Manley SLAM! Stereo Limiter and Mic Pre

SLAM!stands for "Stereo Limiter And Micpre" and it pretty much describes what it will do to most VU meters. We had to put a switch on this limiter to drop its internal VU meters down 3 & 6dB to keep the poor little needles from bending– it'll get LOUD fast (hence the exclamation mark). And on top of being an amazing pair of (actually four) Limiters, and Class A tube mic preamps, it also has masters degrees in DI, AtoD and DtoA, VU and PPM but that was too much for an acronym.

When it comes to classic gear, especially classic limiters, we can all agree numbers like LA2s, 1176s, 2264s, and others come to mind. What if all those sounds were to be found in one stereo tube unit. Interested? We combined our favorite Electro-Optical circuit (ELOP®) with a damn fast FET based brick-wall limiter reminiscent of some cool classics. And we added a sidechain filter that can remove low frequencies that makes it more useful for a variety of tasks, while retaining that easy, "it just works" quality that has made it a favorite for vocals everywhere. The FET-based limiter has different desirable characteristics that both optimise the signal for digital recording, due to the brick-wall capability, and let you dial in how clean or crunchy, or punchy, and how loud and proud you want it. Manley is excited to announce the SLAM!

GET IN: The SLAM! starts with tube Mic Preamps on both channels with selectable phase reverse, high pass filter, and switchable phantom power (on the back panel). We’ve got plenty of gain for you in this newtube circuit, up to 60dB with 20dB more in the limiter– plenty for soft singers using low output ribbon mics. The input attenuator is right up front, like a variable pad so your loud rock ‘n’ rollers won’t cause a problem either. Track direct with the Direct Instrument input or bring your line level inputs into the SLAM! for processing. The DI inputs have selectable input impedance, depending on how far you plug the 1/4" cable into the jack: push it in all the way and the input impedance is 100KOhms, pull it out halfway and it becomes 20 megOhms which reduces pickup loading and allows all the highs to get through. Come on in! 

Get it LOUD... two different and distinct limiters per channel:

The ELOP® LIMITER:Our beloved Electro-Optical Levelling Amplifier circuit, the ELOP®, still the favorite for vocal tracking, shows up here with its new switchable side-chain high pass filter making it much more useful for mixes and other tracks as well. One of the most appreciated aspects of our original ELOP® is the simplicity of a two knob limiter, which is optimum for demanding pro recording pressures. You can still grab two knobs and have complete control of level and dynamics and be ready to hit that record button. Intuitive controls and a straight ahead signal path through quality components simply work. The ELOP® limiter implementation is done like we did in our VOXBOX® which pioneered the idea of limiting the mic signal before it hits any active stage. The ELOP limiter moves to the line stage when LINE is selected. The ELOP sidechain provides different HP filters on a toggle. The 200 Hz setting has been tailored for vocals and adds a touch of de-essing.

The FET LIMITER: Here’s a new fast FET based "brick wall" limiter that spans the range of clean predictable limiting to the grunge and gravel that other FET based limiters are desired for. The FET RELEASE control allows you to dial up a variety of colors. There is even a "CLIP" setting that provides as round a shape as has ever been available. (Read more below about the CLIP settings.) We aimed at "sound" first, then experimented with circuits until we heard what we wanted discovering fresh approaches that provide a very wide range of useful dynamic control. For example, you can set up for a fast brick wall limiter with slower releases for minimum artifacts, or go with the faster releases to get some crunch and then get extreme loudness. The Attack control is unusual too providing the sonic control common with its historic brothers but very unusually retaining the zero overshoot “none-shall-pass” brick wall that defines a real limiter. And it’s punchy! As the saying goes, “Louder is better.” The SLAM! optimizes levels to analog to digital converters with ballz, warmth and loudness verging on silly or just clean and safe.

What is especially nifty is the ability to use the ELOP® and FET limiters together at the same time dialing in whatever amounts of each effect you like. Every limiter has its own character or sound and here we have combined two powerful world-class tools that can be uniquely blended and mixed to taste and purpose. Even the metering is appropriate for each limiter so you can easily see how each is contributing.

Metering on the SLAM! is exceptional. The two full size real VU meters show you the raw input levels, the final output levels, and the ELOP® Limiter gain reduction in the traditionally preferred format for visually displaying apparent loudness. VU’s are cool. Noticing how mastering engineers often need a switchable pad on their VU's because the "optimised" levels tend to pin the needles in the red, the SLAM! VU’s also have a 0, -3dB, -6dB switch to help with this situation. There are also very fast and super accurate multi-color LED ladder peak meters that display the input and output levels and the FET gain reduction, and some combinations. They will do Peak, Peak Hold timed, and Peak Hold until reset, and also allow you to manually set where the individual bars change colors to match your other gear. And with maximum clean I/O levels of +30 dBu, it is a lot more likely you will use these meters to see what’s clipping downstream. It probably won’t be the Manley tube line drivers inside!

The back panel should keep everybody happy. In the sea of jacks including the big power supply connector, you'll find all the usual XLR’s for transformer balanced I/O and 1/4" phone jacks for direct unbalanced I/O. Eight Bantam jacks in a mini-patchbay configuration handle the side-chain inserts and linking functions. You can insert EQs into the ELOP® and FET limiters, and there are more jacks for you to mix in an external signal into the side chains. Another pair of jacks provides for easy linking for 5.1 surround work if you have three or four SLAM!'s. It’s all here.

Speaking of Linking... all previous Manley dynamics units require both channels to be setup for near identical settings rather than the more common "just set the Left side". The SLAM! lets you do it both ways. The "L Side" method is the "STEREO LINK" mode and is quick and convenient but, like others, uses a mono mix for the side-chain. The "BOTH & EXT" adds the rear panel link jacks to the link buss so you could chain up several SLAM!s for linked surround work. Both or all 6 channels should be set similarly. This mode is usually better for mastering as it responds to the true peak level of the channels rather than an artificial mono mix.

Outside the Box, that's where the power supply is. The illuminated power switch turns on the external power supply. Keeping the power supply far away also keeps additional heat and magnetic fields away from the precious audio. The power supply is housed in a sturdy cast aluminum 7" x 4" x 5" box that is easy to mount to or place in the bottom of a rack. A seriously humongous 16 pin connector, and 6' multicore cable connects the two boxes.

The SLAM! also seems a natural as a mastering tool. The FET limiter works great on mixes, and the ELOP® now has a high pass filter switch that is spot-on for mastering. We offer a dedicated mastering version which eliminates the mic preamps, adds detented controls replacing the pots, and provides extra metering options. Manley compressors and EQs seem to be the most popular pieces of gear ever when it comes to mastering. The ultimate analog rack for the final important touches to the mix would have include the SLAM!, The Variable Mu® and The Massive Passive. Combine this with some digital processors of similar calibre and one has a respectable and very effective set of mastering tools. That Manley rack plus some Manley Microphones would make an ultimate stereo recording chain too. This is gear you’ll want to keep for a lifetime.

Features and Specifications

• Input Tubes:2 x 12AT7A NOS GE specially selected by Manley Labs for lo-noise and stable bias
• Output Tubes:2 x 6414W NOS USA dual triodes
• I/O:MANLEY transformer coupled Balanced Inputs and Outputs
• Micpre:Selectable 48V phantom power and PHASE REVERSE
• Gain:60dB max Micpre, 43dB max DI, 20dB max Limiter Gain
• Input Impedance:2000Ω Micpre, 1MegΩ DI
• FET Limiter:Attack: approx. 100μS; Release: 10mS to 2Sec; Ratio: better than 20:1
• ELOP Limiter:Attack: approx. 10mS for 6dB GR; Release: 2.5 Sec; Ratio: 10:1
• Frequency Response:5Hz to 60KHz
• Maximum Output:+32dBm, +30dBm (into 1KΩ load)
• THD+N: Dynamic Range:115dB typical
• Output Impedance:200Ω
• Power Consumption Slam! Analog:
  • 0.480 Amps (480 milliamps = 480mA) @ 120V = 57.6 Watts
  • 0.240 Amps (240 milliamps = 240mA) @ 240V = 57.6 Watts
• Outboard Power Supply:factory set for 100V, 120V or 220-240VAC operation for original destination country's mains voltage.
• Operating Mains Voltage:changeable with power transformer re-wiring via switch and fuse value change.
• Mains Voltage Frequency:50~ 60Hz
• Size:19” X 12” X 3.5” (occupies 2u)
• Shipping Weight: 25 lbs.

Manufacturer's Suggested Retail Price

• Manley SLAM!: $7,400 USD
• Mastering Version SLAM!: $8,600 USD

   

• Digital I/O Interface option: WAS $2,500 and is NO LONGER AVAILABLE since 10/2009

TECH Notes and Upgrade Information

NOTE: The DIGITAL DAC/ADC OPTION for the SLAM! Has been discontinued 10/2009:

There was a slot in the back for our optional 24 Bit/96KHz Analog to Digital converterand 24 Bit/192KHz Digital to Analog converter card. This allowed one to record directly to the computer or workstation and provided a convenient way to use a digital insert tojump into the analog realm and back. An AES digital input received the stereo digital signal to convert to analog for processing by the SLAM! when the DAC input was selected. Or this killer DAC's Output could be used for monitoring, listening, or to be sent to some other analog processors or tape machines. 

The digital inputs and outputs were on XLR's for AES/EBU standards and there was also a BNC for Word Clock and Super Clock input. The BNC was added for users'convenience and not just to help the unit's jitter performance. Thejitter spec was "Almost Zero" and just about unmeasurable regardless of whetherit was locked to its own crystal, the AES input, or the Word/Super Clock. A 7-way rotary switch selected 44.1, 48, 88.2 or 96K, AES IN, Word Clock andSuper Clock, for the AES output data rate. Notice we didn't say "samplerate". This ADC and DAC always used the internal 192K sample rate! Trulyamazing sample rate converters based on SHARC DSPs were used to almosteliminate (not just reduce) jitter and provided the sonic benefits of 192Ksampling, even at 48K or 44.1 data rates. These were very fast exceptionalsounding world-class converters.

The ADC also had a pair of switches for DITHER and NOISE SHAPING and WORDLENGTH, plus something unusual: a filter select switch. We believed that themanufacturer's choice of filter was one of the biggest differences thataffected the sound of a converter, so we gave the user 3 choices, 80K minimaleffect, 20K Analog, (a passive filter similar to the 18K on the Massivo),and a 20K Digital FIR for something comparable to other converters. Therewere also 3 filters on the DAC side set for 20K, 40K gentle, and 80K. All ofthese were passive and quite subtle and useful. There was a dedicated balanced+4 phone jack output for the DAC, so these could be tube-converters or highend solid state. The ADC was similar, but tube or passive(!) front end, thatallowed a mastering EQ to be cleverly used in this chain.The DAC output was always available. There was even a twisted way (called "Expert Mode") to patch the SLAM! so that one could have the MIC PRE/ELOP feeding an EQ, (or someother thing), returning back to the SLAM! to drive the FET LIMITER, meters, and the A to D converter. Digital I/O Interface option: was $2,500 and is NO LONGER AVAILABLE since 10/2009

SLAM! Digital I/O Interface Option Specs:

• DAC: AES 24Bit 192KHz upsampling from 44.1, 48, 88.2, & 96KHz
• A to D: 24Bit 96KHz
• Power Consumption Slam! Digital:
    0.688 Amps (688 milliamps = 688mA) @ 120V = 82.56 Watts
    0.344 Amps (344 milliamps = 344mA) @ 240V = 82.56 Watts

UPGRADE: MANLEY SLAM! Digital Versions before 1/2006

PURPOSE:To add cooling fans to Manley SLAM Digital Version so the converter board does not overheat.

Serial numbers for units that will benefit from this update are:

SLAM 000-291 that have a "D" suffix which had the factory-installed Digital I/O card installed before 1/1/2006.MMSLAM 000-043D that have a "D" suffix which had the factory-installed Digital I/O card installed before 1/1/2006.

ANALOG ONLY UNITS ARE NOT AFFECTED. ALL DIGITAL SLAMS SHIPPED AFTER 1/1/2006 ALREADY HAVE THE FAN KIT INSTALLED.

For FAN KIT installation procedure, please click here.

TECH NOTES: MANLEY SLAM! Digital Versions

PURPOSE:To replace the Anagram Technologies ADC Module if it has failed due to high heat conditions.

Serial numbers for units that this applies to are: Any serial numbers suffixed by "D" with a failed A to D module.

For module replacement procedure, please click here.

UPGRADE: MANLEY SLAM! ADDING DIGITAL CONVERTERS:

PURPOSE:Add the Anagram Technologies ADC/DAC converters Digital I/O to an analog SLAM!

Serial numbers for units that will benefit from this update are: Any serial numbers suffixed by "A"

To add Digital I/O to your analog SLAM, please click here.

UPGRADE: MANLEY SLAM! Digital Versions only 5/1/2003

PURPOSE:Update the Anagram Technologies ADC Module to Software Version 3 to speed up clock estimation routine.

Serial numbers for units that will benefit from this update are:

SLAM 000-005; 022-095, and 119-121 which had a factory-installed Digital I/O card installed before 5/1/2003. ANALOG ONLY UNITS ARE NOT AFFECTED. MASTERING VERSIONS ARE NOT AFFECTED.DIGITAL CARDS SHIPPED AFTER 5/1/2003 ALREADY HAVE VERSION 3 SOFTWARE UPDATE.

For update procedure, please click here.

Some SLAMMY questions. Actual email correspondence:

From the FAQ page:

FAQ #11. Are the micpreamps in the SLAM! the same as the ones in the Voxbox?

Read the this page again and note the word "new". The mic preamps in the SLAM! are similar to the input stage of the Steelhead actually.

Question: The SC Filter seems to work backwards from the description in themanual. When I switch it to the 200Hz position, I get more gainreduction on a bass heavy program sources not less.

It is not backward, it is level compensated (more or less). The 100 hzsetting has more gain than the 'flat' and the 200 hz setting has more gainagain plus has some extra HF boost for mild de-essing. On mixes, it oftendoes seem over-compensated, and on many individual tracks it seemsunder-compensated so it is just a compromise. Without the compensation, thethreshold jump tends to be more disturbing and annoying, especially the 200setting which often made one turn the ELOP® threshold to max. Adding more potrange just made it more difficult and touchy to set.

Question: What's the deal with the CLIP thing on the Release switch and the CLIPand BOTH settings on the Mastering Version SLAM!?

The Slam! Mastering Version includes a mode called "CLIP" on a rotary switchwhere the Input Switch lives on the regular version. This "CLIP" setting isa fixed ceiling that is included to prevent "Overs" or digital clippingthat, in general, is less desirable sound-wise. This clip is set about 0.5dB below digital full scale and is absolutely non-adjustable other than howhard one drives a signal into it. The "CLIP" setting engages two parallelclippers, one in the A to D path and another in the XLR output path afterany transformers. With effort and extreme hot levels one can still reachdigital full scale, mostly due to the effects of severely squaring up asignal and its effect on anti-alias filter ringing plus the fact that whilenon-feedback analog clipping may be somewhat preferable, it is still analogand is not 'perfectly mathematical'.

Besides that mode, there is a mode called "BOTH" which engages both the"CLIP" setting described above plus the "LP LIMIT" mode. "LP LIMIT" is setup for some of the extremely hot pre-compressed tracks a mastering engineergets today. This is a fairly complex setting that watches the input signaland alters the FET limit character once the signal is sufficiently loud.Then it allows the FET limiter to pass a little extra highs and mostly limitlows, thus preventing some of the 'dullness' often occurring with limitersand some of the distortion that can occur with FET limiters in particular(but one of the reasons we like them). Each of these three modes (LP LIMIT,BOTH, CLIP) are intended for the challenge of loud aggressive rock masteringor other situations where a little artistic freedom to shape the sound andpush the 'limits' may be called for. Of course, it all depends on how hardone drives the Slam! and pushes into these circuits so there can be a largerange of effect from virtually transparent to in-your-face craziness. Thesemodes will not suit every task, nor is "CLIP" a magic process thatguarantees a hit - it still requires taste and discretion, and our bestadvice is that a little goes a long way.

OK, on the RELEASE knob of both the regular and mastering versions of theSLAM! is a setting (furthest counter-clockwise) also called CLIP. This is avariable clipper where you are encouraged to adjust the threshold and attackswitch to set up an effect. Essentially it speeds up the timing circuit forthe limiter and the FET is allowed to operate on the waveform rather thanthe envelope of the signal. There is still some shaping of the controlsignal so that it isn't as 'buzzy' as pure clipping. So this clippingcircuit is essentially a bonus effect that can be used to radically treat asignal and probably should be mixed with the straight signal at the console,where one can also EQ both the raw and clipped channels and mix to taste. Weshould point out that the "LINK" switch should not be used and that it wouldbe rare that using say the right signal to clip the left might be useful.This effect is intended for individual tracks where a severe distortioneffect might be cool (like raunchy guitar, synth, or snare) but won't havemuch usefulness on mixes or essentially clean tracks. It seems to works beston sounds that are already somewhat crunchy and where some added nastinessor edge could be fun.

So, there you have it, two or three types of clipping available in themastering SLAM!. Which brings up a good point - 'CLIPPING' is a prettygeneral term for overload distortion where the tops of waveforms are choppedoff because it can't go any higher. Why it is pretty general is becauseevery device seems to have its own signature clip sound, even though simpletheory might suggest much more similarity sound-wise, we know in practicethat this isn't the case and each device has its own overload character.Some distortions are OK for some sounds, some to be avoided at all costs andsome that might be described as magic when you get it set up just right -and there is the big hint - usually the sweet spot is pretty small and itreally is a matter of being fussy with levels and faders to make it workover the distance of a song. You don't have the benefit a guitar player haswith immediate touch and pick force to create the sound, you are probablyworking with recorded tracks or the musician may not be hearing thedistortion effect (it helps if they can) so play those faders to set updrive and return levels throughout the song to really make it work. That'san old trick. Unfortunately, there is some buzz on the net suggesting that"all you need" is some vintage whatever overdriven and you get amazingsounds, like push this button for a platinum record or take this pill forinstant enlightenment. The reality is that it still takes work, musicaltaste and chops. The magic is not only what you use but when and where andthen how you use it.

Hi guys,I just got off the phone with my dealer. Finished the order of what I thinkis the finest analog front end/plug-in for any DAW on the market today:Manley Gold Reference Stereo Mic.Manley SLAM! with digital I/OManley Massive PassiveWouldn't you agree? ;-)

Hutch: I envy you. That is a good chunk of my dream system. Add a DACS mic pre and a pair of Royers and a few pieces from GML or Cranesong and Hutch would be a happy camper recording anything that moved.

Well, I have a question about the combo SLAM!/Massivo:

Using the Massivo as a side chain EQ is a no-brainer, but what arethe hookup options to use it pre or post limiter in a:"front-end" or (D/A > Massivo > Limiter > A/D or D/A > limiter > Massivo > A/D)

You can EQ before limiting like this:

a) SLAM!’s DAC 1/4” balanced O/P to Massivo IN. Massivo O/P to SLAM line in which automatically goes to ADC.

Or EQ post-limiting like this using "Expert Mode":

b) SLAM!’s DAC selected, “expert mode jumpers set” , 1/4” O/P to Massivo XLR in. Massivo 1/4” O/P to SLAM! XLR balanced O/P (yes, feed the output) (actually just the transformer is bypassed using the 1/4”, you are actually feeding the FET drive pick-off point and ADC)

What about "insert" application ( Mic pre > Massivo > limiter > A/D or Mic pre > limiter > Massivo > A/D)

Sorry, no way to insert between the Mic Pre section and ELOP Limiter section. We debated having this feature but it would have required twice the number of tubes, or chips in the signal path, so we passed on officially offering that option. The issue was/is “would it hurt the sound?’ Extra stuff in the path would have compromised it. The ELOP is married to the first stage, just like we did in the VOXBOX.

But as described above, you can get inbetween the ELOP and the FET limiters and insert between those two.

Also, is there any way to go:Mic pre (or instrument in) > A/D > digital into FX unit > D/A > limiter > analog out ??????

Same basic reason as above. While it would have been nice to allow any of these modes and it was certainly possible (if we were a company that marketed features and not sound quality), it would have countered sound quality. Sometimes, we have to make difficult decisions to eliminate features and flexibility if those features would overtake primary goals of ‘minimalist class A all-tube signal path’ or end up costing outlandish amounts.

While we knew we were putting in a fine Mic Pre, to be honest, we didn’t expect the reaction it has gotten.We might be able to add a chip based Mic Pre line driver for you. It might be weird though because there is some ‘magic switching’ made possible by integrating the unit and NOT having a dedicated Pre out. For example the ELOP® limiter switches to the ‘Pre-MicPre’ signal point (taking its sidechain from the limiter section as usual). Also we have ‘muting’ circuits on the mic pre for non-mic/di modes to prevent weird leakages and low level distortions. These would have to be removed and you would have to live with reduced performance for the sake of an interesting patch. One also has to take a good hard look at input and output jacks, and how the controls are shared. For example, Mic & line share the XLR, Mic Pre and Limiter share the INPUT level, and Source selection switch, isn’t ‘just’ simply switching which path goes to the Limiters. Simple yet complex.

So it would appear it's notpossible to insert an external compressor into the chain after the micpre and still use the ADC option - am I reading it right?

Sort of - The mic pre circuit has a high impedance output not suitable as aline driver, and adding circuitry there would have compromised performance,so there is no standard brainless "Insert" per se. Plus the mic pre, dependson X amount of gain added in the line driver, plus benefits from having bothgain controls, etc.

However, if one reads the manual, there is an "Expert Mode", where onechanges a few jumpers indicated on that block diagram, which allows one totake the signal from the 1/4" output, route it through any desiredprocessing, and return it through the Male XLR "output" (doubles as an Inputto the ADC and FET drive point because transformers are bi-directional) andaccomplish what you are actually trying to do, yet better than a simplestraightforward traditional "Insert".

The advantage is that one gets the limiting where it is optimal, thesidechain and metering where they are optimal (each side of the externalbox) plus gets a more-or-less passive input to the ADC. The downsides are:1) It requires a special cable sometimes, 2) It does require a bit of brainsbecause the SLAM! will seem flakey if the 1/4" is used without a return tothe Male XLRs in "Expert mode", 3) The Slam! needs to be opened to changethose jumpers, and may require needle nose pliers and the manual.

Anyways, about 25% use SLAM! in "Expert Mode" daily and do what you aretrying to do, and are loving it. Of course, we'll also repeat that, "TheSlam! is not the easiest of all audio gear to use, and limiters in generalare challenging and may not be appropriate for everything. That said, howmany guys overuse the L2 or digital brick wall multi-band limiters, giventhat they damage music worse? Its just a tool, and the tools we need to relyon most, we carry above our necks. That is the best real hint we can give.

I am still so confused. How do I hook in this "Expert Mode" and patch my Massive Passive EQ in between the ELOP and the FET?How do I go: MICPRE -> ELOP -> MASSIVO -> FET -> A/D? Spell it out Step by step for me:

Set the jumpers inside the unit to "expert mode" following instructions on page 21 section 4 "changing jumpers" of the owner's manual. (It is also called "advanced mode".)

Get yourself a pair of male XLR to TRS 1/4" cables and a pair of female XLR to TRS 1/4" cables. These should be wired tip to pin 2, ring to pin 3, and sleeve to pin 1.

Run a set of your 1/4" to male XLR cables out of the unbalanced 1/4" outputs from the SLAM! into your Massivo input XLR jacks. Come out of the Massive Passive 1/4" Outputs using your 1/4" TRS to female XLR cables and run those into the SLAM!'s XLR transformer balanced output jacks. Yes! You come back into the SLAM! through the output XLR jacks.

This would be for a "live" application with a multi-channel FX unit inthe mic-path.

True, but we already build the VOXBOX® which is a tried and true live work-horse and a stand-alone AD/DA is in the planning stage. For complete routing flexibility, this would be a cool live combo (overkill perhaps). The SLAM! was intended more as a DAW insert or front end or back end (notice that we used the word OR rather than AND). To do all 3 approaches at once, would ‘best’ be done by several boxes rather than just one. Of course, if sound quality were not an issue, it would be easy and cheap to do with a bag of op-amps and jacks.

And finally, I want to put both units in separate standard SKB rack cases. Is the"headroom" in 2 (SLAM!) and 3 (Massivo) unit cases enough to providethe needed air flow?

We suggest at least 3/4” clearance top and bottom for each unit to maximise life and minimise repairs. They are not as hot as some tube gear, so you might get away with no clearance, but expect reduced life on the electrolytic caps. There are 4 tubes in each, plus the SLAM!® has many regulators & relays to help make it warm.

Thanks for your help and most of all for providing me with such outstandingrecording equipment.

I’m sure you’ll get the results you expect. Thanks for the kind words and support.

--Hutch

More SLAMMY questions and answers...

Question: There's no word clock out, correct? How would one send clock to otherunits? And why is it that Hutch seems to advise to use an external clocksync for other units vs. using the Manley? Or did I misunderstand?

Conventional wisdom suggests that the best approach is to use an A to D'sinternal crystal for clock and send that clock everywhere, especially the Dto A's. The reasoning is that A to Ds are most jitter-proof when locked totheir internal crystal clock rather than if marginally locked to someexternal sync. This is generally true but the SLAM! converters are a specialcase. They are permanently locked to a high multiple of 192K crystal clockand use sample rate conversion to talk to the outside world. This softwareprocess effectively cleans up both clocks. Standard operating procedures dueto typical inadequacies of jitter tolerance may not apply to SLAM!.

There are several reasons for not including a dedicated word clock out. Someof this is in that gray area between fact and opinion. I have rather strongfeelings about jitter and the fact that so few converters are particularlygood at rejecting it and that most need a word clock box to sound OK. Thisis why we took the approach of using ASRC instead of PLLs, and this made ourconverters jitter immune (Benchmark does a similar thing). Our process alsopretty much eliminates the typical phase noise issues of the internalcrystal oscillators so they don't need to be "special". This means that ifwe had put on a word clock output it would be just a buffered version of theraw crystal which may be OK, but wouldn't really compare to a good wordclock box like so many people use today and really is the best answer forbig set-ups and multiple converters.

The primary benefits of good word clock distribution is 3 fold.A) It allows a studio or complex to use a form of house sync, whether thisis video, a stable precision word clock or simple clock (cleaned up) asgenerated by the DAW. The latter is used when clock rates change daily.B) The DAWs and other digital devices are generally considered to be jittertolerant and the next device where clock becomes important sound-wise is theDACs. The SLAM! DACs are jitter immune and so there is a wild card scenarioas to whether SLAM!s generic crystal is adequate for unknown DACs after Xamount of raw distribution. An external DAC that is that jitter sensitive isprobably best served by a word clock box intended for that function.C) One can usually use the SLAM!'s DAC to monitor or as a key section in thechain and achieve almost perfect jitter immunity without all the aboveconvolutions in many scenarios. Just set the DAW to follow SLAM!.D) Word clock wouldn't be needed if other converters were like SLAM!'s.

The above overview has a few implications:1) The SLAM! A to D does not suffer when following an external clock.2) An external clock box will probably be a more stable a clean referencethan the SLAM!'s crystal output due to the fact that the external box has afew PLLs to clean up their own crystal phase noise or outside references.3) The word clock box allows easy selection from several sync sources.4) Distribution of clocks is equally important and this requires a healthynumber of BNC and XLR outputs - again the proper job of a word clock box.5) The SLAM! AES/EBU output can feed said clock box and the mission of clockdistribution is then elegant rather than possibly jury rigged.

In other words, if you need a great word clock distribution scenario thenthe right method includes a box designed expressly for that task. It shouldbe up to accepting AES as a master clock which SLAM! can supply. Slam! isalso meant to follow it, if a different reference is desired, but withoutthe usual draw backs. We didn't create the SLAM! to fix a lot of otherquestionable converters, and primarily not to have the same flaws itself.

Question: And as for Maximum integration - does that include meteringintegration? If not, there's a reason to go with a standalone box. Whatpart of the maximum integration with the SLAM! would make one feel thatit is too advantageous to pass up, especially if ("if" being the operativeword at this point) there are other factors lending advantage to anexternal box?

Can't ask EveAnna exactly what she meant there right now but my concept of"integration" is a single box that interfaces easily with any DAW (includingas an "analog insert"), and acts as a very high quality versatile interfaceto and from other gear. This is why there are mic pres, DI, and limitingdone where it is most beneficial, metering both in VUs and LED peak styleand a tasty amount of class A tube color. Of course, HEDD is a lovelyconverter with my favorite simulations of tubes, but frankly the SLAM!converters have an advantage in the time domain where goove and feel aresignificant concerns, for me anyways. The LED meter also indicates A to Dovers in every mode. While it may be tempting to draw comparisons, methinksthe SLAM! is in a rare category in the sense that it is all in one box anddesigned to work synergistically - the diverse sections blend in a nice way.

Question: Again, thanks for the response. I have no doubt in the quality of theconverters, judging from your reputation, Hutch's documentation in themanual (fascinating!) and what folks like Lynn and Fletcher have said(as well as most involved with the 3db comparison.) I'm sorry forpestering, but for me this is a sizeable investment, and I just want toresearch as much as possible.

Very wise to ask and ponder.Just read your original questions below and maybe I can help there too.As described above the clock question depends on how much distribution isneeded and other converters that might benefit. I am guessing that you havea clean simple high quality set-up that doesn't require much in terms ofsync distribution. The SLAM! can be the AES master clock to your DAW, andthe usual benefits of word clock won't apply to SLAM!s jitter immuneconverters. With other converters you may find that a word clock box ismandatory but not so with SLAM!s converters. Makes it easier, less concernsapply.

In regards to "minimalist" the A to D uses the SLAM! (shared) outputtransformer as the front end to the converter. It doesn't get any closer tominimalist and passive than that. Other converters use X number of op-amps.There are a few advantages & disadvantages compared with a stand-aloneconverter. The advantage is that the limiter and LED meters watches that"input/output" and reacts accordingly even if an EQ were inserted in thatpath. The disadvantage is that as a stand-alone it is so minimalist that youdon't get gain controls (we thought there were too many already). Of coursethis only applies if the main part of SLAM! is used for something else andis in a different path.

Hope this helps, sorry for complex answers but these are complex questionsnot because of you, but because of the hoops most guys go thru because somany converters are too jitter sensitive. Methinks excessive band-aids, andanother page we'll cringe over in the upcoming history of digital audio.

--Hutch 9/2003

PS - We will be introducing a stand-alone converter (someday) based on SLAM!s with 3independent word clock inputs and outputs clocked by astoundingly cool DDSsystem (Bigger Ben). So this will do everything that you are thinking aboutthe best way possible today but won't be as good of a $ deal as just goingfor the SLAM! converter option, nor be quite as minimalist. It is also 6 to9 months down the road at least as opposed to 'available now'.

Question: I'm curious about the need for two compressors? I'm just starting to get into mastering. I have ordered the mastering version of the SLAM! Do you think that the SLAM! would cover that need since it has two limiters per channel? I understand the need for two EQ's, and the only reason I could see having two compressors, is to use one to compress, and one as a brickwall limiter.

Hutch's Reply: Interesting question. The answer hinges on the fact (or opinion) that all compressors do sound different, plus the desirability for different sounds in ones’s arsenal. Perhaps, we don’t “need” multiple compressors, and perhaps we don’t even “need” to compress at all, but there does seem to be advantages to having the choice and it seems that with hundreds of compressor/limiters on the market and most studios owning X number of different units and most mastering studios regularly chaining 2 or 3, that its all about options and getting the most out of each, if indeed compressors are different sounding and if most are at their worst when over-used. Put another way, each generally begins to show its character beyond a few dB of gain reduction, so some guys use several units to minimize obvious characterizations. All of us designer-types wish we could build that ONE compressor that did everything and did it all best, but so far none of us has. We can make one that maybe does a few things best tho’.

Question: Can the FET and ELOP® limiters be used simultaneously and independentlyof the converters? I was wondering if, during mixing, I would be able to use the convertersto go to my Masterlink.... and at the same time use the ELOP and FET on various individual tracks.

Yes on the ELOP®, no on the FETs and Peak meters. The latter are tied to theXLR output which can be used as an input to the ADC. The DACs can always be usedindependently.

Question: Can I insert an EQ in between the ELOP and FET limiters?

No, one cannot insert between the ELOP and FET sections basically because the levels around there are -20 dB, and are at rather high impedance, and they are unbalanced, and absolutely unsuited as a line driver. And No, one can't use the bantam jacks that way, as they are just meant as side-chain inserts to control how the EOP or FET responds (for example inserting an EQ with a 7K boost can provide de-essing).

However, one can use "Expert Mode" to accomplish a similar goal, but the actual effect is rather optimised and thought out - but somewhat un-intuitive (and why we call it "expert mode"). It is described in the manual and is worth a try. The end result is that:

1) The ELOP is first in the chain and responds to the raw un-EQ'ed signal.

2) The FET is next in the chain and responds to the EQ'ed signal.

3) The EQ is next in the chain (see I told you it was non-intuitive) What is cool is the signal is limited before the EQ electronics but the sidechain is responding to the final output that the A to D would see. That prevents much EQ overloads.

4) The SLAM! LED output meter and A to D converter is last in the chain. The FET side chain is here too - strange loop.

Anyways, this should really do what you need. Why didn't we add 4 or 6 op-amps into a simple passive class A signal path to provide some convenient but compromised and questionable patching path? We were just trying to maintain some sound quality, and we saw a clever way to pull it off, that had a little drawback of being a bit complicated.

Question: This will probably sound dumb, but when setting the input level on theSLAM!, what should I be looking for on the meters? Should the VU hangaround 0? That seems pretty quiet. Should it be in the red? Should I belooking at the LED meter instead? I'm using a Roctron Prophesy guitarprocessor XLRs with "MIC" selected as the source on the SLAM!. "9:00" onthe input knob sets the VU to around 0 for most sounds, while the 12:00position pins the needle to the right.

The first rule is: "There are no rules". OK maybe "If it sounds right, it isright". The second rule is "with music, trust your ears, more than youreyes".

Regarding the meters and input / output levels: Setting the input level sothat the VU in INPUT mode is near ZERO will usually be the safest especiallyin regard to FET Limiter distortion. One can then easily set the THRESHOLDSand OUTPUT Levels. If you are sending to an A to D converter, I mightsuggest using the LED Meters to set the OUTPUT LEVEL rather than the VU inOUTPUT mode (the latter requires a bit of interpretation/experience becauseVUs are slow).

So all we are saying is, if the VUs (in INPUT) are pinned, one might begetting a little more FET distortion (they do that), but one might not knowhow far the VUs are pinned and when extreme, might result in the preampclipping, but there is quite a bit of headroom before that is likely. For araunchy guitar sound, you have my blessing to go nuts and just go after asound that you like, regardless of what the meters tell you, after all, somedistortions can be a lot of fun. I might warn you, that if you are clippingthe A to D converters occasionally, that at least you verify, that the soundof your A to D clipping is OK, because it often isn't. This is why one mightwant to use the LEDs both on the SLAM! and A to D to prevent too hot of anoutput. Generally, one can be -10 below Digital Full Scale without problems.

Regarding the knob positions: Don't worry about 9:00, 12:00, etc. If it werepossible to always have pots set at 12:00 we wouldn't need them then, and wecould have saved a bunch of money and panel space. They are there to beused. Input and Output levels are supposed to be kinda flexible. In fact theway you are using it, that Input Level in MIC mode has about 60 dB of range.This makes it a bit touchy, but it needs to accommodate a many situations.

Hope this helps, and hope the SLAM! is helping get some killer tone.

-- hutch

I love the Slam overall, however when using it across my 2 mix I noticed it really dulling the mix. I hired a good tech and we did some tests.We mix on a Neve 8058 that has been modded and tricked out by Pat Schnieder.We put an oscillator straight into line input as a control and the other side through the unit. We found when engaging our (perfectly calibrated) mix insert that the 15k dropped by 1.75 db. We also found when running a tone into the unit and then into a normal line in (a scenario we might use when utilizing the unit as a buss comp) that the 15k dropped by 3.5db. The tech therefore contends that the Slam can't drive a 600 Ohm load. Is this possible?

Hi

Yes, you and your tech pretty much diagnosed the problem. Admirable.The root cause is the output transformers which droop HF somewhat into 600 Ohms.

One cure might be to use the unbalanced outputs, which won't droop that way.

On a similar note, interestingly, capacitively loading the outputtransformers (like a very long cable) tends to create a HF rise or peak.Both of these symptoms are very common for output transformers and arealmost never noticed but might explain different results for differentpeople or situations. About once every 2 years somebody notices.

If the tech is feeling brave, there is a network (R&C) loading the OPtransformers optimally set for 10K loads and these might be changed to getcloser for 600. We use a 3K & 10K square wave plus freq sweeps and a RCsubstitution box to arrive at good values. Doubt whether it would be enough.

However, the best answer is to drive a higher impedance input, because tubecircuits in general (ours included) lose some headroom driving lowimpedances. And while it should still should be fine, one hopes to optimise.

Lastly, limiting may be perceived as a dulling factor because it reducestransients and a lot of HF is in those pesky transients. One can use alonger attack time or boost a bit of HF EQ to compensate.

-- hutch