Analog Input with DAQ - inpedance problem

[notimperial]

---

@SnowMule wrote:

The input impedance of the 6008/6009 is about 140kΩ.  That low impedance just soaks all the current the little photodector can pump out. 

I built a voltage follower out of an LM358 and piped that into the DAQ's input... works real well.  Opamp has a real high input impedance and provides enough current that the daq's low input won't cause it to crowbar.

---

Thanks for your reply! I am going to follow your advice.

Some posts ago, Henrik Volkers said I should take a look at aliasing. Could you comment on that?

Also, I know now this is getting a bit off-topic but I am a noob and I don't have a power supply available at the lab. Do you know if I can power the LM358 using a battery, which is easier to buy? 

[K C]

The Thorlab can drive 10V into 50Ohm !! (2W!!) So the input resistance of the 6009 of 140KOhm is no problem.

I don't think it is necesarry to add any hardware.

See the spec sheet for output configurations.

Kees

[notimperial]

---

@K C wrote:

The Thorlab can drive 10V into 50Ohm !! (2W!!) So the input resistance of the 6009 of 140KOhm is no problem.

I don't think it is necesarry to add any hardware.

See the spec sheet for output configurations.

Kees

---

The spec sheet of the photodetector says that after a 50ohm coaxial cable (which I am using) there should be a 50 ohm terminating resistance.

If the input impedance of the DAQ card is 144kohm, shouldn't I connect a 50ohm between its terminals? So that 144k//50~50 ohm.

[K C]

Yes that is correct, but the spec tells you more.

If you want to use the full bandwidth you should use a 50 Ohm termination resistor at the DAQ input. But this will also empty your battery quickly. Beside that the resistor must be able to dissipate 2 Watts if the output is at maximum.

If you do not need the full bandwidth you can use a higher resistance. This will result in a mismatch but at a lower frequency this is no problem.

Read this section of the spec. !!

[notimperial]

---

@K C wrote:

Yes that is correct, but the spec tells you more.

If you want to use the full bandwidth you should use a 50 Ohm termination resistor at the DAQ input. But this will also empty your battery quickly. Beside that the resistor must be able to dissipate 2 Watts if the output is at maximum.

If you do not need the full bandwidth you can use a higher resistance. This will result in a mismatch but at a lower frequency this is no problem.

Read this section of the spec. !!

---

Thank you for your advice.

With a 470ohm resistor I was still getting very low signals so I decided to use a voltage follower, as other users suggested. I built one with a LM741, which now causes the output to saturate at 10V - the maximum voltage the daq card can read 😞

Do you have any suggestion about what I can do next?

[Henrik_Volkers]

741 😮 ... You have a sport car engine (PD) and put it in bobby car (poor 741 OP) with drifting, offset voltages etc.  (At least the low bandwidth of the OP match the 6009 😉 )

So back to start:

First check the power supply of the PD!

1)You want to use that PD to monitor what? 

2)How fast (and in what range) do you expect your signal of interest will change?

3)You said you have seen a signal at the scope: How was the setting of the scope?

4)Did you use the 1M input of the scope?

(Some scopes offer to switch the input impedance to 50Ohm)

5)How does the signal (using what light source?) at the scope look like when you terminate the cable at the scope with 50, 470, 1k Ohm?

As KC noted , I also claim that you don't need a high input impedance for that PD.

Depending on what you want to measure (how strong and how fast the light to the PD changes, what resolution you need) I would add a simple cap as an aliasing filter. Suggested values after you answer at least Q2 and Q5.

[notimperial]

---

@Henrik Volkers wrote:

741 😮 ... You have a sport car engine (PD) and put it in bobby car (poor 741 OP) with drifting, offset voltages etc.  (At least the low bandwidth of the OP match the 6009 😉 )

So back to start:

First check the power supply of the PD!

1)You want to use that PD to monitor what? 

2)How fast (and in what range) do you expect your signal of interest will change?

3)You said you have seen a signal at the scope: How was the setting of the scope?

4)Did you use the 1M input of the scope?

(Some scopes offer to switch the input impedance to 50Ohm)

5)How does the signal (using what light source?) at the scope look like when you terminate the cable at the scope with 50, 470, 1k Ohm?

As KC noted , I also claim that you don't need a high input impedance for that PD.

Depending on what you want to measure (how strong and how fast the light to the PD changes, what resolution you need) I would add a simple cap as an aliasing filter. Suggested values after you answer at least Q2 and Q5.

---

Hey again!

1) light output of an optical fibre

2) I'm plotting autocorrelation curves of laser light.  For that I'm planning to use about 50kS for 2 sec.

3) not sure this is what you mean, but I just had the PD connected to the scope via a 50ohm coax

4) at 1Mohm input impedance I could see a signal, when I switched to 50 the amplitude was lower (obviously :p) but I could still see sth

5) is this achieved by placing resistors with these values across the terminals at wich the scope is connected?

[johnsold · Accepted solution]

2) 50 kS/s is faster than the maximum rate of the USB-6009 (48 kS/s).

4) The input impedance of the 6009 is 140 kilohms. This is slightly lower than the scope impedance.  Since the photodetector works OK with the 1 megohm scope, it will work OK with the 6009.  You do not need the resistor.  Try connecting the scope and the 6009 at the same time via a Tee connector and see if the scope signal changes when the 6009 is connected or disconnected.  I think it will not change.

Lynn