The 0.0000 output only came about if the ADC input was shorted to ground. The other values are from a shorted op amp input and seeing the minimum output that the op amp could achieve.
-
Welcome back Guest! Did you know you can mentor other members here at H-M? If not, please check out our Relaunch of Hobby Machinist Mentoring Program!
Needing more than a spark test?
- Thread starter graham-xrf
- Start date
Yes.
First question is, do I have the driver part number correct? AD8655 does not seem to be in the recent updated LTSpice database.
Next question is, is it operating between 0V and +Vdd(something), single ended?
Regardless, I will put somethiing else in there if the update does not suipply a model.
On a different subject, my second set of 8 sources arrived today. Also, I'm happy to report that ABS machines like a dream.
Yes. AD8655, that is what Mark used. It is running off AVDD which is roughly 5V and Vss is connected to GND.
Can you use an AD8021 instead? It's wider BW and probably more expensive though. It does operate over a wider supply range than the AD8655. But it has higher offsets. In 1K quantities it is 50% more costly. $1.50 vs $1.00.
The AD8021 definitely is more expensive. I checked the price today and it's around $12 for one. Could explain why it's in stock
.ADI also mentions the AD8610 as a input buffer. Its GBW is similar to the 8655 and has a higher power supply voltage range. It's a bit less money.
@WobblyHand :
Hi Bruce
OK - it works, and I have made a discovery about negative supplies on the way.
Also - Woah! Too fast. I already simulated using AD1807 (differential drivers for AD7622).
You need to press ESC a couple of times to stop it micro-stepping. Also, I did a .OP pass first.
The experiment has to be changing the negative supply. At -0.5V, things messed up, I think because the integrator needed it better. I tried -1V. and then -2.5V. OK change to AD8021. It does not matter that it has higher offsets. We are driving them out anyway. The one attached is a disappointment. Offset is 397nV.
Hi Bruce
OK - it works, and I have made a discovery about negative supplies on the way.
Also - Woah! Too fast. I already simulated using AD1807 (differential drivers for AD7622).
You need to press ESC a couple of times to stop it micro-stepping. Also, I did a .OP pass first.
The experiment has to be changing the negative supply. At -0.5V, things messed up, I think because the integrator needed it better. I tried -1V. and then -2.5V. OK change to AD8021. It does not matter that it has higher offsets. We are driving them out anyway. The one attached is a disappointment. Offset is 397nV.
The AD8610 has offset adjust pins - so that tells you something. About +/- 3mV of trimming. It's not that cheap.The AD8021 definitely is more expensive. I checked the price today and it's around $12 for one. Could explain why it's in stock
ADI also mentions the AD8610 as a input buffer. Its GBW is similar to the 8655 and has a higher power supply voltage range. It's a bit less money.
Too early to claim victory. Simulate out to 1msec or 10msec and tell me if it is still good. When I simulated the integrator in the loop adjuster, it was ok short term, but not so nice longer term. Integrators are not always well behaved over temperature and voltage. Maybe your experience is different, but I've been burned more than once...
AD8021 is not in the database anyway - nor is AD8610.
So - still using my LT1807 driver.
OK - It's still working fine, but with an improvement.
Moving the integrator output to drive the U4 inverting input, instead of the stage before, reduces the offset to -2.87nV.
In a test variation I disconnected the U2 output, and I put a 0V short at R4 input, and I let the simulation run to the end.
I get the startup transient, and it flatlines at -39pV !
Directly shorting the U4 non-inverting input yields -32pV at the driver output.
I defy the ADC to even notice that!
Now going for longer simulations
So - still using my LT1807 driver.
OK - It's still working fine, but with an improvement.
Moving the integrator output to drive the U4 inverting input, instead of the stage before, reduces the offset to -2.87nV.
In a test variation I disconnected the U2 output, and I put a 0V short at R4 input, and I let the simulation run to the end.
I get the startup transient, and it flatlines at -39pV !
Directly shorting the U4 non-inverting input yields -32pV at the driver output.
I defy the ADC to even notice that!
Now going for longer simulations