Replacing/duplicating a Meyer analog processor?

S

Simon Coffin

Sophomore
Mar 2, 2011
133
1
18
State College, PA
www.medianengineering.com
Folks,

Thinking of possible paths for a monitor update later this summer. We have 8 Meyer UM-1C wedges with 4 of the M1-A processors. Getting all the drivers reconed is probably necessary and easy enough, but the processors are no longer made and hard to find on the used market. Meyer tech support indicates the M1-A uses analog phase and amplitude correction that cannot be replicated on a DSP, not even their own Galileo. Which means if I want to get more mixes we would have to find a way to translate the analog crossover to a DSP?

Does anyone have any experience doing that?


Alternately, we could buy all new wedges, but as is always the problem, the budget requires we stay in the used market. Whatever we get will need to be as rider friendly and sound as good as the Meyers.


Thoughts?
 
Re: Replacing/duplicating a Meyer analog processor?

I'd get a measurement program (I use SMAART) and run an RTA, Transfer Function and Impulse Response on each input to output of the M1-A . Store all the traces, then run the same on the new DSP and try to match them to the Meyer. Once matched on the screen, then measure the same on the live speaker through the M1-A and then the new DSP. Hopefully they should be close.

-Mark
 
Re: Replacing/duplicating a Meyer analog processor?

Mark Oakley said:
I'd get a measurement program (I use SMAART) and run an RTA, Transfer Function and Impulse Response on each input to output of the M1-A . Store all the traces, then run the same on the new DSP and try to match them to the Meyer. Once matched on the screen, then measure the same on the live speaker through the M1-A and then the new DSP. Hopefully they should be close.

-Mark
Click to expand...

This assumes that what the Meyer processor is doing can be represented by a single transfer function (i.e. it is an LTI system). Given that there are sense lines involved, I'll bet that some of the filters have response that change depending on level. That can't be replicated on any commercial DSP that I'm aware of, or with any single transfer function.
 
  • Like
Reactions: Guy Graham
Re: Replacing/duplicating a Meyer analog processor?

Rob Timmerman said:
This assumes that what the Meyer processor is doing can be represented by a single transfer function (i.e. it is an LTI system). Given that there are sense lines involved, I'll bet that some of the filters have response that change depending on level. That can't be replicated on any commercial DSP that I'm aware of, or with any single transfer function.
Click to expand...
I have heard (cannot confirm) that also. For example-when the levels get to a certain point, the crossover points start shifting to protect the drivers.

Of course at that level-it could be considered "protection" and any sonic issues would take a back seat to keeping it working.

IF you get to that level, you should move to a louder product.
 
  • Like
Reactions: Guy Graham
Re: Replacing/duplicating a Meyer analog processor?

This topic almost ties directly into the question I had in a different post. Out of pure curiosity, I would like to know what is going on in this processor. What about hooking this processor up to an amp, returning the sense lines, and then measuring a transfer function at different drive levels? It still won't tell you the whole story, especially in regards to fast gain reductions, ie limiting. But if the crossover is dancing, you should sure be able to see that.
 
Re: Replacing/duplicating a Meyer analog processor?

Rob Timmerman said:
This assumes that what the Meyer processor is doing can be represented by a single transfer function (i.e. it is an LTI system). Given that there are sense lines involved, I'll bet that some of the filters have response that change depending on level. That can't be replicated on any commercial DSP that I'm aware of, or with any single transfer function.
Click to expand...

You could absolutely do this kind of thing with a MediaMatrix NION processor. You just need to thoroughly define what you want to have happen beforehand, and that, I believe, is the real problem with trying to replicate this particular processor's function. Only the designers at Meyer know what the processor does, and I doubt you are going to get them to tell you. If you could get that information from them, you could absolutely replicate the behavior with a NION.

The trick is with getting the sense lines back into the NION. You can use a line or mic/line input card and add some external pad to drop the signal level coming from the amplifier output down to something that is reasonable. Basically, you want to drop it to the point where when the amp is running full tilt boogie, you have about 30 to 33dBu at the input of the NION. Then you can start setting up conditional situations where the crossover point moves, limiting engages, EQ changes, or phase adjustment changes all based on the input level at the returns from the amplifier.

Basically, it would take a LOT of woodshedding to sort out, but it could be completely handled. As I said above, the key is getting a good definition up front regarding what the processor is actually doing. That could take the form of getting real engineering information from Meyer (yeah, good luck with that one!!! There is at least one spec sheet where they list the amplifier power in a product as "Adequate") or take a processor, hook a loudspeaker or two up to it and start taking measurements at all different signal levels and maybe with different spectral content: noise burst, sine sweep, pink noise, and even different music selections. Obviously you will need both magnitude and phase information for all these measurements.

Once you have a good idea what the processor is actually doing, only then can you start building your signal flow and setting up the real time adjustments that need to happen. It could be done, it is just really, really time consuming. Ergo, unless you are game for putting in all that time and effort to reverse engineer what the analog box is doing, you are far ahead of the curve to just buy more of the analog controllers. If you can find them.

Have fun and good luck!
 
Re: Replacing/duplicating a Meyer analog processor?

David Sturzenbecher said:
This topic almost ties directly into the question I had in a different post. Out of pure curiosity, I would like to know what is going on in this processor. What about hooking this processor up to an amp, returning the sense lines, and then measuring a transfer function at different drive levels? It still won't tell you the whole story, especially in regards to fast gain reductions, ie limiting. But if the crossover is dancing, you should sure be able to see that.
Click to expand...

I expect you are on the right path there David. I would suggest also hooking up an actual loudspeaker to the amplifier too, just to provide a complete "Meyer system" on that end of the rig. After all, the sense lines coming out of the amplifier are also connected to the loudspeakers.

As I commented in my other post in this thread, I expect one could develop a pretty good model of what the processor is doing if they can take enough measurements and interpret them appropriately.
 
Re: Replacing/duplicating a Meyer analog processor?

I've heard that Meyer processors do weird tricks with the crossover points at high levels, and I've also heard that that was an old tale told about some very early product and no longer true.

I wouldn't expect there to be any changes with dynamics except at extremes, and we should be avoiding extremes anyways. Trying to match the processor outputs using SMAART would be a good learning exercise, and probably would yield pretty good results. The system wouldn't be a "true" Meyer system as far as rider friendliness goes, but it would probably still sound pretty good.
 
Re: Replacing/duplicating a Meyer analog processor?

Here's a snip from an old Meyer Design Reference with the M-1a's amplitude correction.

M-1a snip.JPG

All the text info regarding the sense lines reads like they are used only for limiting.

Here's a block diagram of the M-1a.
M-1a block.JPG

Certainly don't know, but I really have a hard time believing anything beyond limiting is occurring ...
 
Re: Replacing/duplicating a Meyer analog processor?

Ivan Beaver said:
I have heard (cannot confirm) that also. For example-when the levels get to a certain point, the crossover points start shifting to protect the drivers.
Click to expand...
Scott Helmke said:
I've heard that Meyer processors do weird tricks with the crossover points at high levels, and I've also heard that that was an old tale told about some very early product and no longer true.
Click to expand...

I think Meyer Sound Labs has refuted this claim many times.

At this point any products that rely on these processors are so old they are unlikely to perform as originally intended anyway.

Mac
 
  • Like
Reactions: Guy Graham
Re: Replacing/duplicating a Meyer analog processor?

Mark Wilkinson said:
Here's a snip from an old Meyer Design Reference with the M-1a's amplitude correction.

View attachment 14407

All the text info regarding the sense lines reads like they are used only for limiting.

Here's a block diagram of the M-1a.
View attachment 14408

Certainly don't know, but I really have a hard time believing anything beyond limiting is occurring ...
Click to expand...
With only a 6 dB per octave crossover, HF limiting would effectively move the acoustic crossover up an octave when 6 dB of HF limiting is applied.

Regardless of Meyer tech support indicating "the M1-A uses analog phase and amplitude correction that cannot be replicated on a DSP", I think it could easily be improved upon.
 
Re: Replacing/duplicating a Meyer analog processor?

Art Welter said:
With only a 6 dB per octave crossover, HF limiting would effectively move the acoustic crossover up an octave when 6 dB of HF limiting is applied.

Regardless of Meyer tech support indicating "the M1-A uses analog phase and amplitude correction that cannot be replicated on a DSP", I think it could easily be improved upon.
Click to expand...

+1 :thumbup:
 
Re: Replacing/duplicating a Meyer analog processor?

Remy G. Santerre said:
Meyer no longer recones Ms-12 drivers
Click to expand...

FTFY. Meyer doesn't recone drivers, period. The cost for them to recone a driver, put it through the same testing as a new unit and still have it (potentially) fail is more expensive than manufacturing a new one, especially if it does fail. This is why there are also no recone kits available - no one in the field can put them through the testing that Meyer does. But you can absolutely get replacement driver units for unpowered UM.

OTOH, I believe they may still accept trade-ins on HF driver units for credit towards a new one.

If you're ever in the SF Bay area and have several hours to burn, set up a tour of the Meyer facility. The driver manufacturing and QC line alone is worth it.


To the OP:

Simon, the sense lines are what make the old analog processors unique - replicating them is a difficult process, and you still may not get the same results as the original analog processors. At that point you'd be better served picking up more M-1A processors used, even if you have to send them back to Meyer for service. Which, if you have any questions about the functionality of your existing processors, Meyer WILL service them.
 
  • Like
Reactions: Guy Graham
Re: Replacing/duplicating a Meyer analog processor?

Folks,Thanks for the info. Duplicating the Meyer processor would be an interesting intellectual exercise, but in balance is more work then I can invest. That being said:

Simon, the sense lines are what make the old analog processors unique - replicating them is a difficult process
Click to expand...

Brian, if the sense lines are only being used as feedback in the limiting circuitry, wouldn't it be fairly straightforward to measure when they engage and replicate that in a DSP, assuming the amp gain is known? Comparing the input signal to the processor and output from the amp, should indicate when the limiting starts and how aggressive it is. It seems to me, since Meyer didn't know what amps were being used, they could not just set a static limiter threshold in the processor, so had to use a feedback loop via the sense lines. The diagram posted earlier doesn't indicate any function for them beyond limiting, but if they do anything fancier, all bets are off.

And ref, the original intent of this post, we are probably going to replace the Meyers with RCF TT powered wedges.

Cheers,
 
Re: Replacing/duplicating a Meyer analog processor?

Art Welter said:
With only a 6 dB per octave crossover, HF limiting would effectively move the acoustic crossover up an octave when 6 dB of HF limiting is applied.

Regardless of Meyer tech support indicating "the M1-A uses analog phase and amplitude correction that cannot be replicated on a DSP", I think it could easily be improved upon.
Click to expand...

Hi Art, +2, DSP replication.... and even improvement, is bound to be possible..

Thinking about the acoustic crossover shift....and what that means as far as power delivered to the drivers...
I can see a relative power shift, depending on which section is limiting harder (and I can easily see the accompanying acoustic crossover shift)

...but as far as thinking about absolute power delivered to the drivers, it seems the acoustic crossover shift would not effect anything ....each driver is still getting whatever its limiting section allows independent of the other section.
I mean you could turn one driver off, and it would still limit exactly the same, wouldn't it?
 
Top Bottom