• If you like antique engines, vintage tractors or old iron, please register and join us. When you register on Smokstak, give complete answers and fill in all blanks. IF YOU ARE ON WIRELESS. GIVE YOUR CITY AND STATE! NO ZIPCODES! We get hundreds of applications, so there is no time to deal with missing information.

Rectifier Electrical / Electronic Question

DMeed

Subscriber
Last Subscription Date
07/12/2019
Here are a couple of diagrams depicting current flow during each half cycle of the supply, of course ignoring peak vs RMS etc etc.
It looks right. But how is the current flowing both ways through that winding - what is the phasing of those coils? Typically I expect coils connected like that to have the +ve end at the top, -ve end of that coil connected to +ve end of the coil below and -ve end of bottom coil to the other side of the supply. So the coils are connected in phase with each other and the 60 volts runs from the top all the way to the bottom, with 30volts at the centre. The current rises in the two coils in sync and flows top to bottom (for 1/2 the AC waveform)

(And actually the current is always flowing in one direction in both those windings in both pictures. When does it flow in the other direction for the other half of the AC waveform which is what I would normally expect to see?)

In your diagram if I put 60 volts at the top, 30 volts at the middle, 0 volts at the bottom - how does 50 amps of current flow from 60 to 30 + to - and 50 amps from 0 to 30 - to + unless the coils are connected in reverse phase. (but then it's only 30 volts of coils across 60 coming in?!)

Sorry to be obtuse. I'm sure someone has done it and proved it. I'm just trying to get my brain to figure how it does that, probably mixing up resistors and coils. I'll have to study it out more later. I gotta go to bed. Wifey is getting up very early tomorrow morning for some reason.

Because of this, the frequency will be 30 cycles. Therefore I need a transformer with twice as much iron, ie designed for twice as much voltage. A 60 cycle transformer designed with two 60 volt windings (that I will run on 30 volts, 30 cycles), each capable of 50 amps, would be perfect. If it existed lol. Will likely end up winding my own. :shrug:

Umm - I think at 30 cycles it will be rated about 1/2 as many watts as it is on 60 cycles?? - again - it takes more iron at 30 cycles so the core would have to be twice as large. (And maybe more than that, I don't know)
 

armandh

Sponsor
Last Subscription Date
09/02/2010
will the engine supply 3 KW at half speed
will the generation VR function at 30 HZ

I like the idea of large buck/boost XF(s) at correct speed/hz.
 

Vanman

Subscriber
Last Subscription Date
07/10/2019
It should do 3kw, give or take. The engine is 12-1/2 hp @ 1800 rpm, the generator 6-1/2 kW. No avr. For the reduced speed running it will be separately excited.

Being it's just going to a big battery and inverter the actual amount of power it generates is not critical, nor the exact speed it wants to run. We'll adjust the throttle to the point where everything is happiest and most efficient.

Whenever I should need more power, we can just throw a switch and revert back to full speed operation powering the AC loads directly, the inverter then charging the battery as normal.
 

Jim McIntyre

Subscriber
Last Subscription Date
07/10/2019
One thing not mentioned (or maybe I missed it) is the current waveform. If the DC source is being used to charge batteries, the current probably won't be sinusoidal. If it's not sinusoidal, keep in mind that the transformer's current rating is amps, RMS.

Current waveforms with high peak current have surprisingly high RMS values.

LTSpice to the rescue... ;)
 
Last edited:

Vanman

Subscriber
Last Subscription Date
07/10/2019
/\ Yes! This has been a concern of mine more importantly for my generator!

I downloaded the LT Spice but am not smart enough to figure out how to use it. :bonk:

Some reactance in series with the output of the rectifier could certainly help. But too much results in something akin to a square wave on the AC side? :shrug:
 

Jim McIntyre

Subscriber
Last Subscription Date
07/10/2019
I recall that adding inductive reactance is a way to improve this, but the extra weight and cost of the ginormous inductor might make it impractical. It might be that transformers used for battery chargers have a lot of leakage inductance...
 

armandh

Sponsor
Last Subscription Date
09/02/2010
as I recall
or someone once told me
when I asked why no cap.
ripply DC was better for battery charging
BS or not?

turns out to be BS, a quick google finds this
https://www.cdtechno.com/pdf/ref/41_2131_0212.pdf
but the problem is worse at lower Hz. [near the end]

if one thinks of the battery as like a big cap all those half wave bumps sum together charging
the only question is keeping up with the load. consider the inverter generator.
load goes on engine speeds up to deliver more power to the input caps.
here one of those 400Hz 3 phase sets might be better [if all it did was charging]
 
Last edited:

Jim McIntyre

Subscriber
Last Subscription Date
07/10/2019
Another aspect is that a full wave bridge presents a better load to the transformer than the full wave center tap shown in the first post. Again, this arises from the rms value of the current waveforms. Lots available on the google on this topic...
 

Vanman

Subscriber
Last Subscription Date
07/10/2019
In my case, in the diagram shown, that is not a transformer secondary, but rather an autotransformer. The supply is a generator at ~60 volts AC. The autotransformer could be used to supply a four diode full wave rectifier. By my analysis the load on it would be the same. But the rectifier losses would be doubled.

That article on battery charging ripple current is fantastic. While flooded cell batteries are not discussed, one can assume that the same principles apply. Flooded cells are likely much more tolerant, but I definitely don't want the battery undergoing two charge / discharge cycles for every AC cycle! I would have imagined that the battery capacitance would have been far and away more than enough to prevent this. Curiously that was not discussed...

A large reactor looks like the only practical way to ensure a steady current on both AC and DC circuits. :shrug:
 
Last edited:

Jim McIntyre

Subscriber
Last Subscription Date
07/10/2019
I ran a quick simulation of the autotransformer idea, and your initial hunch is correct.



With the V1 source at 60 Vrms (84 V peak) we see a load voltage of 25 VDC average (28 Vrms). The current in each winding of the autotransformer is 14 Arms, as is the current supplied by the V1 source.

I don't have a handy model for a battery, but hanging a large capacitor across R1 is gonna make the RMS current go sky high, I'm pretty sure...
 
Last edited:
Top