How are you handling line-level sources through a mic-splitter?

[lampmeister]

Hey people,

I'm trying to think ahead and anticipate potential problems here and I'm wondering what the best way to handle line-level sources (keyboards, drum machines, samplers, laptops audio interfaces, etc) when live recording.

Mic splitting transformers aren't usually designed to handle line-level sources so presumably the line-level signal needs to be buffered down to mic-level and then put through the mic-splitter. Either that or the line signal needs to go through a seperate line-level split.

How are you handling this out there?

Any and all info/experiences appreciated.

Cheers!

Dave.

[Jim vanBergen]

Check the specs of your transformer. What are you using?

My split is VERY happy with line level as well as mic level signals. It can do +24 before the THD skyrockets.

Hope this helps!

[lampmeister]

Cheers Jim!

In that case I'm glad I asked now before I buy the transformers for my splitter!

Just out of interest what transformers are in your splitter Jim?

Cheers,

Dave.

[Gilliland]

While it may not be difficult to split a line level signal at line level, I've never actually seen it done in practice. Every sound company that I've ever worked with has simply used a DI to convert the signal back to mic level, then fed it into the splitter.

So the key is to make sure you've got some good DIs.

[lampmeister]

Cheers Gilliland, appreciated.

[foldback]

The dirty little secret of every audio transformer is, THD increases for low(er) frequencies. The best audio transformers available will have problems with signal level at 20Hz.

A mix of Nickel and Iron in the laminations can reduce distortion but cuts down signal level. High Iron content will produce a transformer that can handle a lot of level but will have higher distortion below 50 Hz.

Jensen is brutally honest about their spec's for their splitter transformers, they tell you max input is +2 dB at 20 Hz. Thats it. If you go up to 30 Hz, distortion decreases by a whole order of magnitude, probably why API uses 30 Hz for a lot of their specs, they usually quote some pretty high signal levels.

Fortunately, transformer distortion is a friendly beast. It's not like a solid state device, it's soft sounding. Not good, but not like a digital over.

The real problem is for the device that is feeding a transformer when it saturates. A transformer begins to look more like a dead short to the source. Instead of reflecting the impedance connected to the secondary, a transformer stops being a transformer when you go outside its boundaries. When you hear gross distortion, it is probably the source device clipping because the transformer looks like a dead short, afterall, a transformer primary is just a piece of wire wrapped in a coil.

Before you buy a bunch of splitter transformers, buy one as a sample and experiment with it. That is the best way to understand the performance.

Some splitters include pads to knock big signals down closer to mic level. There are bridging transformers made just for line levels too. I have some in separate boxes that I run down an unused snake channel if I have a big signal to deal with.

Broadcast situations are the worst for big signal levels. The old Betacam decks often output +8 nominal.

Best of luck to you. Go do some experimenting and listening. Cheers.

Mark

[lampmeister]

Great info Foldback.

Getting 2 or 3 likely candidate transformers and experimenting is the way forwards me thinks.

Cheers!

[Remoteness]

Originally Posted by Jim vanBergen

Check the specs of your transformer. What are you using?

My split is VERY happy with line level as well as mic level signals. It can do +24 before the THD skyrockets.

Hope this helps!

What JvB said. Many XFMRs can handle +24, but in any case you must check the specs first.

Originally Posted by Gilliland

While it may not be difficult to split a line level signal at line level, I've never actually seen it done in practice. Every sound company that I've ever worked with has simply used a DI to convert the signal back to mic level, then fed it into the splitter.

So the key is to make sure you've got some good DIs.

IMO, using a DI is the simplest way to go about this situation.

Originally Posted by foldback

The dirty little secret of every audio transformer is, THD increases for low(er) frequencies. The best audio transformers available will have problems with signal level at 20Hz.

A mix of Nickel and Iron in the laminations can reduce distortion but cuts down signal level. High Iron content will produce a transformer that can handle a lot of level but will have higher distortion below 50 Hz.

Jensen is brutally honest about their spec's for their splitter transformers, they tell you max input is +2 dB at 20 Hz. Thats it. If you go up to 30 Hz, distortion decreases by a whole order of magnitude, probably why API uses 30 Hz for a lot of their specs, they usually quote some pretty high signal levels.

Fortunately, transformer distortion is a friendly beast. It's not like a solid state device, it's soft sounding. Not good, but not like a digital over.

The real problem is for the device that is feeding a transformer when it saturates. A transformer begins to look more like a dead short to the source. Instead of reflecting the impedance connected to the secondary, a transformer stops being a transformer when you go outside its boundaries. When you hear gross distortion, it is probably the source device clipping because the transformer looks like a dead short, afterall, a transformer primary is just a piece of wire wrapped in a coil.

Before you buy a bunch of splitter transformers, buy one as a sample and experiment with it. That is the best way to understand the performance.

Some splitters include pads to knock big signals down closer to mic level. There are bridging transformers made just for line levels too. I have some in separate boxes that I run down an unused snake channel if I have a big signal to deal with.

Broadcast situations are the worst for big signal levels. The old Betacam decks often output +8 nominal.

Best of luck to you. Go do some experimenting and listening. Cheers.

Mark

Marky Mark, you RAWK!

[Karl Jackson]

I did a tour once that involved recording the same show every night for several weeks, using MKH800s and an HV3D. On two occasions, situations demanded that I run the output of the pre down a XFMR split to the FOH position (typically I run my own snake). On both occasions, I experienced distortion when (and only when) the orchestral bass drum was played fff. My guess in both cases was either core saturation at low frequencies or some other strange loading situation.

Definitely something to watch out for.

Karl Jackson

[Jim vanBergen]

Originally Posted by lampmeister

Cheers Jim!

In that case I'm glad I asked now before I buy the transformers for my splitter!

Just out of interest what transformers are in your splitter Jim?

Cheers,

Dave.

1) Jensen. Can't be beat!
2) Lundahl, which has the Jensen spec at a slightly lower cost, but requires time to get, Imported thru one guy in Florida, while Jensen is wound here in the good ol' USA. I like the way these sound!

I don't know what is in the Radial Engineering 8x unit, but it looks tasty...

To be frank: I ALSO HAVE WHIRLWIND. They work, and sound perfectly fine, but can't handle past +20. The spec is nowhere near as good as Jensen or Lundahl. But they work. (Last choice, sometimes it's what is available.)

Sorry for thhe delay in response... have not been able to access the board

[lampmeister]

Cheers Jim,

I'm in the UK actually where Lundahl transformers can be ordered from Canford Audio.

Others I've looked at are:

Sowter (UK)

Cinemag (US)

Crimson (US)

Stevens & Billington (UK) who I've not heard of before some googling threw them up.

...and of course Jensen.

I'll post links to any other sources as I find them just for the information pool.

Cheers!

[foldback]

John at Crimson is a friend, but more importantly, he makes some great parts for reasonable prices and it will be hard to find a nicer person on the planet.

I've been to his shop many times, it is totally old school. Beautiful ladies winding and assembling precision audio transformers, one at a time.

[Juan T.]

Hi. New guy to this forum but have been reading it for a few weeks now.

We have a Whirwlind 56 pair Lundahl three way splitter that has done well with line level sources. We even have some Lundahl single channel splits that in built into some DI size boxes by Whirwlind. Not sure if WW gets them from that guy in Florida as JvB noted but they seem to be readily available from WW. That said for many of broadcast remote events it is common to have a dedicated line level splitting snake system also referred to as a high level splitter. I have heard that a good quality dedicated line level splitting transformer is a Cinemag Audio Transformers By Cinemag Inc.. We intend to build one in the near future. Cinemag does seem to require a long lead time and also made in the USA.

Juan

[lampmeister]

Good info chaps, thanks very much.

I chatted with a guy from Stevens & Billington (UK) the other day and they seem pretty good to me at a reasonable price.

The specs of the particular one I'm interested in are as follows (from the S&B website):

Microphone splitter transformer [model] 904

This microphone splitter transformer has a 1ct:1+1+1 ratio. The device is fitted with inter-winding screens and has a CMRR of -76dB @ 10Khz. The .5% THD figure is reached at +5dB @ 40Hz and the .5dB band-width is 15Hz - 46 Khz.



If anything there rings alarm bells with anyone then let me know quick! :-)


Cheers!

[Jim vanBergen]

http://www.jensen-transformers.com/mic_sp.html


DING DING DING DING DING!!!!!!!!!

[lampmeister]

I'll take that as a hint then Jim...

[Remoteness]

We should do a shoot out between a variety of mic splitting XFMRs...

Double blind test and really figure it out.

We can put the ear against the gear & deside on the bang for buck ratio once and for all.

We can also set up a few different groups of folks taking notes on what they heard.
Then we can tally up all our thoughts at the end of the discovery.

[foldback]

Great idea Steve. When I return from this trip, I would love to help with some parts to listen to.

Mark

[Jim vanBergen]

Really, the best thing to do is to put 600ohm, 1:1 wound transformers on an Audio Precision or Neutrik A2 and measure THD across the frequency spectrum and from -60 to +24dbu, as well as measuring consistent frequency response. The biggest problem I have seen is poor low end frequency response.

JvB

[Remoteness]

Excellent idea JvB.

Setting test rig is a great way to cross relate the measurements with what we hear.

[lampmeister]

That sounds brilliant!

[Jim vanBergen]

Thanks for the kind words, but for me it was kind of growing up in regards to taking or "owning" the responsibility for every element of my work. I needed to find out how to really quantify between an acceptable device and a great device.

Using the Neutrik A2 to compare splitter transformers was actually what convinced me to spend the copious $$ required.... then, when we compared low end response by ear alone , it seemed kinda obvious WHY it was worth the extra money.