Widebands That You Would or Wouldn't Buy

[DelSolid]

OK, I am assuming that you referenced the LSU4.2 datasheet rather than the LSU4.9 and if that's true then I am assuming you are referring to the part I bolded below.

4.3.1 Design measures:
- Locate sensor as close to the engine as possible, respecting max. allowed temperature range
- The exhaust pipe in front of the sensor must not contain any pockets, projections, protrusions, edges flex-tubes etc. to avoid accumulation of condensation water. A downwards slope of the pipe is recommended.
- Make sure, that the front hole of the double protection tube does not point against exhaust gas stream.
- Attempt to achieve rapid heating-up of the exhaust pipes in the area in front of the sensor and also of the complete sensor thread boss area, to avoid developing of condensation water
- The sensor thread boss must be designed as shown in 4.9 to reach a rapid heat up of the sensor protection tube area. Make sure, that the protection tube is fully reaching into the exhaust gas stream.

Is that it?

If so then yes, my rig goes against Bosch design specs for OEM installations. I consciously did that because of 3 things:
1. I am using dry test gas so moisture is not an issue
2. I wanted to reduce the test gas volume as low as possible.
3. Different sensor P/N's have different hole configurations in the outer protection tube and I wanted to negate that effect. They are all similar on top.

[DelSolid]

We obviously both posted at the same time.

This device is designed as a desktop "apples to apples" response test rig. I believe the specific number does not hold a lot of importance and the real data is in comparing "Unit A" to "Unit B". All devices tested use the same sensor mounting so whats good for one is good for all. Secondly, the test gas flow is limited to 3 litres/min. That is substantially less than blowing on the sensor with your mouth. The pressure sensor placed in the housing does not even register a pressure rise when the rig is turned on.

My main rig uses a conventional exhaust pipe section with multiple bungs allowing simultaneous sensors to be run concurrently. But the test gas volume is high and the response time becomes more about gas switching efficiency rather than sensor/controller response time measurement. Great for calibration validation but not good for response measurement.

Do you want me to run your rig across this bench and send you the data? Do you want to run mine across your test bench and you post the data here? I am open to whatever.

[DelSolid]

dupe

[toalan]

If during a normal real world applications the end customer will never see 20ms response times, then why even talk about 20ms? The end customer will see response times an order of magnitude slower in a real world installation, 20 ms it is not even in the realm of honesty.

[DelSolid]

If during a normal real world applications the end customer will never see 20ms response times, then why even talk about 20ms? The end customer will see response times an order of magnitude slower in a real world installation, 20 ms it is not even in the realm of honesty.

I completely disagree and believe you are basing your conclusion on incorrect assumptions.

Earlier in this thread another person posted their impression of the unit: (viewtopic.php?p=40813#p40813)

Thanks.
4. It is EXTREMELY fast to respond. Especially given how clean the signal is. The instruction sheet says 20ms. But, it covers the span of 20.9% o2 to Lambda 1.000 in one or two samples ( 100s/s on the CAN). when hit with a spray of CO2. So, the t63 time is more like 10ms. Which is how it looks on the DAC ( 500s/s ). But, the DAC only goes up to ~ Lambda 1.22. So, going by slew rate. This is faster than an LC-1 in "instant mode"

You have to remember that the sensor is not hit with any sort of a pressure wave or high flow, it is literally in a gas flow slower that a human breath.

Like I proposed earlier, do you want me to run your rig across this bench and send you the data? Do you want to run mine across your test bench and you post the data here? I am open to whatever. I think if you played with it you would be surprised just how fast it really is. You obviously have a lot of time with O2 controllers so I would be interested in your feedback.

[toalan]

If during a normal real world applications the end customer will never see 20ms response times, then why even talk about 20ms? The end customer will see response times an order of magnitude slower in a real world installation, 20 ms it is not even in the realm of honesty.

I completely disagree and believe you are basing your conclusion on incorrect assumptions.

Earlier in this thread another person posted their impression of the unit: (viewtopic.php?p=40813#p40813)

Thanks.
4. It is EXTREMELY fast to respond. Especially given how clean the signal is. The instruction sheet says 20ms. But, it covers the span of 20.9% o2 to Lambda 1.000 in one or two samples ( 100s/s on the CAN). when hit with a spray of CO2. So, the t63 time is more like 10ms. Which is how it looks on the DAC ( 500s/s ). But, the DAC only goes up to ~ Lambda 1.22. So, going by slew rate. This is faster than an LC-1 in "instant mode"

You have to remember that the sensor is not hit with any sort of a pressure wave or high flow, it is literally in a gas flow slower that a human breath.

Like I proposed earlier, do you want me to run your rig across this bench and send you the data? Do you want to run mine across your test bench and you post the data here? I am open to whatever. I think if you played with it you would be surprised just how fast it really is. You obviously have a lot of time with O2 controllers so I would be interested in your feedback.

Come on.

That guy hit the tip of the sensor with CO2 not using a jig, not metering anything, does not have a signal to indicate when the CO2 was released so to even begin to calculate a response time. All you can tell from the picture that he posted at best is the slew rate of the DAC on your system, if you see more than that in the picture he posted then tell me.

All you have to do is mount the sensor correctly, and post honest results, that is the easiest thing you can do, the first thing you should have done.

[toalan]

Also if you reference 20ms in your advertisement and or product manual/pages, you should remove it. It is dishonest.

[toalan]

Also you are probably violating the 6-9ma limit into and out of the pump cell as outlined in the Bosch datasheet to get your artificial results.

[toalan]

I am looking at your product manual,

"Response Time typical 20.00 ms"

You have to tell me AEM's definition of "typical".

[DelSolid]

Wow, I go away from from the keyboard for a while and come back to find myself accused of everything except the Lindberg baby kidnapping.

Exceeding the 6-9ma limit? are you just pulling stuff out of the sky now?

I thought offering to send you a unit to test yourself was pretty up-front and fair. Obviously you think otherwise. To each his own I guess.