V radios out there.

Dan

PS any websites or Mag articles on it would also be a good read...


‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾
Valve radios, They just don't make them like they used to

 
 Return to top of page · Post #: 2 · Written at 10:13:34 PM on 2 May 2017.
Marcc's avatar
 Location: Wangaratta, VIC
 Member since 21 February 2009
 Member #: 438
 Postcount: 4337

What one needs to do is have a hunt to see if you can spot some models that are Vibrator (Operatic) sometimes they bother to tell you the current draw.

You need to look at the fine print with regulators. Many as you suspect, loose control with no load. Typical examples are 7805, 7905 which need 5mA of load to maintain control. HV regulator LR8 needs 0.5mA (Its voltage divider is part of that load: But may not be enough.

Then there is the mains: Whilst some will argue that it never happens: It does here. The way things are wired with the transmission lines, can cause havoc with the Solid State Regulators. Should either one of Negative, or Positive, be earthed to anything, this will result in a big spark & the regulator to go out of control.

I note that both Harold Sheeran & I fit a 0.01mfd (103) Line Cap to our Regulated power supplies & you earth the set via that.


 
 Return to top of page · Post #: 3 · Written at 6:25:35 AM on 3 May 2017.
Wa2ise's avatar
 Location: Oradell, US
 Member since 2 April 2010
 Member #: 643
 Postcount: 766

Computer printers often use separate power bricks that are around 30VDC. However, they are switching power supplies, and likely will produce RFI into the radio. And you probably need more current than they usually are rated for.


 
 Return to top of page · Post #: 4 · Written at 9:43:21 AM on 3 May 2017.
Marcc's avatar
 Location: Wangaratta, VIC
 Member since 21 February 2009
 Member #: 438
 Postcount: 4337

Leaving things not running with a PSU that does not need to be running is rather wasteful long term and I don't really consider is desirable, or really necessary to use a regulated supply with such a dirty system as a mechanical vibrator.

The original 32V power supply would have no surge 12+12+6 Lead acid storage batteries (not car). It may be better to consider a conventional system with perhaps globes or Zeners to cut down the surge. A couple of old devices here use a relay system to handle over voltage.

Switchmode power supplies, NBN boxes (wireless), mains powered LED floodlights, Fax machines, Computers, Compact Florescent Lights, are along with a variety of other switch mode devices, the enemy of radio reception. We no longer have RFI standards: If we did the place would be in darkness & grind to halt.

Do not use LV caps with anything like a Mechanical Vibrator, (or Kettering points ignition): The back emf can be 200V, or more. That's why the caps are that big & one is over 1KV & notoriously unreliable. If I consider that there will be back emf onto regulators, I will often add a series diode in addition to what should be on the regulator.

As an aside, all of the computer peripherals here are fed via a contactor box, one switch kills the lot. All of it has a delay timer so that if the power fails (common), or the "re-closer" operates 3 minutes of uninterrupted supply must elapse, before the system gets power.

With that contactor, none of the non essential devices are powered, when they don't have to be.


 
 Return to top of page · Post #: 5 · Written at 10:38:38 AM on 3 May 2017.
Ian Robertson's Gravatar
 Location: Belrose, NSW
 Member since 31 December 2015
 Member #: 1844
 Postcount: 1904

Here's a tested design with suggested component source references.

The heatsink could be bigger than suggested in the circuit, the bigger the better. U1 is rated a 5A continuous and will shut down on over current and over temperature. It's probably better bolted directly to the heatsink without washers, making the heatsink hot to 32v.

There are diodes to protect the regulator against accidental shorts of all capacitors, making the power supply bulletproof.

R1 reduces conducted emissions into the mains caused by the peak diode currents in the bridge. These emissions otherwise centre on about 450kHz.

32 volt, 4 amp Power Supply


 
 Return to top of page · Post #: 6 · Written at 10:49:02 AM on 3 May 2017.
Ian Robertson's Gravatar
 Location: Belrose, NSW
 Member since 31 December 2015
 Member #: 1844
 Postcount: 1904

I should add a comment about the TO-220 package pinout.

With the LM338T part number facing you, and pins facing down, pin numbers go 1, 2, 3.

Don't even think about substituting another regulator for the LM338T, by the way. Anything else won't cut the mustard for this job.


 
 Return to top of page · Post #: 7 · Written at 8:13:37 PM on 3 May 2017.
Flakes's avatar
 Location: Adelaide, SA
 Member since 27 February 2010
 Member #: 630
 Postcount: 392

Hi Ian

Look forward to seeing it. I am starting to collect a few of the "Farm" radios. I think that they should be restored as they were and not "Converted" like a radio of mine in another post.


‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾
Valve radios, They just don't make them like they used to

 
 Return to top of page · Post #: 8 · Written at 9:14:41 PM on 3 May 2017.
Brad's avatar
 Administrator
 Location: Greenwich, NSW
 Member since 15 November 2005
 Member #: 1
 Postcount: 6347

Circuit diagram uploaded to Post 5.


‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾
A valve a day keeps the transistor away...

 
 Return to top of page · Post #: 9 · Written at 8:11:25 PM on 4 May 2017.
Marcc's avatar
 Location: Wangaratta, VIC
 Member since 21 February 2009
 Member #: 438
 Postcount: 4337

Do not ever overlook the humble CPU fan. I have used more than one in a PSU to keep air moving across heat sinks. One unit I made has a heat sink on the actual board. So the fan went into the base of the vintage looking PSU box and actually blows cold air into the box aimed at the heat sink, cooling everything (Box is of course vented).

Marc


 
 Return to top of page · Post #: 10 · Written at 1:55:43 PM on 5 May 2017.
Ian Robertson's Gravatar
 Location: Belrose, NSW
 Member since 31 December 2015
 Member #: 1844
 Postcount: 1904

Yes Marc, very useful for smaller heatsinks and if there is no natural airflow.

You can calculate the working temperature of your heatsink, with and without fan.

With a 4A load, the regulator will be dissipating 28 watts. You can find heatsink calculators on the web.

You will get a number for Theta Sink to Air. It will be in degrees C per watt. Add that to the maximum ambient and if it's less than about 95 degrees you are safe.

You will need to enter airflow, either forced or natural.


 
 Return to top of page · Post #: 11 · Written at 9:57:06 AM on 6 May 2017.
Marcc's avatar
 Location: Wangaratta, VIC
 Member since 21 February 2009
 Member #: 438
 Postcount: 4337

Fans are a great boon in keeping things cool in really hot temperatures. One example I saw was a tube heat sink so everything got moving air.

The CM Valiant had its (sealed) engine computer in the engines air intake.


 
 Return to top of page · Post #: 12 · Written at 4:51:30 PM on 6 May 2017.
Flakes's avatar
 Location: Adelaide, SA
 Member since 27 February 2010
 Member #: 630
 Postcount: 392

Got sidetracked today trying to align electrically and mechanically a Philips Radioplayer 124 so haven't had a chance to build the power supply.

I have a 3A transformer here that I will use for testing. and I will put an Amp meter in series just to see what the radio draws when I get a chance to start on it....

Still need a 25hr 8 day week.


‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾
Valve radios, They just don't make them like they used to

 
 Return to top of page · Post #: 13 · Written at 11:44:14 PM on 6 May 2017.
Marcc's avatar
 Location: Wangaratta, VIC
 Member since 21 February 2009
 Member #: 438
 Postcount: 4337

I don't like the sound of trying? The worst one I have had recently was the 92X Egg Crate, that had been fully adjusted (anything that moved). The IFt's did adjust per plan in seconds. Re-resonating the aerial coil trimmer C3 & fixed cap C1 was annoying but ......That happens (rarely).

Be very careful as to how much signal you stuff down the throat of the signal grid. You only need around 50uV. Too much will cause the AVC / AGC to kick in and that will kill any chance of getting the IF transformers right.

With the dial if its been messed with eg. The string broke. Some sets do have a "set pointer here" and loose drums do slip.Otherwise its get it in the same place each end of the dial. Then you play with the oscillator.

If you really need a 32V welder, somewhere I have a 32V 20A charger transformer with a long duty cycle (and a weight to match) primary current 5A.

I know the day problem: One neighbour landed a slasher here, with a broken yoke on its propeller shaft; With another one I spent a large slice of the day removing the cylinder head from his tractor, which means that after it's sorted, there goes several hours putting it back & then there is a horse float that needs wheel bearing work & new tyres. No boredom this week.


 
 Return to top of page · Post #: 14 · Written at 2:58:22 PM on 7 May 2017.
Flakes's avatar
 Location: Adelaide, SA
 Member since 27 February 2010
 Member #: 630
 Postcount: 392

Well it's 1/2 way through Sunday. All the Trying finally sorted out the Philips. Someone had unsuccessfully replaced the Gang. The Shaft on the Gang was shorter than the original one and the dial string popped off when the baffle board was screwed in the case. I had to "Fix" and then re string both dial and knob strings, IF alignment, Osc alignment, now sounds sweet..... 32V stuff can wait another day. Thanks to Ian for the Circuit.


‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾‾
Valve radios, They just don't make them like they used to

 
 Return to top of page · Post #: 15 · Written at 10:27:14 AM on 9 May 2017.
Ian Robertson's Gravatar
 Location: Belrose, NSW
 Member since 31 December 2015
 Member #: 1844
 Postcount: 1904

Just thinking, Flakes, your concern about using an unregulated supply is probably unjustified. Cold valve heaters have a lower resistance and will drop the voltage on turn-on until they warm up a bit. You'd probably get away with a brute force R - C filter - adjust the R to get your 32V and add C until the hum stops!

But then, you'd have to fiddle the R when changing radios. Most 32V radios have a lot of valves (hence heaters) to make up for the reduced stage gain / power output when running on a B+ of only 32 volts.

I've never seen a vibrator 32V radio. 6V, yes, not 32V. Even so, the load of the radio will be rendered predominately resistive due to the valve heaters and the heavy filtering on the input to the Royer converter (vibrator). No "back EMF" issues.....

I use an unregulated power supply for my 6V vibrator radios that originally powered an inkjet printer. It's a large "lump in a lead" 12V, 2A supply with conventional transformer, bridge, cap. Literally found on the side of the road! I just added a 5w resistor between the bridge and the cap. Can't remember how many ohms, 4.7 I think Just to get the 6 volts.


 
« Back · 1 · 2 · Next »
 You need to be a member to post comments on this forum.

Sign In

Username:
Password:
 Keep me logged in.
Do not tick box on a computer with public access.