3
\$\begingroup\$

I'm using a TLV61220 IC boost converter to get 3.3V from a 1.5V AAA battery. I've built the circuit as specified on the datasheet (p. 1 of http://www.ti.com/lit/ds/symlink/tlv61220.pdf) with R1 at 1MΩ and the appropriate value for R2. The converter outputs 3.3V as expected with a small load, but appears not to be able to handle loads that draw more than a couple of mA (the output drops to about 1.8V with a 1K load, for example). The battery should of course be capable of supplying much more current than this, and figure 3 on p. 7 of the datasheet suggests that the converter should be too.

I was wondering if this sort of performance is to be expected given that the prototype circuit is on a solderless breadboard (and that the TLV61220 is hooked up via one of the crappy SOT-23 to DIP converters sold by Digikey). I've read that high frequency circuits do not perform well on breadboards, but I'm unsure whether it's plausible to put the extremely poor performance down to this factor alone. On the other hand, if the circuit was not hooked up correctly, then it seems unlikely that I'd be getting 3.3V output with small loads.

(I've tried with two different TL61220s with exactly the same results.)

\$\endgroup\$
  • \$\begingroup\$ Update: It seems that the breadboard is a least a factor. I tried replacing some unnecessarily long wires with shorter ones and got it up to 2V from 1.8V for a 1K load. \$\endgroup\$ – foldl May 31 '14 at 23:08
  • \$\begingroup\$ What kind of inductor are you using? \$\endgroup\$ – Zuofu May 31 '14 at 23:15
  • \$\begingroup\$ One of these: digikey.com/product-detail/en/TSL0808RA-4R7M3R5-PF/… \$\endgroup\$ – foldl May 31 '14 at 23:17
4
\$\begingroup\$

Solderless breadboards are useless for prototyping circuits over a few hundred kHz. For high frequencies, you need to construct your prototype circuits "deadbug" style over a copper clad ground plane, like so:

deadbug construction

\$\endgroup\$
  • \$\begingroup\$ Thanks. Just in case this info is useful to anyone else, I can confirm that the solderless breadboard was the issue. I tried constructing the circuit on a soldered breadboard (but not using a ground plane as in the picture) and the output went up to 2.5V with a 1K load. So the layout and construction of the circuit does indeed make a huge difference in this case. \$\endgroup\$ – foldl Jun 1 '14 at 2:10
  • 2
    \$\begingroup\$ @Bitrex OK - I'll bite - what's "Oliver"? \$\endgroup\$ – Russell McMahon Jun 2 '14 at 0:04
5
\$\begingroup\$

I wouldn't expect a 1MHz boost converter to work well on a breadboard. The SOT to DIP adapter would add parasitic inductance. The breadboard itself would add parasitic capacitance.

In general, switch mode power supplies can be layout sensitive. If I were doing a PCB layout, I would just replicate the layout of the evaluation kit for TLV61220 (pp.6-7).

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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