woodward 2301 gov on electronic diesel 1

[ivymike]

hi, I'm wondering whether anyone has experience configuring a woodward 2301 governor to work w/ an electronically governed diesel engine (Caterpillar C280). I've got a C280 recently added in with several 3616s sharing load in a stationary powergen application. Each has its own 2301 governor, and I'm trying to get the gains set correctly for the electronic engine. I've actually never messed with a 2301 governor before, so this is new to me. Planning to use single-point gain maps to get a feel for what settings give startup/no-load stability, with the gain ratio set to 1, then move those settings to the lowest point on a 5pt gain map, populate the rest of the map, and try to to take load. I gather that it's typical to have gain higher at the higher load points, perhaps dropping back a bit at 100% load. I'm not so sure what values of "actuator compensation" and "reset" are appropriate for an electronic engine. Seems to be most stable with very high actuator compensation, but some feedback I got was "if that was a mechanical actuator, it would have broken hours ago- it would be buzzing back and forth." Very low "reset" also seems to help.

Trouble is that I've so far had trouble getting gain high enough to avoid overshoot on startup while still getting stable idle operation... perhaps the gain ratio can help me there? "rpm window" setting of 17rpm currently...

any pointers/experiences would be appreciated.

 

[catserveng]

I'm assuming you're using 2301D (digital) governors since an analog 2301A wouldn't have the correct output to bias the CAT's ECM.

Most of my experience on 3600 engines in power gen is using Woodward 721 or 723 governors and DSLC's, only site I did work on 2301A LS equipped 3600's was in Florida at a wastewater plant and I wasn't overly happy with end results.

Only 2301D's I've done are on 3500 and 3400 mechancial engines, both gas and diesel.

Using a 2301D over an engine's electronic governor wouldn't be my choice for doing a load share system, an LSM (Load Share Module) is the product Woodwards uses for that, have done a lot of mixed mechanical and electronic engine projects with 2301A's on the mechanical units and LSM's on the electroninc engines.

my contact info is on my website,

http://www.pactectric.com,

this is more than I can help you with over the forum but if you want to drop me an email with some more details I may be able to help. You can also try Jeff Rongish at Woodward in Fort Collins, he's about the best guy for that product line.

Mike L.

 

[ivymike]

it's actually an analog one, and we're using the pwm converter from Cat to get from 0-200ma to pwm input into direct rack. The 2301 is exactly the same as is controlling the mech engines, we're just using the pwm converter into direct rack in place of the rack actuator. naturally the dynamics of the control input system are completely different (no momentum that I can imagine, even though there's a tiny bit of a response lag built in somewhere).

The engine itself is also a bit different than its neighbors - same displacement and inertia, but different turbos and injectors.

The 2301 at each engine gets a speed bias from a DSLC upstairs.

I'll have to ask around about the LSM, and see what the options might be for getting one to site. Remote location, so everything is in slo-mo.

 

[catserveng]

OK, so the 2301A's are speed controls being controlled by DSLC's, right? Then this is easy, use a DSLC with PWM output direct to the engine ECM and don't use the 2301, just adding ao lot of extra problems you don't need.

9905-373 is DSLC part number for open delta pt's, can't remember the one for wye pt's.

Sounds like you have a lash up destined for failure, a DSLC to an analog governor to an analog to digital convertor to an embedded electronic governor.

The LSM is an option if you're using 2301A load share governors that use the nominal 3VDC load share lines. If you're using DSLC's then all you need is the correct DSLC.

With the electronic engine you may have to reduce the loadshare gain in the DSLC down from default of .70 to around .35 (this settings works for most CATs I've done).

Mike L.

 

[ivymike]

thanks for that, I'll see what I can do about it

 

[catserveng]

One other possible solution, although it has not been a very good in my experience, is to use your existing DSLC with the voltage output and use an Axiomatic or Pheonix signal convertor. I used both but have never really been happy with the response.

I can dig up the part numbers if that is the way your customer wants to go, but your best solution is the correct DSLC.

Mike L.

 

[ivymike]

we're trying to get a guy to buy a different DSLC and fly here with it tomorrow. trying to look up the spec for him for the ECM speed bias input signal...

 

[catserveng]

A current C280 should have an ADEM 3 ECM, and the speed bias should be selectable from the configuration menu in CAT ET from 500Hz PWM, 4-20mA or 1-5VDC.

There is also a new Woodward 9905-373 DSLC with PWM output for open delta PT's on eBay, had a project where the indicated he wanted to move forward then disappeared, so I was looking. If I had the spare money I'd buy it myself since the older DSLC will likely go out of production next year and I know of plants that have them.

Good luck!

 

[ivymike]

It turned out to be fairly straightforward to get it working w/the 2301A and the older DSLC, once we figured out that a couple of our speed sensors were not providing a good signal (oops). The gains were nearly identical to those for the mechanical engine.