This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TL026 in single-supply application

Intellectual 270 points

Replies: 3

Views: 1059

I'm looking at using the TL026 in a single-supply application. As far as I can tell, the part really only requires mid-point reference for input biasing; output coupling is AC, and the AGC input is referenced to REF OUT. Further, IN+ and IN- are relatively high-impedance pins (> 10kohm) so the main impedance concern for the input bias network is the input resistor.

So - I'm thinking of using two TL026s in an IF strip (or perhaps just one TL026, depending on noise), using a simple resistive divider, capacitively-decoupled to system ground (negative supply rail).  REF OUT would also be referenced to ground through a resistive divider to generate the ~7.2V for the AGC circuit.

Any reason why this shouldn't work?

Thanks -

Dana

3 Replies

  • I am not familiar with this part, but I reviewed the datasheet and I think that if you are careful with the details you should be able to make this work on a single supply as you have described. 

    The inputs need to stay within 1V of the mid supply point so the voltage dividers should be accurate.  This part has pretty high maximum gain, so input biasing voltage is important and matching between the two inputs will be pretty important.  You will want fairly low impedance on the voltage dividers so that bias currents and any asymmetry in the signal does not cause it to drift. 

    Having modified a number of eval boards for single supply operation I know it ends being a bit of a compromise compared to using split supplies.  Is it the cost of a negative supply that is the prime consideration?  We have switching supplies that can generate a negative supply rail with only a few passive components.  I am actually in the process of designing a split supply evaluation board for one of our single supply parts.  I personally like the LMR70503 because it uses only 7 passive components.  The LM2611 gives a lot more flexibility but it needs two inductors (9 passives total). 

     

  • In reply to Loren Siebert 1:

    Quoting:

    The inputs need to stay within 1V of the mid supply point so the voltage dividers should be accurate.  This part has pretty high maximum gain, so input biasing voltage is important and matching between the two inputs will be pretty important.  You will want fairly low impedance on the voltage dividers so that bias currents and any asymmetry in the signal does not cause it to drift. 

    Thanks. I actually plan to use the part in a single-ended configuration, meaning the input signal would only be applied to one of the input pins, with identical input-bias resistors to the mid-point bias. The output is capacitively-coupled so small amounts of DC drift aren't important.

    Is it the cost of a negative supply that is the prime consideration?  We have switching supplies that can generate a negative supply rail with only a few passive components.

    Cost is not the prime consideration, and I contemplated the addition of a negative supply as you describe, however, this is in a sensitive radio receiver and I'm concerned about RFI generation in a switching supply, but I'll take a further look.

    Thanks -

    Dana

  • In reply to Dana Myers16512:

    Hi Dana,

    We have had good luck on our ADC boards by using a switcher followed by an LDO and some additional filtering.  I talked to a power apps guy who claimed  you could filter the switcher well enough to skip the LDO, but I have not had a chance to test this.  

     

     

     

     

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.