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.
Hi,
I am currently working on a project which requires sending some very high frequency (up to 1GHz bandwidth) signals down a very (very!) small cable. The cable in question is a 150um co-ax cable which as you can probably imagine is not brilliant, but it's what we need to use.
Anyway, long story short I inherited some of the work from someone else, and as far as I was aware the cable was a 50Ohm coax which made finding amplifiers not too difficult. I settled on the LMH3401 for its low external part count - we need 12 of them in a small space, so minimal components is key (we also can't use ferrite material, so balun transformers are out of the question!). I have one of the EVMs and all looks pretty nice. So anyway I went and hooked up the micro co-ax cable and was amazed at how terrible the signal reflections I was seeing (a 5ns square pulse became two!). After much head scratching and some fiddling with a network analyser it turns out that the cable is actually around 90Ohm! After consulting with the manufacturer they specify it as being 75Ohm co-ax, so I'll go with that for the time being and do some testing on another sample when it comes in (I think the cable may have previously been kinked changing its impedance).
So having discovered that it causes a bit of an issue trying to find an amplifier which is suitable to do the conversion required - a different co-ax is not an option, a 50Ohm sample that we got later is terrible (it's about -15dB at 1GHz!!). Building a matching circuit over the 1MHz to 1GHz range needed is probably pretty much impossible. Essentially we are stuck with using the 75Ohm stuff and now I need to find an amplifier which will take that (single ended) and convert it to a 100Ohm differential signal with a minimal part count and without using any ferrite material.
I've seen that the LMH3401 has a close cousin in the LMH5401, the difference being that the latter uses external gain, feedback and termination resistors rather than internal, so I am hoping this will allow the input impedance to adjusted (I hope!). The trouble is I am having a hard time trying to figure out how I should go about selecting these resistors for a given input impedance.
Basically we want the full amplification possible which if I am not mistaken should be 12dB with a 200 Ohm output load (100 Ohm differential with the 40 Ohm series resistors). The input impedance needs to be 75 Ohm. So we will be setting it up as in Figure 65 of the datasheet.
So the question is, how do I go about calculating the resistor networks for this requirement?
Thanks.
Hello Thomas,
The preferred approach for picking Rf is to establish a maximum gain (Avmax) for the single to differential stage. As you have noted, the key is to select the desired gain (Av) less than the Avmax.
I would recommend picking Rf larger than 200ohms, because with Rs of 75ohms the desired gain (4V/V) that you are targeting becomes equal to the maximum gain with Rf = 200ohms. Picking Rf = 250ohms should give you the adequate BW response. I have attached the below circuit with Rf = 250 ohms.
Best Regards,
Rohit