• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 49 subscribers
  • Views 641 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

PGA870 Prototyping

Jonathan Donovan
Prodigy 20 points
Other Parts Discussed in Thread: PGA870, LMV7239, LMH6702

I am working with a team of engineers to design and build a time-domain reflectometer. We need to input a signal into the base of a bjt with a range of 0 to 5V. We want to use the PGA870 to control the gain on an input signal to the bjt. We decided to use the PGA870 because it has a large bandwidth and a large range of gain steps (B0 to B5). However, when prototyping it, we were getting very erratic results. We setup the circuit like the schematic on page 18 of the datasheet (single end input to differential output).  When measuring the differential across the Out+ and Out- pins, we obtained at most -200mV difference. The circuit also drew around 150mA, which seems like a lot. To change the gain, we simply set gain strobe and latch mode high. The gain should then change instantaneously. We also tried toggling gain strobe high and low to set the gain. However, we weren't able to control the output with a range of 0-5V. We didn't place any resistors on the gain pins. Do we need to place resistors on those pins to limit the current into the gain pins? Could this be why there is so much current being drawn? We are also applying 5V to the gain pins to set them high. Is this okay?

  • 0
    TI__Genius 13935 points

    Hello Jonathan,

    Can you provide a complete schematic?  I do not know what you mean by a 0-5V output range - is this the actual DC voltage?  What are the emitter and collector of the BJT biased to?

    The 150mA is the typical supply current for the PGA870, so this is normal.  The PGA870 outputs are differential, but their individual output voltages can only range from 1.3V to 3.7V.  The two outputs change out of phase from each other, to create a larger differential swing, but the DC level they operate at does not go from 0V to 5V.  

    If you want to drive a 0V to 5V signal you will either need a rail to rail amplifier (will still not achieve 0V or 5V) or you will need an amplifier that uses a supply voltage that is more than 5V.  

    Do you really need a differential amplifier?  Do you really need variable gain?  

    Regards,

    Loren

  • 0 in reply to Loren Siebert 1
    Prodigy 20 points
    Hello Loren,

    We want to generate a 50ns pulse through a DC blocking capacitor on the collector side of the bjt and control its amplitude between 0 and -60V. 60V is sent into the collector (rail) of the bjt. The bjt is set up as a common emitter amplifier and will amplify the signal into the base of the bjt with negative gain. This base signal should be the output of the PGA870. By controlling the gain of this signal, we can control the amplitude of the pulse we generate. When the bjt is off (the base signal is below the saturated voltage necessary to "turn on" the bjt), the dc blocking capacitor blocks the 60V. When we turn on the bjt by outputting a signal from the PGA into the base of the bjt, a pulse will be generated based on its amplitude. This is why we need variable gain. I am not sure however, if we need differential output. The base signal should be DC, but the PGA seems to be outputting AC voltage (the voltage is fluctuating). We also don't need 0-5V, but rather a signal low enough to put the bjt in cutoff mode (<0.7) up to around 2.5V for a full swing in the pulse. Do you not think the PGA870 can be used for this application? Thank you for your help.

    Regards,
    Jonathan
  • 0 in reply to Jonathan Donovan
    TI__Genius 13935 points

    Hello Jonathan,

    I am not familiar with that application, however it does not appear to me that you require a differential output.  

    I suppose you could use the varialbe gain to change a fixed DC voltage from <0.7V to 2.5V.  

    The issue you are seeing with the output voltage fluctuating is comon with AC coupled signals.  You need to find a way to provide a reference voltage on the BJT side of the blocking capacitor.  The pulse generagted by the DVGA would then ride on this DC reference voltage. 

    One potential solution would be usa a high speed comparator (LMV7239) (http://www.ti.com/lsds/ti/amplifiers-linear/comparator-products.page )  and then buffer it (if necessary) with an amplifier like the LMH6702 which is a very fast single ended op amp.   This would be basically a single bit DAC that would convert one logic bit into the pulse  you desire.  You will still need to establish the proper DC voltage on the other side of the blocking capacitor no matter how you create the pulse. 

    If I were implementing this circuit I would not use the PGA870. 

    Regards,

    Loren