Because of the holidays, TI E2E™ design support forum responses will be delayed from Dec. 25 through Jan. 2. Thank you for your patience.

  • State Resolved
  • Locked Locked
  • Replies 17 replies
  • Subscribers 49 subscribers
  • Views 1515 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.

LM7171: Circuit Board Material vs Rise Time, Stability,Etc.(FR1 vs. FR4) and LT1166 Stability For Function Generator Output

Fred Doe
Intellectual 655 points
Part Number: LM7171
Other Parts Discussed in Thread: THS3491, LM334, BUF634A, LM6171

Hello!

I am designing a PCB with the LM7171 used in accordance with the Schematic in Figure 7 in the LT1166 datasheet (see below) and I am using it as the output amplifier in  a sine Square triangle (maximum freq. 5MHz) function generator (I am shooting for >50ns rise time on square but i don't think that effects the LM7171 for the purposes of this question) the out put would be a maximum of +-15Vpp.  I have two main questions, (a) would using the LM7171 be appropriate in the first place, and (b) would using FR1 vs FR4(for cost reasons) on the PCB affect The LM7171 in stability, rise time etc.?

Thanks for any help!

p.s. Here is the LT1166 datasheet((https://www.analog.com/media/en/technical-documentation/data-sheets/1166fa.pdf)).

  • 0
    Guru 140200 points

    Hi Fred,

    I have not fully understood what you want to do with the LM7171. Can you post a schematic?

    Concerning FR1 and FR4: If you run your application always at room temperature and at low humidity, FR1 will work. Many years ago I have designed HF-circuits running up to the GHz range with FR1 and it worked very well. But when it comes to multilayer boards with plated through holes, I would always use FR4, because you will get much more reliable boards with tighter tolerances. And if high temperatur changes are to be expected, FR4 (or better) is a must. Also remember that FR1 is very hygroscopic. Happily the price difference between FR1 and FR4 has beome much smaller in the last years.

    Kai

  • 0
    Guru 48395 points

    Well Fred, 

    That LT1166 is a pretty odd part to start from, Figure 7 is a current source output? 

    Anyway, most AWG output use a current feedback part, your speed needs are pretty modest but 15Vpp will probably need at least +/-10V supplies for CFA. 

    Lot's of choices driven by required load drive - are you using a 50ohm output. The latest (way overkill for you) is the THS3491, but there are approx. 100Mhz CFA as well to choose from. 

  • 0 in reply to kai klaas69
    Intellectual 655 points

    I will try to have a schematic done in a day or two if that is okay.  By the way, is it okay to attach an EAGLE schematic file to e2e?

  • 0 in reply to Fred Doe
    Guru 48395 points

    Unlikely we would be able to open an Eagle file, but a schematic snip might be enough - most e2e folks work in TINA. 

  • 0 in reply to Michael Steffes
    Intellectual 655 points

    Yes, I am using a 50 Ohm (ultimately BNC) output.

  • 0 in reply to Michael Steffes
    Intellectual 655 points

    Would doing a screen shot be okay-- or is that what you are referring to with the "schematic snip"? *By the way, I do not have TINA, I only have EAGLE as my schematic drawing program.*

  • 0 in reply to Fred Doe
    Guru 48395 points

    Yes, that is what I meant. 

  • 0 in reply to Michael Steffes
    Intellectual 655 points

    Here is the schematic. I think it is readable, if it is not please let me know.  By the way, I think the LM334 is appropiate but if it is not, it is not a big deal to change it to a different current source/sink; also it is set to ~64 mA but also that would subject to change if that is not the appropriate value. 

  • 0 in reply to Michael Steffes
    Intellectual 655 points

    Are there any CF amplifiers in a DIP with reasonable speed that could drive a 50 Ohm load like you had mentioned?

  • 0 in reply to Fred Doe
    Guru 48395 points

    Well DIPs were largely abandoned above 100MHz due to lead parasitics

    I really can't adapt your circuit to a more typical CFA design - and, what exactly are your design targets. 

  • 0 in reply to Fred Doe
    Guru 140200 points

    Hi Fred,

    hhm, this approach with the LT1166 looks very exotic to me. I would go a different way.

    As Michael already asked, what exactly do you want to develop?

    Kai

  • 0 in reply to kai klaas69
    Intellectual 655 points

    My function generator design has: a +-5Vpp input to output amplifier(as in the waveform generating circuitry has a +-5Vpp output, which goes into the output amplifier); 0.1Hz to 5MHz output; the output is switchable between sine, square, triangle, ramp, pulse and white noise (to 5MHz); 50 Ohm BNC output; a separate TTL level 0.1Hz to 20 MHz oscillator (non-synchronized so I don't think it particularly affects the output amplifier); an adjustable output level from +-50mVpp to +-15Vpp(I am thinking of doing this with a voltage follower and potentiometer between the +-5Vpp output and the input to the output amplifier [advice on this is welcome]); any of the waveform switching circuitry would be done prior to the +-5Vpp output; a desired rise (or fall) time of ~25ns or less (for the waveforms to which it applies); and finally it has a sweep of ~25:1 range, never exceeding 0.1Hz or 5MHz in any case (again,not sure if it matters to the output amplifier particularly, but if it does matter...).  

    Again,any thing available in through hole is better, but it is not a big deal as long as I don't have to worry about hand soldering QFNs, DFNs, or BGAs, i.e. anything with hidden leads.

  • 0 in reply to Michael Steffes
    Intellectual 655 points

    I think the BUF634A will work as an output buffer, used in coordination with the LM7171 in a closed loop circuit; do you think that would work well or not?

    By the way the rise/ fall time is not a critical specification, the circuit would be fine with a somewhat slower rise/fall time; I think the circuit with the BUF634A would probably have a slightly slower rise/fall time like that anyhow.  

  • 0 in reply to Fred Doe
    TI__Mastermind 25625 points

    Hello Fred,

        From my understanding you are looking at an input signal of between +/-500mVpp to +/-15Vpp at 0.1Hz to 5MHz. Which specifications of the LM7171 drew you in choosing this product for your application? You do not seem to need a very high slew rate if you are aiming for >50ns for these voltage levels at unity gain. Also, having the output buffer is significant for increasing output drive which I do not see as necessary for your application. I would suggest looking at amplifiers lower than 100MHz with a minimum of 300V/us slew rate. As Michael mentioned earlier, you would need atleast +-10V supply, and you will have more luck in finding through hole amplifiers at this range. Let me know if I am missing considering an important specification for your application.

    Thank you,

    Sima

  • 0 in reply to Sima Jalaleddine
    Intellectual 655 points

    The specifications that I was particularly interested in on the LM7171 were basically: large current output drive(<100mAfor MOSFET gates in the original circuit); the DIP option; the high speed/slew rate (greater than 10x than the 5Mhz upper frequency limit) and fast settling time; the +-15V supply capability;and the inexpensive price ($3.21 @1pc).  

    By the way, I would not particularly go to a great extreme to get a settling time/output slew rate, but I was interested in the LM7171 because it was inexpensive, fit the above specifications, and I could easily substitute a lower frequency amplifier (like the LM6171) if stability became an issue.  Also,  the whole application is relatively non-critical in specifications.  ( I am designing the whole thing as a homebrew function generator for a home experimenting and designing lab.)

    P.S. Sorry for the delay in replying.

  • 0 in reply to Fred Doe
    TI__Mastermind 25625 points

    Hello Fred,

        I was looking into other products; it does seem like the LM7171 is the only one that fits all your specifications you have listed. If you would be able to go with another package other than the DIP option; there are about 60 devices that could fit your application with half being current feedback amplifiers. Current feedback amplifiers are more traditionally used with function generator designs due to their high slew rate and ability to be used for higher gains. In your case, this might not be too critical, and as Michael already pointed out, it does not seem like it could fit into your schematic. To correct my first reply, you would actually need about 150mA of output current at +-15Vpp. You are correct earlier in needing an output buffer, the BUF634A is a good recommendation. I am unfamiliar with the use of the LT1166. I would have to look into it more, and I will let you know if I find a way you could simulate this design. If you are open to other options, there are plenty of resources for creating function generators: TIDA-00684, TIDA-00075, AWG System Design.

    Thank you,

    Sima

  • 0 in reply to Sima Jalaleddine
    Intellectual 655 points

    Thanks for the feedback! I will look at the links you provided (I haven't had time yet).  By the way, I appreciate any effort regarding the LT1166, but, as I have generally mentioned in previous posts I am flexible with the output amplifier, so if it would be too much trouble about the LT1166, please don't feel that you have to go to great lengths on it.  Again, I appreciate any effort you put into it!