3
\$\begingroup\$

I'm using the TVN5-2421WI DC-DC converter where I need an output of at least +/-6V (which will then go into 2 LDOs that will drop it to +/-5V), so I'm trying to use the trim pin for this, however the datasheet doesn't specify what value resistor to use: enter image description here Datasheet: http://docs-asia.electrocomponents.com/webdocs/1341/0900766b81341219.pdf

How do you figure this out? I'm very hesitant to use a potentiometer.

EDIT: I found a datasheet for another DC-DC converter from Traco, and they have a formula for determining the trim output: enter image description here Datasheet: http://www.tracopower.com/products/ten25wi.pdf What's the likelihood I can use this same formula?

\$\endgroup\$
4
  • \$\begingroup\$ not going to work, TVN5-2421WI is a dual output supply, and only the single output models have the trim input. \$\endgroup\$
    – Jasen Слава Україні
    Commented Apr 12, 2017 at 5:27
  • \$\begingroup\$ The pinout has "TRIM" for both 'single' and 'dual' as available. \$\endgroup\$
    – deki
    Commented Apr 12, 2017 at 5:39
  • \$\begingroup\$ Have you chosen an LDO with high-frequency Power Supply Rejection to reduce the SwitchReg high frequency trash? To assist the LDO, you need Low Pass Filters before the LDO. Use resistors and capacitors in those LPF. Your attenuation will be set by cap(ESR + Z(ESL)) / R_lpf. Thus 0.01 Ohm ESR + 5nH at 30MHz (or J1 ohm), with R_lpf of 1_ohm, achieves only 6dB attenuation at 30MHz Fring (lotta SwitchRegs have parasitic LC ringing 10X or 100X the switching frequency). \$\endgroup\$
    – analogsystemsrf
    Commented Apr 12, 2017 at 7:08
  • \$\begingroup\$ This was a challenge actually. I managed to find a good positive regualtor (ADM7150) that goes to 60dB at 1MHz (I'm only going to 300k). But I couldn't find anything as good for the negative regulator. Best I could find was the TPS7A3301 which starts at 72dB for 10kHz. Thanks for the other info. \$\endgroup\$
    – deki
    Commented Apr 12, 2017 at 7:25

2 Answers 2

Reset to default
3
\$\begingroup\$

What's the likelihood I can use this same formula?

You beat me to the same data sheet I found. More than likely, yes. Manufacturers do not reinvent the wheel unnecessarily, and will use proven designs over and over.

But the TVN 5-2421WI is a dual supply model. And the data sheet has conflicting information if the adjustment can be done, or how it could be done. It says single-output models only in two spots, but also shows a range for the dual output models (±10%). This makes me believe that only one direction of adjustment is available. And since its ±10%, that means only 0.5V or 0.6V. I.E. the positive output can only go up to 5.5V from 5V, and the negative output will match the change equally from -5V to -5.5V. No dropping down below 5V to 4.5V for example.

But since you need ±6V, you likely won't be able to use this for your needs even if the trim works.

\$\endgroup\$
1
  • \$\begingroup\$ Thanks, I just noticed that percentage rating. I'll go ahead and use the formula and if it doesn't work I can always try another resistor value. As for the 6V rating, that was more of an upper limit, I may be able to get by with +/-5.5V or close to it. Thanks. \$\endgroup\$
    – deki
    Commented Apr 12, 2017 at 6:07
1
\$\begingroup\$

Why? Connect a suitable dummy load, a high value potentiometer in rheostat configuration between Trim and Vout+ and start tuning. If you are getting close to the bottom, change the potentiometer to one with lower value. A resistor in series won't hurt too in case it can't take a short between Trim in Vout+.

\$\endgroup\$
2
  • \$\begingroup\$ I'm worried about noise, since this will be used for an application that requires minimal noise. I've already tried to optimise it for as little noise as possible, and don't want to have my output affected by a poteniometer. I've updated my post to include a formula I found. \$\endgroup\$
    – deki
    Commented Apr 12, 2017 at 5:44
  • \$\begingroup\$ You only use your potentiometer to set the value, then replace it with a fixed resistor. \$\endgroup\$
    – winny
    Commented Apr 12, 2017 at 6:37

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.