0
\$\begingroup\$

I have been doing some work with an energy harvesting IC from Linear Technology. It is the LTC3588-1.

In the application information section of the datasheet, there's the following setup for the IC (which I am not using/implementing, it just made me wonder...):

enter image description here

The 150K resistors are there to keep the maximum current within the limits of the IC. The PZ1 and PZ2 are the input pins for an ac source and there is a bridge rectifier inside the IC—hooked up to those pins.

My question, would it make any difference if instead of having two 150K resistors in series with each lead of the voltage source, you had all four resistors in series with a single lead?

Something like this:

schematic

simulate this circuit – Schematic created using CircuitLab

I'd just like to have someone's input on what the differences are between the two configurations. Thanks in advance.

ADD: An isolation transformer is actually added in the real circuit for obvious safety reasons. And the actual implementation I am working one DOES NOT use mains power like this. I was just curious about the resistors and thought about asking.

\$\endgroup\$
6
  • 1
    \$\begingroup\$ I am not too keen on either of them.... \$\endgroup\$
    – Trevor_G
    Commented Dec 11, 2017 at 3:48
  • 1
    \$\begingroup\$ If the end user can contact the output, even the original circuit won't meet certain safety requirements (0.25mA maximum leakage current). Your one is a potential death trap. The two resistors in series are probably intended to be safer but if someone uses very small or poorly made parts they're also potentially deadly. Jim has explained why yours is worse. \$\endgroup\$ Commented Dec 11, 2017 at 4:19
  • \$\begingroup\$ @SpehroPefhany An isolation transformer is actually being used in the real circuit. This is only for testing (replicating a source with a high internal impedance source) \$\endgroup\$
    – Big6
    Commented Dec 11, 2017 at 4:29
  • 2
    \$\begingroup\$ @sixcab Good to hear. It is irritating when example circuits in datasheets are impractical or dangerous. \$\endgroup\$ Commented Dec 11, 2017 at 4:31
  • 1
    \$\begingroup\$ An isolation transformer does not stop a circuit from killing you.... \$\endgroup\$
    – Trevor_G
    Commented Dec 11, 2017 at 4:31

3 Answers 3

2
\$\begingroup\$

Imagine a case where your 120 VAC 60 Hz wall outlet is wired incorrectly (or the power cord is wired incorrectly), with the hot (120 VAC) and neutral (~0 VAC) lines reversed. In your figure where R1 through R4 are wired in series and connected only to P1, this reverse wired 120 VAC power source would apply 120 VAC directly to U2 pin P2, which would be extremely dangerous. Placing two 150 kohm resistors in series with U2.P1, and two 150 kohm resistors in series with U2.P2, prevents the direct application of 120 VAC to either the P1 or P2 inputs, among other things.

\$\endgroup\$
0
3
\$\begingroup\$

First of all I have to point out both circuits are extremely hazardous and you should not even think about implementing something like this if you are not extremely aware of how this works and the dangers involved both to you and whomever may ultimately be using anything tied to this circuit.

The circuit indicated by the manufacturer does however provide a significantly higher margin of "safety" over your four in series circuit.

Having two resistors in series in each path increases the overall breakdown voltage versus having a single resistor. Having two in each line also provides a limited amount of isolation between the regulator and the high voltage supply.

By using proper PCB layout techniques you can limit the incursion of the high voltage to a localized point somewhere near the edge of the board and, if need be, shield it from touch.

In your four in series example, one side of the high voltage must be passed all the way to the regulator and becomes an extreme hazard. It also makes it very difficult to lay out the PCB to the appropriate clearances and creapage distances.

Further with the four in series, only a single fault to short the series to ground would put the full mains voltage onto the regulator. This would no doubt destroy it, but would also have a high probability of feeding that high voltage out to whatever the regulator is feeding, and potentially at the user.

The recommended circuit, on the other hand, requires two faults in order for that to happen.

Again, I can not stress it enough, this circuit is deadly in any form. Also, do not rely on an isolation transformer to save you.

\$\endgroup\$
2
  • \$\begingroup\$ I am not using this circuit in reality. I was just curious about the resistors and if they make any difference. This is not something I am implementing. Thanks for the response \$\endgroup\$
    – Big6
    Commented Dec 11, 2017 at 4:38
  • 1
    \$\begingroup\$ @sixcab that's good to hear. Trouble is this is a global website. The next guy that thinks about doing this, and finds this question, may not be so wise. \$\endgroup\$
    – Trevor_G
    Commented Dec 11, 2017 at 4:39
0
\$\begingroup\$

The chips specs for PZ1 PZ2 are 0V to Vin on pin 4, with pin4 rated at –0.3V to 18V or 25mA pk <1ms and < 1% d.f.

Thus even with 10% of line voltage on Neutral ( max line current) and 10x line voltage for routine inductive power line transients, there is no way to meet the IC specifications with your suggestion.

BTW the 4x150k resistors are mainly to increase the dielectric breakdown voltage and for small SMD share the small dissipation.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.