Choke not working

[RyreInc]

I've constructed a circuit to charge a capacitor bank. A choke is in series with the 12V secondary of a transformer (pre-bridge rectifier) to limit current to about 5mA. The choke is 5.75H, 330ohm, which, at 60Hz, should have an impedance of about 2200ohm, right? 12V/2200ohm=5.4mA.

However, the 400uF of capacitors are charged in the first few cycles, instead of about 10s. Additionally, when I replace the caps with a 100k potentiometer, the current is definitely not limited to 5mA as resistance is decreased.

I measured both the inductance and resistance with meters, so I know the choke isn't defective.

Also, the active current limiter I use (2 BJTs, 2 resistors) as an alternative to the choke works as designed.

What is going on here??? Thanks for your help.

 

[Skogsgurra]

Is the choke in the DC path? Or AC?

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[RyreInc]

AC - straight off the transformer secondary, before rectifier. Rectifier is bridge-type, no center tap on transformer.

 

[VEBill]

...about 10s.

Where did 10 seconds come from?

 

[Skogsgurra]

The math seems OK. Around 10 seconds to charge 400 uF to around 10 V.

A 12 V choke with a 330 ohms resistance seems to be a very small one. My bet is that the core saturates heavily. Use a "better" choke. One with a larger core.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[DHambley]

Does your choke data sheet show a rated current? I bet Skogsgurra is right, it's probably saturating.

 

[RyreInc]

Choke rating is 40mA, transformer rating is 950mA on secondary.

One thing I had previously dismissed was another bridge rectifier coming off the secondary, prior to the choke, which supplies a DC bus for control circuitry. Schematic is attached. (The zener diode is actually a 3-terminal voltage regulator; capacitor is after regulator)

 

[RyreInc]

I just realized how strange this circuit may appear. Perhaps it helps to know that this is a low voltage version of a pulsed power charging circuit; the high voltage equivalent we use charges 400u in about 10s (approximate), using a choke to restrict in-rush current.

 

[RyreInc]

Aha! The common ground, in combination with the first bridge rectifier, is the culprit. There is a current path I hadn't noticed before, that uses both rectifiers.

I can either lift the charging caps from ground, or change the first rectifier to a single diode.

Thanks for everyone's help!

 

[Skogsgurra]

I fail to spot that path. Can you describe it?

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[RyreInc]

On the negative half of the AC wave the current flows out the bottom of L2, then through D6, C1, and D3--not D7 like it "should". The common ground provides that path. The choke therefore only works only on the positive half of the AC wave.

 

[Skogsgurra]

Quite! So it is saturated from DC, all the same. Didn't see that.

The same is true for the positive half-wave.

Looks like you have to separate those two circits.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[zzzjim]

Spotted the the parts in parallel = not needed, but worried about D9 as it is fed directly by (BR1)=D1 to D4 !

Reminds me of voltage multiplier circuit based on Bridge rectifiers isolated by caps in ac feed .

 

[zzzjim]

Sorry oops ..
( just spotted zener (D9) is really a regulator comment )