[ARC5] 28 volt high-current supplies?

Sun Feb 27 21:35:56 EST 2022

Success! I found the problem. It just seemed like SOMETHING was being 
told to draw way too much, so I started looking at the control circuitry.

There is a small (10 watt) variable resistor (remember the cigar-sized 
Ohmite "Dividohm" with the exposed resistance wire and a sliding clamp)? 
A small version of one of those. It was completely open circuit! And its 
function is to set the voltage output via a carbon pile regulator which 
drove the 400 Hz generator field. So with it open, the field (rotor) was 
driving wide-open all the time. Which means the RPM was excessively high 
and likewise lots of power going somewhere it shouldn't. Glad I didn't 
fry it.

Given the clamp position, it looked like it should be somewhere around 
350-400 ohms. So I tacked in a handy 383 ohm power resistor, and left 
the 150 ohm wirewound adjustment pot (which also had some oxidation but 
improved after working it back and forth) in the center where it had been.

Fired it up with the usual horrendous starting surge, near 20 amps 
primary transformer current on the Amprobe, transformer ratio 18:210 
taps = 11.7:1. That 220 amps I measured on the transformer secondary 
wasn't so far off after all! Primary current dropped to 10 amps as the 
machine spun up, as before... but suddenly, there was an audible entry 
to closed-loop operation just like letting off the gas, and the primary 
current immediately dropped to just 3.5 amps. That translates to 41 
secondary amps which is (estimated 41/1.8)* = 23 DC amps... not bad for 
nameplate 115 full load amps. *(Signal Transformer's power supply design 
catalog says the RMS secondary current is 1.8 amps the DC current in a 
capacitor-input, bridge rectifier system).

DC power supply volts under that load now up to 23 with less ripple, RPM 
sounding more reasonable too. Should be much better once my 220,000 uF 
filter cap gets here later this week.

And without even tweaking the pot, there is exactly 115 VAC measured 
with my Triplett 630-A on all three line-to-lines (A-B/gnd, C-B/gnd and 
A-C)! So it really is a 115 delta with corner ground. I still can't tell 
if that will work with a Collins 618T-2 or not. The schematic says it 
should, showing a 115 delta input to the HVPS, but the literature keeps 
referring to 115/208Y which is the current aircraft standard...

Haven't loaded it yet because I don't have the bridge rectifier properly 
cooled (just a 3x4" heat sink that might not be enough for frequent or 
prolonged use).
I will check the frequency too, although it's not adjustable (fixed 
parallel-resonant LC).

About the resistor, I don't have an exact replacement (750 ohm)...does 
anyone have one on hand? 500 ohm would also work fine. Or I'll just 
mount the 383 fixed resistor in there! :)

On 2/26/22 15:23, Charles wrote:
>
> It's not the capacitor on phase A.  Oh pooh. Now I have to look for 
> zebras rather than horses...
>
> Yes, I am aware of the basic motor physics ;) That "about a second" is 
> with a stiff supply. Mine is apparently not stiff enough <that's what 
> she said> :D
>
> My original plan was to acquire a 618T with the 400 Hz three-phase 
> supply - usually much cheaper than the 28 vdc or the 400 Hz single 
> phase versions. But this inverter is becoming a real white elephant. 
> It would be running all the time, not PTT like the ART-13.
>
> Anyway I am tempted to go buy a pair of car batteries and find out, 
> just because I'm stubborn that way! (My two "ARC-5" receivers have 
> their dynamotors, but I only run one at a time, and the 28v supply I 
> built with a handy Stancor RT-204 transformer has no trouble).
>
> Charles, WB3JOK/0 since '76
>
>
> On 2/26/22 15:13, Christopher Bowne wrote:
>> No, just a an Ames (Harbor Freight) 80 buck multimeter with a clamp 
>> on AC or DC ammeter.
>>
>> Peak current is always at initial 0 RPM, as the motor winds up the 
>> counter EMF brings the current down in about a second. Basic
>> DC motor physics.  Considering the instrument pedigree I take no 
>> responsibility for the accuracy of the reported result.
>>
>> As an alternative to floating batteries on the output of a DC supply, 
>> one could implement a step starter circuit, but you would need
>> some hefty dropping resistors and contactor relays.  Wouldn’t lend 
>> itself to quick PTT operation on voice.
>>
>> Chris AJ1G
>> Stonington CT
>>
>> Sent from my iPhone
>>
>>> On Feb 26, 2022, at 15:38, Charles <charlesmorris800 at centurytel.net> 
>>> wrote:
>>>
>>> Thanks, there's a data point of 5 times the nameplate.. What was 
>>> the decay time of that 160 amp peak (did you use a scope)?
>>>
>>> If I HAD a couple of spare car batteries I'd certainly try them ;) 
>>> but have you seen the prices lately? they start at $55 plus core 
>>> plus tax at Wally World...
>>>
>>>
>>> On 2/26/22 13:37, Christopher Bowne wrote:
>>>> I’ve measured the peak inrush current on my ART-13’s DY-17A 
>>>> dynamotor,  it was a nominal 160 amps.  Just float a couple of 12 
>>>> batteries across the output of your supply, make sure that they are 
>>>> fully charged and full sized automotive or deep cycle marine type, 
>>>> small one may not deliver enough peak current to preclude tripping 
>>>> the output breaker in on your high current supply.
>>>>
>>>> Chris AJ1G
>>>> Stonington CT
>>>>
>>>> Sent from my iPhone
>>>>
>>>>> On Feb 26, 2022, at 13:57, Charles via ARC5 <arc5 at mailman.qth.net> 
>>>>> wrote:
>>>>>
>>>>> I am sure I would have no trouble starting a DY-17 with a 33 amp 
>>>>> full load rating. This beast is 3.5 times that full load draw and 
>>>>> consequently any surge current would likely also be 3.5 times that...
>>>>>
>>>>> Anyway I am going to investigate shorts on the AC side, and my 
>>>>> prime suspect is the (noise reduction?) capacitor on A phase. Hope 
>>>>> it's not the armature. It does have a repair request tag from the 
>>>>> 80's or 90's (don't remember at moment) and an unreadable 
>>>>> "Serviceable" tag. The electromechanical regulator (carbon piles!) 
>>>>> will get a look after I check the cap. Something else I have no 
>>>>> experience with.
>>>>>
>>>>>
>>>>>
>>>>>> ______________________________________________________________
>>>>>> ARC5 mailing list
>>>>>> Home: http://mailman.qth.net/mailman/listinfo/arc5
>>>>>> Help: http://mailman.qth.net/mmfaq.htm
>>>>>> Post: mailto:ARC5 at mailman.qth.net
>>>>>>
>>>>>> This list hosted by: http://www.qsl.net
>>>>>> Please help support this email list: https://www.qsl.net/donate.html
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.qth.net/pipermail/arc5/attachments/20220227/1df20cb6/attachment-0001.html>