Adding Spindle Probe to HMC. Blum or Renishaw?

What would you guys go with?
I have the OMP40 on my 2 5 axis machines. Works for what I use it for. I have a bunch of production jobs I'm trying to move to the horizontal and it just got out of warranty so Mori can't limit me who I can use to add the probe now.
Wish I would have bought it with the machine as it has sucked without having one.

IMO, I would just go Renishaw.

Why? because I am familiar with them and the support is good.

Have you priced out the cost difference or any difference in features?

m98custom1212 said:

IMO, I would just go Renishaw.

Why? because I am familiar with them and the support is good.

Have you priced out the cost difference or any difference in features?

Click to expand...

Just starting to research. Haven't received any price quotes yet.

Side note: Did you ever get the probing in NX to work with your posts? That would be a huge time saver I'm sure. Right now for my Siemens controlled machine I hand write all the probing :/

dstryr said:

Just starting to research. Haven't received any price quotes yet.

Side note: Did you ever get the probing in NX to work with your posts? That would be a huge time saver I'm sure. Right now for my Siemens controlled machine I hand write all the probing :/

Click to expand...

Yes, I did not easy but I can help you if you need any.

Personally, I would stick with Remishaw - just get another OMP40 for the horizontal, and call it good. Support from the Renishaw Chicago office is great. And the macros are really pretty easy to use after a brief reading of the manual.

If you can find an old Renishaw probe, they will take it in trade and give you half price. I think part of this is that you must deal with them direct, so you are cutting out the MTB. I saved $5k by trading in an ebay probe I bought just for the purpose.

I prefer BLUM. The reason is one-hit probing. I'm probing at 2500 mm/min (98"/min), single hit (Brother control) and it has a Cp over 12 at that speed. Total recorded variation against a cast aluminum surface was <30 microns (0.001") over 750 measured parts. Most of that variation can be attributed to the fact they were measured on a Zeiss Calypso Scanning head CMM and where the CMM hit may not have been where I hit on the as-cast surface with the probe. It was also cheaper by about $1k at the time I bought the first one. I'm using the TC62 which is more expensive because it's radio, but the radio is worth it because you don't have to have the receiver in the messy zone.

I just bought a used blum tool setter on ebay. When it came it had gotten coolant in it so I have my doubts as to weather it is trash or not. It is the Z-nano and is still made. The part that should be replaced is a pc board about 30mm in diameter. Otherwise the thing looks fine inside. I called blum and they said new price on the probe is about $1900 and it will cost more than that to replace the pc board. Hopefully Renishaw would support their product better than this.

Tonytn36 said:

I prefer BLUM. The reason is one-hit probing. I'm probing at 2500 mm/min (98"/min), single hit (Brother control) and it has a Cp over 12 at that speed. Total recorded variation against a cast aluminum surface was <30 microns (0.001") over 750 measured parts. Most of that variation can be attributed to the fact they were measured on a Zeiss Calypso Scanning head CMM and where the CMM hit may not have been where I hit on the as-cast surface with the probe.

Click to expand...

Tony,

What kind of variation would you estimate on machined top/bottom surfaces, and in comparison, in your experience, how would the Renishaw fare strictly in terms of repeatability?

Thanks.

dstryr said:

What would you guys go with?
I have the OMP40 on my 2 5 axis machines. Works for what I use it for. I have a bunch of production jobs I'm trying to move to the horizontal and it just got out of warranty so Mori can't limit me who I can use to add the probe now.
Wish I would have bought it with the machine as it has sucked without having one.

Click to expand...

What size is your Mori? Will optical system be sufficient for space coverage, or you have to go radio? Do you have all necessary control options open (G31 skip, user macro, sufficient memory)? I would consider Marposs as a supplier too.

Orange Vise said:

Tony,

What kind of variation would you estimate on machined top/bottom surfaces, and in comparison, in your experience, how would the Renishaw fare strictly in terms of repeatability?

Thanks.

Click to expand...

I am probably not understanding the question entirely. The Renishaw here repeats to about .0002" on a machined surface...? Wouldn't this be normal. Again, this is kind of why I didn't understand the question. The repeatability of ours at work is great, are you asking the difference between what the probe 'says' vs an actual measurement?

It is not one hit measuring though, but I think there are options in the manual to make it a one hit. This would probably be a better comparison. I have no idea how good the Renishaw is on one hits.

For the record it is about 5 seconds to touch a single surface with 2 hits.

Orange Vise said:

Tony,

What kind of variation would you estimate on machined top/bottom surfaces, and in comparison, in your experience, how would the Renishaw fare strictly in terms of repeatability?

Thanks.

Click to expand...

I would expect <5 microns on a machined surface from my experience. How much of that is measuring error and how much would be probe measuring and how much would be machine error would be the question that needed an answer. I have a Blum video that shows it measuring through a big glob of grease and it has a repeatability of 0.002 mm over something like 20 hits. I've been using them in a production environment since 2010. They do a measurement basically twice a minute, 24/7.

Mike1974 said:

For the record it is about 5 seconds to touch a single surface with 2 hits.

Click to expand...

I'm turning the probe on, (it is already in spindle from end of previous cycle) making a rapid to the part from home then a 70 mm protected approach in Z to the part. 1 hit each, 2 places 30 mm apart, adjusting C for angularity and back to tool change. Total time, cycle start to #2 tool loaded is <5 seconds.

Renishaw will exchange. He a crashed probe for a brand new one for about $1000. Crash, destroy, pull probe tip and battery, overnight to Renishaw, pay $75 extra and they overnight a NEW one back to you WITH probe tip and battery. That was a big selling point for me. I break things. Nice to have someone replace with NEW probe for that cheap.

PROBE said:

What size is your Mori? Will optical system be sufficient for space coverage, or you have to go radio? Do you have all necessary control options open (G31 skip, user macro, sufficient memory)? I would consider Marposs as a supplier too.

Click to expand...

500mm

Im sure i have everything I need since it has a nc4 laser on it.

Orange Vise said:

Tony,

What kind of variation would you estimate on machined top/bottom surfaces, and in comparison, in your experience, how would the Renishaw fare strictly in terms of repeatability?

Thanks.

Click to expand...

Not Tony, but I've done some recent testing on our Mazak Lathes, as we're investigating using in-machine probing to collect diameter measurments, and send the data to an SPC system. We're using a Renishaw RLP40 (Radio Lathe Probe 40mm body diamter...) in our machine. Using a 4"/100mm per minute feed rate, the probe will record the diameter, with less than .0015mm variation. This was done with the same part in the machine, simply looping the measuring cycle, and watching the #137? macro-variable, which records the "measured X-value" from the probing cycle. I think that's pretty good, since a .0015mm total variation actually translates to half that value when you think about the diameter/radius relationship of diameter to the probe.

The trick is to do your initial "protected-position" approach move with a fast-feedrate, but then to the actual measuring move with a slow feedrate, in order to minimize the probe-skip-feed hold delay...

Orange Vise said:

Tony,

What kind of variation would you estimate on machined top/bottom surfaces, and in comparison, in your experience, how would the Renishaw fare strictly in terms of repeatability?

Thanks.

Click to expand...

I'm not sure how you are measuring these things, possibly the details make all the difference. On my DMG with HH glass scales and a Renishaw OMP40-2, if I repeatedly probe a machined surface (moving all axis in between hits) I get a scatter enclosed by about +/- 0.000030. That is moving slowly on the actual probe approach and doing a spindle orient to ensure the same switch and kinematics are in play. If I do a tool change in between hits, generally less than 0.0001, I have seen the occasional outlier at 0.0002 - I presume due to seating the tool holder in some different dirt .

If the spindle is oriented for each probe hit, the only inaccuracy for which the probe system itself is responsible is switch accuracy and re-centering, and flight time of the signals to the control. If the flight time is consistent (it may not be in radio or IR schemes) it would be comp'd out in the calibration procedure. If the spindle is not oriented then the polar variation in switching accuracy of the probe comes into play. The machine is responsible for the rest of the inaccuracy: lost motion in measuring system, lost motion in the ways/guides, signal processing time variations. Those will be the same for any probe in the spindle.

When probing at fast speeds, the flight time may become the largest source of error. The Renishaw OMI documentation states that the transmission delay is 144 µs, and an addition 10 µs rise time. It does not state if that is average, maximum, or what the min-max variation is. At 100 in/min probe speed, the axis has moved 0.00026 in that time. If calibrated at the same speed the average might be comp'd out, but the variation appears as measurement error. There are similar variations in control processing time, I have never seen that spec'd. The slower you probe, the less these affect accuracy.

Perhaps everyone already knows these things, in that case Milacron can come in and label me Captain Obvious (again).

Probes are as repeatable as stated in respective data sheets, generally 1 micron on 500 mm/min hit feed and 50 mm long stylus. The OMI (or any transmitter) delay is constant, therefore it has no influence on repeatability scatter. 1 hit is the measuring software topic, mainly used to shorten the measuring process, but not always efficient.
The cause of results scatter in repeatable measurement of same point is machines PLC and/or CNC cycle time. Usual PLC cycle time is 10 milliseconds, CNC - 40 microsecond. As timing of the trigger in cycle is random, the delay from the hit moment to log moment can vary from 0 to 10040 microseconds for systems using standard skip, or from 0 to 40 microseconds for High Speed Skip. The higher hit feed, the greater scatter. Renishaws measuring routines use 30 mm/min hit feed.
At this feed the "per definition" result scatter is 5 micron [30/60]/[1/0.0104] for standard skip. If High Speed skip option is available, the hit feed can be enlarged
25 times for result of same 5 micron scatter.

PROBE said:

Probes are as repeatable as stated in respective data sheets, generally 1 micron on 500 mm/min hit feed and 50 mm long stylus. The OMI (or any transmitter) delay is constant, therefore it has no influence on repeatability scatter.

Click to expand...

This cannot be strictly true can it? It is not spec'ed in Renishaw's public documentation, but both the probe and interface are using clocked logic, and the IR link has acquisition time, all leading to some variability. It may be inconsequential or not, but it must be there. For the OMI-2 interface, switching times alone are listed as "max 100 µs".

I am probing at 250mm/min feed, and getting about 1.5 micron total scatter vs. 2 micron worst case spec for probe. This is with a HH 430 control, probably a little tighter than your average Fanuc . It doesn't say in the spec, but I assume the 1 micron spec includes polar variability?

Having now used both, i'd stick to the OMP40 for spare parts purposes unless there's a significant upfront cost difference. They both appear to work well, but not different enough to care.