Fan speed Measurements...

Thread Starter

dmusoke

Joined Jan 4, 2010
11
SgtWookie:

I'm trying to build a hopefully simple tachometer to gauge fan speeds ranging from 0 to 30,000 RPMs. I am using the LM2907-8 with a desired output voltage of 0 to 10 volts (into my ADCs) and I've attached my schematic for your comments, if possible, along with the datasheet for the 28V fans.

I have tried to find a spice model for the 2907 to use pSpice in OrCAD but can't seem to find any. Would you know where i could find one?
There was one model that had a geocities link address that is no longer valid as it seems yahoo has closed all goecities accounts.

I added C24 as to remove a perceived DC offset based on what I interpreted from the fan datasheet where it said that its tachometer amplitude output is 5V ± 1v. I wasn't sure if they meant the amplitute is 5V with an offset of ±1v or the amplitude goes from 0-4volts to 0-6volts. If the latter, then I can remove R34 & C24. Can the 2907 work reliably with ground referenced devices? I am concerned about the comparator NOT switching if my input signal is within its voltage offset range.

Your help and comments are most welcome:)!

Thanks & Regards,
David Musoke
 

Attachments

SgtWookie

Joined Jul 17, 2007
22,230
You are in luck.

I happened to find the model awhile back, and saved a copy of it, plus the image that he'd posted of how the model worked.

I haven't figured out how to get the thing working in LTSpice, though. :rolleyes:

Here's the image:


The model is attached as a .txt file.

This note is in the model:
*NOTE: THE REAL DEVICE OUTPUT ON PIN 3 SATURATES AT ABOUT VCC/2
*THIS MODEL OUTPUT ON PIN 3 HAS NO LIMIT ALTHO IT WILL BECOME
* NONLINEAR IF THE UPPER FREQUENCY IS EXCEEDED

I don't have any of these devices, so I can't test them to verify what he's implying. It doesn't sound quite right to me, and that's not explicitly stated in the datasheets that I've found for it. There might be a limitation in the real device due to the charge pump - but I just don't know.

It also looks like he might've reversed pins 11 and 1 on the .SUBCKT line; either that or his notes are backwards in the comment line below it.
 

Attachments

Last edited:

someonesdad

Joined Jul 7, 2009
1,583
I can't help with the circuit, but I'll pass on a tip -- it might be useful to other folks. When I was a student, we used the General Radio 1531 stroboscopes to study motion with Polaroid cameras. Those strobes were also useful for stopping motion of rotating objects.

If you're working at your bench and you need a quickie stroboscope, consider hooking up and LED to a function generator or a 555 circuit set up to oscillate at a variable frequency. It is fast to set up and works fine, especially if you dim the room lights. I used this a while back to figure out which fan was making noise inside my computer -- it was obvious once the strobed light was on the offending fan, as I could see the hesitation in the blades (it was probably a bad bearing). I used a white LED, but of course a monochromatic LED would work fine too.
 

Thread Starter

dmusoke

Joined Jan 4, 2010
11
Thanks SgtWookie ...now to the task of getting it to work in pSpice.

Do you know whether the comparator in these devices switch reliably with ground referenced sources, in my case a variable frequency square wave from 0 - 4 to 0 -6 volts?
 

SgtWookie

Joined Jul 17, 2007
22,230
Look in the datasheet.

The -8 devices are specifically rated for inputs that traverse ground. Their inputs are protected against such traverses; the noninverting input to the comparator is grounded.

The non-8 devices require that the inputs to the comparator stay within the power rails, or it will likely be damaged.

The differences in the internal circuitry are documented in the Equivalent Schematic Diagram on page 18 of National Semiconductor's datasheet. Be sure to download the latest version from National's site. The one I have is from December 9, 2008
 
Top