First Experience - QuadTech LR2000 milliohm Tester
For those of you that aren't familiar with my writing style, this article is what I call my first experiences. This does not mean this product is brand new on the market, but rather the first time I've personally had an integration application using this product. I will also share my learning curve ratings at the end of my first experience articles.
Have you ever milliohm tested anything? Looks like something that would be pretty simple, find a tester, integrate your tester to your PLC, done. Sometimes things can be deceiving unless you look at your application really close.
My boss turned me on to this milliohm tester from Quadtech. We purchased the LR2000 milliohm tester in hopes it would be a good choice for our application after finding this unit in a Google search. Our application was to measure the weld connection in two places on a part that we automated. Of course, these units can only measure one connection at a time like your Fluke meter would, unless I came up with a switching circuit or other means of automating the concept of measuring two different connections. Nonetheless, here is the low down on this tester.
When I decided to order this tester, I obtained the manual before the unit arrived. I always do this and study how to integrate a PLC to a unit and handshake some good/bad signals with any new unit I integrate. Again one of my worst pet peeves, you have to give your whole life history just to download the manual. I can understand that some companies want to generate sales leads from their websites, but I just don't think this should be one of those ways in my opinion. So I asked the salesman to e-mail the manual to me directly.
My customer also wanted to connect a laptop to a tester of choice and log some SPC data. Upon calling QuadTech and talking about my PLC integration and customer requirements, they told me they had some free software I could download from their website which would allow this. Without much further investigation, I took their word it would function as they said. Later I found out some issues using the software that I will share.
My first objective was to figure out how to handshake with the unit using PLC digital signals. I guess normally QuadTech is use to other engineers connecting through the IEEE interface and writing up some snazzy Visual Basic application to handle all the signals required. Little does QuadTech understand that PLC people just need a simple dry contact, tell us a 1 or 0? Pretty please?
When I looked at the interface diagrams in the manual, I was confused. Not as simple as an Associated Research tester. If you want to interface a PLC with this tester, don't forget to purchase the "Handler Interface" option. This is about a $545.00 adder option wow. Also another word of caution, you don't get an IEEE cable with your PLC "Handler Interface". You only get the IEEE jack on the back of your tester. Luckily I had enough time in my project to locate an IEEE cable from cablestogo.com. Of course all I could find was an extension type cable, but we just cut off one end of the cable and made the interface cable a flying lead type to tie into the PLC.
For the PLC connection I used the "Compare" mode and wiring method as described in the manual in Table 3-8. I connected my own 24VDC to pin 8 and 10. The results from the tester turned on pin 21 (passed) or pin 15 (failed). These outputs from the tester are NPN outputs. So I had to use small MURR terminal style relays to convert the NPN signal from the tester to PNP for my PLC, since my machine was designed with PNP I/O in mind. Of course in normal order was the trigger signal. All the unit needed was a dry contact closure on pins 1 and 5 to start the test. I also connected a relay to the "End of Test" output so the PLC knew when the tester was finished. Because of the not so user-friendly outputs, you'll need to supplies relays to isolate your PLC inputs. Unless your designing an NPN input type PLC system where you could possibly drive your input PLC signal direct with this output from the tester.
So what about the tester itself? The front panel menus were amazingly simple to use. I breezed through the setup process in a matter of minutes. Just toggling through the menus I was able to select the options I needed to set my limits that I was looking for and the tester was responding with pass or fail outputs. Not bad to get the unit up and running before the day's first Mountain Dew break. Where I had some problems was milliohm testing in general. This was my first time trying.
I already knew that I was measuring a value of about 14 milliohms, which is like 1 foot of 22AWG wire. See this handy wire resistance chart here. I needed to measure two contact points in my part so I thought I would be smart and design a switching circuit to switch the test probe leads to test point #1, trigger, wait for a pass or fail, then change the switching circuit to test point #2, trigger, wait for the second pass or fail. I even told my assembly team that the wiring harness had to be exactly the same length on both sides of the test points, and to use very heavy gauge wire, about 12AWG, to cut down on resistance from my test leads. Everything worked like a champ, except the relay contacts in the switching circuit. I used a very high end Magnacraft relay that measures 2 milliohms across the dry contacts, but sometimes it would be 3 or 4 milliohms. Granted this is still an awesome connection, but my pass and fail parameters were +/- 3 milliohms. With the relays fluctuating a different contact resistance every time they opened and closed, the switching circuit idea wasn't going to work. I tried different relays only to find some relays had a much worst dry contact connection around 200 milliohms! So needless to say, I bought a second tester and dedicated one test point per tester. This was the best solution.
Once we decided to have two testers, we still had test probe resistance to deal with. Luckily the tester supports a test lead zeroing function. Now with the test leads always being consistence and not going through a switching circuit, it was easy to zero the leads and probes on the tester then getting an accurate measurement.
Ok, now with the application working correctly, I had to go back to my customer's request. My customer wanted to log SPC data from this tester as the machine was running. He only wanted to do this every few days by walking up to the unit and connecting with a laptop and start logging for about fifteen minutes then disconnect. I download the Virtual Front Panel Wizard software as per recommended by Quadtech. The software does log data correctly, but when you connect using your laptop the software takes over control of the tester, and well, frankly, nothing else works. You can't press any buttons on the front panel and worse, your PLC handler interface stops responding to the PLC. There is no way to trigger the unit from PLC or read outputs to the PLC while the laptop is connected. So much for "real time" logging of SPC data. I don't think Quadtech understood what I was asking them in that first phone call. It was too late in the game to change testers so I scrambled to try and figure out what could be done. Upon digging into the tester, I started talking to the tester through the RS-232 port using HyperTerminal and thought I was on to something. I was going in a direction to connect the tester to my DF1 (RS-232) port on my PLC and drive the entire tester using ASCII functions in my PLC program.
Using a standard DB-9 cable you can dump commands to the tester and get responses back within HyperTerminal.
Here is a list of commands I had fun with:
Needless to say I couldn't come up with anything and ran into a roadblock with some commands in the LR2000 manual I needed that were not support in the Handler Interface side, but were supported in the IEEE side. I had to break the news to the customer that logs could not be done when the machine was running, but rather one part at a time. My customer being the nice guy he was said that was ok and let me off the hook.
Even after all that there was still another problem. Once you close the software and unplug the RS-232 cable, the front panel was still locked up or locked you out. I called QuadTech and they said there was a secret handshake to unlock the panel or release it back to you. At the time of this writing, I forgot what key sequence you press to unlock the panel. If I figure it out again, I'll revise the article and let you know or you can call QuadTech as ask them if you run into this problem yourself.
In summary, the tester is easy to use, setup and did a great job measuring milliohms like my application required. I had some issues with the handler interface being only offered in NPN, and I felt I was originally mis-lead by Quadtech about what their "free" software could do. I also had problems with the software not releasing the front panel back to the operator once the tester was connected via a laptop with Quadtech's Virtual Front Panel Wizard. You'll need to call Quadtech and ask them what the secret button sequence is on the front panel to release the front panel back to you and enable the handler interface again after connecting the laptop. Keep in your mind what you are doing when milliohm testing. Understand that 1 foot of 22AWG wire will measure 16 milliohms, that's just the BARE wire from end to end. Milliohm testing is a great solution to verify weld connections and could tell when a weld was good or bad by measuring the milliohm value.
Learning curve rating:
Curve 0 = Walk in the park.
Curve 10 = Get out the scholastic cap and crash in the classroom.