Forte Models 4 and 6 solid-state amplifiers (review) (Nov. 1992)

FORTE ACRES AND A MULE

Tube slave CG listens to the Forte Models 4 and 6 solid-state amplifiers.

Forte Model 4 power amplifier Forte Model 4 stereo class-A solid-state amplifier. Rated output: 50Wpc into 8 ohms (17dBW), 20Hz-20kHz at no more than 0.1% THD. Input impedance: 47k ohms. Output impedance: 0.02 ohm. Damping factor: 400. Current capability: 50A peak, 16A continuous into 0.1 ohm. Output noise: <300µV. Price: $1595.

Forte Model 6 stereo class-A/B solid-state amplifier. Rated output: 200Wpc into 8 ohms (23dBVV), 20Hz-20kHz at no more than 0.1% THD. Input impedance: 47k ohms. Output impedance: 0.02 ohm. Damping factor: 400.

Current capability: 50A peak, 16A continuous into 0.1 ohm.

Output noise: <300µV. Price: $1695.

Both: Approximate number of dealers: 50.

Manufacturer: in-Concert, 7325 Roseville Road, Sacramento, CA 95832.

Tel: (916) 348-0600.

Fax: (916) 348-0678.

ALL THINGS MUST PASS

Most high-end manufacturers are known almost as much by their designers as by their products; personalities like VTI’s David Manley, Thiers Jim Thiel, and Theta's Mike Moffat have become as well-known to audiophiles as their tube amps, speakers, and digital gear. And because these talented designers put so much of their own personalities into their designs, it's big news when one of them leaves the company they've become associated with.

"Didja hear? The Sixers traded Barkley to Phoenix!" "That's nothing; Adcom traded Campos to NAD for two draft picks plus a minor league feedback topologist to be named later!" Think Threshold and you think big, meaty, solid-state muscle amps like the mighty SA 12/e. You also think Nelson Pass, the man who designed this and other classic Thresh old amplifiers and preamps. Well, think Nel son no more; Pass has left Threshold, handing over the Chief Designer mantle to a 26-year-old Swede named Mikael "Mike" Bladelius. I spent some time last April hanging with Mike at Audiophile High End Show in LA, and was very impressed with both his technical expertise and the fact that a Swede could hold his tequila. Mike, who joined Threshold a couple of years before Nelson left, hit the ground running, whipping out Threshold's DAC 1/e digital processor as well as a whole line of gear for Threshold's more affordable Forté division. The Forté gear can be thought of as Threshold's version of Conrad-Johnson's Sonographe, or Levinson's Proceed: simpler cosmetics, lower prices, but with the same high-quality design philosophy as the top o' the line stuff.

COFFEE, IGBT, OR ME?

The Forté Models 4 and 6 amps are the first Threshold designs to use Toshiba's new IGBT, or Insulated Gate Bipolar Transistor, as output devices. IGBTs are hybrid devices, combining the most desirable characteristics of both MOSFETs (high input impedance, tube-like sound, easy to drive) and bipolar transistors (low output impedance, high current handling) in a single package? IGBTs are also highly linear li'l guys, and Mike's taken advantage of this by totally eliminating all global feedback from the amplifier circuit; only a small amount of local feedback around the input stage is employed.

Both Forté amps are DC-coupled from input to output without the need for DC servos, and, unlike most solid-state amps, there's no series output inductor to stabilize the amplifier by defining the ultimate high-frequency rolloff with capacitive loads. When a solid state amp lacks a series output inductor you can be sure that either the amplifier is very stable or the designer is not.

Both Forté amps are built to a high standard, and the rounded corners and low-key, elegant cosmetics are a welcome sight in this price range. Mounted on the thick brushed aluminum faceplate are just the centrally mounted power switch and the teenie weenie red LED "on" indicator, and both Fortés kindly turn on and off without any deadly DC belches. The beefy power supply features a 500VA toroidal transformer, 60,000µF of supply capacitance, and dual 35A rectifier bridges, one per channel. Parts and build qualities are first-rate. Surprisingly for a toroidal transformer, both Fortés had a fair amount of audible hum coming off the chassis; most amps do this, including the VTLs, but toroidal power transformers are usually very quiet. The vibration hum is higher than normal, but inaudible from the listening chair if your amp is situated over by the speakers.

This is the last time I'm going to include "the amplifier's back panel sports a pair of gold-plated Tiffany-style RCA jacks and the customary 5-way speaker binding posts, as well as a detachable IEC-style AC power cord" in an amplifier review, because, with a handful of exceptions, THEY ALL DO!!! Shortly after I received the first review

[1. The great-sounding BUF-03 buffer used by Theta, Kinergetics, and my own Buffered Passive Preamp is a hybrid device with a FET input and a bipolar output stage.

2. Of course, all non-OTL tube amps have series output inductors: their output transformers, duh hey now! ]

samples, Forté called me up to say they'd just added balanced inputs to the amps as standard equipment. I didn't have a balanced pre amp to audition the balanced inputs of their amps, but I sent the amps back anyway and got a couple of new ones to ensure that I was listening to current production units. It should be noted that the Forté amps offer true balanced differential inputs, in stark contrast to most of the faux-balanced gear in the High End. Usually, "balanced" gear either

a) feeds the balanced input signal to an op amp, which then promptly unbalances the signal and sends it along to the rest of the circuit, or b) simply wires XLR pin 2 in parallel with the unbalanced RCA input jack and ties pin 3 to ground. Method A keeps the advantages of balanced-line noise rejection, but adds an op-amp and other devices in series with the audio signal; Method B adds nothing extra to the signal path, but throws away all the noise rejection that balanced lines are used for in the first place! Neither approach is kosher for high-end audio, and it's good to see Forté doing it up right at this price point.

DEAD RINGERS

Now here's where the going gets cool. Forté's Model 4 is a 50W class-A stereo amp, and the Model 6 is a 200W class-A/B stereo amp, even though they're the same size and weight and look identical. As is customary, I asked Forté for both amps' schematics so I could describe to you, Cherished Reader, just what's goin' on inside 'em. A short time later, a lone circuit diagram came curling off the fax machine without any indication as to which amp, the 4 or the 6, it was supposed to represent. So I picked up the phone and called Mike Bladelius.

"Yo, you only sent me one schematic!" "Yo, that's because the two amps are one and the same!" Unbelievable but true: the 50W Model 4 and the 200W Model 6 are exactly the same amp, right down to the last resistor! Both amps even use the same power transformer, but with different secondary taps for the DC supply voltages: the Model 4 rides on 39V rails, while the higher-powered Model 6 gets 76V to play with. And, obviously, the 50W Model 4's output stage is biased into class-A, while the 200W Model 6 is biased a bit lower to the more conventional class-A/B. Other than the different power-supply voltages and bias settings, though, the two Forté amps are identical. Same circuit, same number of out put devices (four pairs of IBGTs per channel), same everything. Well, not the same price.

Why is Forté offering two versions of the same amp? Simple. All things being equal, full class-A operation is definitely the way to go if your primary objective is sound; with both halves of the push-pull output stage conducting at all times, class-A amps are inherently more linear than class-A/B designs.

Unfortunately, because their output stages are always conducting at full power, class A amps run hot as hell even when idling, and waste more wall-juice per watt than A/B designs. That's why the vast majority of amps on the market are biased into class-A/B, which offers a much more reasonable balance between efficiency and power output.

By configuring the bias settings and supply rails, Forté is able to offer users either a lit tle bit o' class-A magic-see Sidebar-with the 50W Model 4, or a lotta class-A/B muscle with the 200W Model 6.

THE SYSTEM

In order to get a better picture of the Forté amps' full-range performance, I unhooked the mighty Muse Model 18 active subwoofer I usually use in cahoots with my VTL Deluxe 225 amps and heave-ho'd it out of the room.

This is getting ridiculous, though; if I've gotta grunt it out of the listening room every time I do an amp or speaker review, I just may put some of those big-ass gold-tone casters on the mofo! Maybe Steve McCormack will design some Audiophile-Approved roll around casters, with thousands of li'l Tip toes to couple the gold balls to the floor.

Maybe I took too much of this cough medicine.

===========

John Atkinson on CLASS-A

Given the misunderstandings about amplifier output-stage biasing, a quick primer is in order. The simplest form of power-amplifier output stage consists of a bipolar transistor, tube, or FET connected as a "voltage follower," with a voltage gain of unity. The signal is amplified by a previous stage to the appropriate high value to feed to the loudspeakers and the follower delivers this same high volt age to the load from a suitably low impedance. The amplifier's output stage therefore amplifies current.

The output stages of nearly all modern amplifiers consist of two follower devices in what is called "push-pull" topology. (The two devices bridge bipolar voltage rails and the output voltage is taken from the common connection be tween the devices. The devices can there fore be said to both "push" and "pull" the load.) The amount of DC current that the designer allows to flow through the two output devices-the "bias" current-has a significant effect on how this stage handles the signal: if sufficient DC current flows all the time to keep both devices conducting during each complete signal cycle, this is termed "class-A" operation; "class-B" operation is when the DC condition at each device's control input is A such that both are completely turned off when no signal is present. Class-A there fore has both devices handling the signal all the time; class-B involves only the device connected to the positive voltage rail being turned on during positive input signal half-cycles, and only the device connected to the negative voltage rail being turned on during negative half cycles.

As no current flows when there is no signal, and, when current does flow, does so only from either rail to the load, class-B operation is very efficient. Class-A, in which devices pass current from one volt age rail to the other all the time, whether there is signal or not, is very inefficient.

A class-A amplifier therefore runs hot test when there is no output-no current is being diverted into the load, and the output devices have to dissipate all the power. A class-B amplifier runs hottest (if you work through the math) at one third power, which is why we precondition amplifiers under test at this level.

Even at its most efficient, however, a class A amplifier can only deliver 50% of the electrical energy it draws from the wall to the load, while a class-B amplifier can, in theory, be 100% efficient.

As it takes a finite voltage to be present at the control input of a device before it will turn on, pure class-B operation results in significant amounts of what is called "crossover distortion," as each device hands the signal over to the other twice per cycle. In practice, therefore, class-B amplifiers are arranged so that just enough bias current flows to keep the devices turned on and at the start of the linear portions of their transfer characteristics, minimizing crossover distortion.

This is termed "class-A/B" operation.

As well as good sound quality, the advantages of hot-running class-A operation are inherently good linearity and constant stress on the power-supply and output devices, which means that there will be no signal-related effects in their transfer functions. The disadvantage is low efficiency, which means large heat sinks, high operating costs, high price, and four times the number of output devices for the same output power as a class-B amplifier (or only one quarter the maximum output power for the same financial investment in output devices). The advantage of cool-running class B operation is efficiency, hence small heatsinks, low price, and low operating costs. Disadvantages are inherently less good linearity (before the application of negative feedback), sometimes question able sound quality (especially in early solid-state amplifiers), and signal-related stress on the power supply and devices.

In class-A operation, the maximum current that can be fed to the load without turning off one of the devices can only equal twice the standing bias. By Ohm's Law, an amplifier must therefore have a standing bias of just under 1.8 amps to be able to deliver a maximum output of 50W into an 8 ohm load. With output voltages that would cause the 8-ohm power to exceed 50W, or with the amplifier driving loads lower than 8 ohms, each of the two devices starts to cut off each signal cycle: class-B operation.

I measured the Forté 4 as having 1010mA of output stage bias (for com parison, the Forté 6 has 300mA of bias current), meaning that the 4 will only deliver 16.3W into 8 ohms in true class-A before its output devices shift into class-B. Given the relatively small size of the 4's heatsinks, I suspect that this is the maximum bias that could be used without the amplifier melting down. Nevertheless, this does appear to endow the 4 with much of the sonic benefits of true class-A operation, as CG and I found from our auditioning.

-J A.

================

LPs were played on a Well-Tempered Record Player fitted with Sumiko's Analog Survival Arm Wrap and their new Blue Point Special cartridge, fed via AudioQuest Silver Lapis to an Audio Research SP-14's phono stage and then on via the Rec Out jacks to my Homemade Buffered Passive Preamp. CDs fed the Theta Data/Pro Basic II combo, linked together with AT&T glass cable real optical like. In addition, I compared both Forté amps with the $1200 Muse Model One Hundred, my first choice in a sub-$1500 solid-state amp. All interconnects except for the Lapis between the 'table and preamp was Kimber KCAG, and everything that didn't have a captive AC cable was plugged into an Audio Power Power Wedge with Kimber's kool man AC Power Kords.

Speakers used with the Fortés included, at one time or another, Spica Angeluses, Scientific Fidelity Teslas, NHT 2.3s, Apo gee Centaur Minors, Eminent Technology LFT-VIIIs, and ProAc Response Twos, with the bulk of the listening done with the ProAcs and NHTs. I tried a bunch of speaker cables ranging from AudioQuest to Cardas to Symo to Straight Wire to Kimber, but the silver Kimber 4AG 8' bi-wired pair to the ProAcs was clearly a cut above the pack. (It's interesting to watch the HOT NEW! BEST-EST! cables like the Lindsay-Geyer come and go while good ol' Kimber just keeps right on chooglin'; card-carrying Audio Shriners must have the latest obscure wire if they're worth their salt, of course, but even after trying the latest "hot" cables on the market, the Kimbers are the best I've heard.) From silver to gold: Richard Lehnert sent me the new MFSL gold-plated Ultradisc reissue CD of Elvis's That's The Way It Is, the soundtrack to the concert film of the same name. This is the King ca 1970, when he was well into the white jumpsuits and karate kicks but before he started to really put on the chunkage; a bona fide Las Vegas extravaganza, with 200 backing musicians, 60 black female backup singers, and Charlie Hodge to hand him his towels. If, like me, you long to hear gold-plated remastered versions of "Patch It Up" and "Mary In The Morning" sung as only Elvis could, you can't afford to be without this monumental disc; bravo, MFSL.

BLIND WILLIE'S JOHNSON

They say that a little knowledge is a dangerous thing. Maybe so, but then Danger is my maiden name. Before I even had the two Forté amps out of their shipping boxes, I had a feeling about how each one would sound; the 50W class-A Model 4 would sound extremely clean and clear at low to medium volume levels, with little deep bass and a smaller dynamic ceiling than I'd need to do Nirvana justice, while the 200W class-A/B Model 6 would kick effortless ass but sound rougher through the mids and highs.

This kind of preconditioned opinionating is what ABX loyalists point to with glee when they gather at AES conventions to compare '70s haircuts and target the High End for maladies ranging from the Black Plague to the cancellation of "My Mother The Car." Because I formed an opinion before I listened to the units, says the ABX crew, it's impossible for me to be truly objective about the sound of the amps. That's why the bumper stickers say "ABX Testers Do It Double-Blind"; to ensure the validity of the test, the identity of the DUT is kept hidden so the subject can be free of any preconceptions that might color his response.

You know what, though? I was tight on the money. The Forté Model 4 was very clean and clear, and didn't go very loud. Maybe it wasn't quite as clean and clear as I'd hoped for from a full class-A design, but then it probably needed a bit of break-in yet.

The Model 6 was a different story. It surpassed my predictions, kicking a much higher level of a much higher quality of ass than I'd ever expected it to do, and doing it with an even greater clarity than the class-A Model 4.

Was it Tim Cahill or Ron Jeremy who said there's no substitute for cubic inches? I'd expected the greater power output of the Model 6 to give it the edge in dynamics, but certainly not in basic sound quality; my experiences with both tube and solid-state gear have shown that the closer you get to full class-A operation, the more effortless circuits seem to sound. Forté claims to bias the Model 6 higher than most AM designs, but it really only operates in class-A for a couple of watts; the rest of the range is A/B. Still, the Model 6 was fast becoming my clear favorite of the two Forté amps.

A CHEF IN THE KITCHEN & A FORTÉ IN THE BEDROOM

When I get new gear in for review I usually break it in by inserting it into my bedroom system, which consists of a Muse Model One Hundred amplifier, NHT 23 speakers, Audio Quest's budget-reference Type 4 speaker cable, [3] a homemade passive preamp, and a JVC XL-Z1050 CD player. Other amps, speakers, and sources replace these units when I need to break them in for the requisite month'r'so, but these pieces make up the basic system that I listen to music on every day when I get up in the morning, get dressed, put on my truss, etc.

Accordingly, I switched off between the two Forté amps to drive the NHTs for several weeks before I moved them over to the listening room for review. The NHT 23 is a three way 4 ohm dynamic speaker with bass capabilities that belie its slim profile, and the Model 4 drove it fairly well. Still, I was bothered by an overall "smallness" to the sound. It wasn't so much a case of the little 50W amp distorting on loud peaks, but more of a closed-in character to the sound at even moderate levels. Coming after the big, open sound of the Muse Model One Hundred, the Forté Model 4 sounded as if it were subtly reigning in the NHTs in all directions. The bass was more restrained; dynamics were less forceful. The sound was very detailed, but it just didn't have the out'n'out pelvis of the system when driven by the Muse.

The Model 6, however, threw a saddle across the 23s and hollered, "gindele!" On bass-heavy thrash like Metallica, the Model 6 drove the NHTs louder and with tighter

[ 3 AudioQuest's 79e/ft F-14 speaker cable is very good, but I think their $2/ft Type 4 is the best value in speaker cables today. This simple four-leg, solid-core cable has wiped some muy expetuivo top o' the line speaker cables from other manufacturers in my various systems; before you spring that long, long green for cable you know you're gonna get rid of six months from now if not sooner, try the AudioQuest.]

deep bass than my usual Muse Model One Hundred. The Forté sounded a bit harder through the upper midrange, but its high end was open, clean, and free of congestion. During the time I had the Scientific Fidelity Teslas in my main listening room driven by the VTL 225s, I actually started retreating to my bedroom whenever I wanted to hear some loud, clean rock'n'roll; I ended up listening to the Teslas only when I had to for review purposes, and spending my music listening time in my bedroom with the Forté Model 6 and the NHTs. Even the big VTLs couldn't ameliorate the problems I covered in my review of the Teslas in Vol.15 No.4, while the much less expensive duo of the Forté Model 6 and NHT 2.3s played louder, cleaner, and with a great deal more musicality.

Impressed as I was with both Fortés, it was definitely time to move them over into the main rig and see what they could do.

AT LEAST I CAN SPELL "POTATO"

Listening to the two Fortés in my listening room, my initial impressions were enforced: whether driving the Spica Angelus, Apogee Centaur Minors, or the ProAc Response 2s, the Model 6 was clearly a notch above the Model 4 in musicality. Neither amp sounded particularly impressive driving the Eminent Technology LFT-VIIIs, where the top end became pretty dark and the dynamic ceiling lowered considerably, so I wouldn't recommend this combo. But with the other three very different-sounding speakers, the sound with the Model 6 was open, clear, and musical.

The Model 4 remained somewhat disappointing in comparison with the larger Model 6. Again, I kept coming back to the sense of "smallness" the Model 4 conferred to my system: the soundstage became both narrower and shallower, and the amp just didn't have the effortless quality I liked so much in the higher-powered Model 6. I spent a lot of time with each amp driving the sys tem, but when I wanted to forget about the gear and enjoy the music, I'd hook up the Model 6.

Finally satisfied that I'd gleaned whatever there was to glean from the two amps, I brought in their shipping boxes from the tool shed and went over to unhook the Model 6 from my system.

Only it wasn't the Model 6. It was the Model 4.

What the-?! Look, these two amps are absolutely identical except for the teeny-tiny lettering at the bottom of the front panel that says either "Model 4" or "Model 6"!!! I know I had their identities straight when I started listening to them in my bedroom, but obviously at some point I got them mixed up. I thought the Model 4 was the Model 6, and versa-vicey.

In other words, it had been Forté's Model 4 that had been the big winner here, not the Model 6. In other words, the 50W amp played louder and sounded better than the 200W amp. And aside from the differences in power-supply rail voltages and bias set tings, they're the same amp.

I was astounded. It made no sense. I put the shipping boxes back into the tool shed and came back to do some more listening.

FORTE ACRES AND A MUSE

Even though I'd already done some quick comparisons with the Muse Model 100, I hauled it back into the listening room for a head-to-head with the Fortés, matching the levels by playing track 1 of the first Audiophile Test CD and measuring the voltage across the speaker terminals. Both Forté amps are a good deal less sensitive than either the Muse or the big VTLs, and I ended up with the volume control of my gainless preamp almost full up for the Fortés, with the Muse requiring a bit less drive to achieve the same levels.

I got perfectly fine results with the combo of my preamp and the Fortés, but if you own speakers with lower sensitivities than the 87dB/W/m ProAcs and/or have a big room, I'd suggest mating the Fortés to a preamp with some gain.

Now that I had the two Fortés' identities straight, I continued to hear the Model 4 walk all over the Model 6. Driving the ProAcs, the 4 threw up a highly detailed, razor-sharp soundstage with a sense of ease that I rarely hear in solid-state amps at any cost, much less at the Model 4's $1500 price tag. It didn't match the VTLs' freedom from grain and congestion, but the Forté Model 4 is clearly one of the best solid-state amps I've heard.

Compared with the Muse 100, the Forté had a very slightly less clear midrange, with Lyle Lovett's vocals sounding a bit more hooded on the Deadicated CD than when played over the Musc Richard Lehnert's spoken voice on tracks 1 and 2 off the new Ste reophile Test CD 2 also sounded a little more veiled than with the Muse, or to a greater degree, than with the VTL 225s.

The Model 4's rendering of space also fell a little short of the mark set by the Muse.

While the soundstage width and depth were excellent for a sub-$1500 solid-state stereo amplifier, I felt the Muse bettered the Forté in absolute terms. The differences weren't great, but recordings that sounded huge on the Muse, like my Eden track on Test CD 2, were slightly smaller when the Forté was inserted into the chain. Again, the difference was slight, but it was there.

It was in the trebles, though, where the Forté edged the Muse. I find the Model 100's top end to be one of the most open I've heard in a sub-$1500 solid-state amp, but the Forté Model 4 was even cleaner. The brushed cymbals on Bob H.'s drum tracks on Test CD 2 had more air than with the Muse, and when the drummer started to really clang them Zildjians, I heard less off-putting hash and more of the true sound of the cymbals. When you really smack a cymbal, its pitch shifts slightly sharp; this was more readily apparent on the Forté than the Muse.

At the other end of the scale, it was more of a toss-up. The Muse is justly legendary for its tight, powerful bottom end, but the little 50W Forté kept right up with it. The biggest difference was in the pitch differentiation; listening to recordings with first-class bass-lines, like Gene Chandler and Jerry Butler's One On One LP (Mercury SR 61330), I became more aware of pitch relationships between the bass and other instruments with the Forté. The tightness and level of the Forté's bottom end didn't sound dramatically different from the Muse, but the Model 4 seemed to be able to better delineate low-end subtleties.

Of course, if I really pushed it on bass-head dope like the Metallica or Jimmie and Stevie Ray Vaughan's Family Style album, the Forté gave up with a crack before the Muse did, but the margin of difference was less than I expected given the power ratings of the 50W Forté and the 100W Muse. In fact, it was in this area, and this area alone, that I felt the Forté Model 6 to compete favorably with the Muse; taken in isolation, it was capable of greater headroom in the low end than either the Model 4 or the Muse. However, taken as a whole, the 6's overall sound sounded smaller than either of the other two amps'. Whether it was distortion products up the scale giving the impression of strain, I can't say. But over all, Forté's Model 6 just wasn't in the same class as their Model 4, or the Muse.

The DARA TEST

In my house, the hardest challenge a piece of gear under review faces is The Dara Test. No, I don't have recordings of my girlfriend warbling Ivanhoe, the Dara Test is whether or not Dara will come sit and listen to music with me if I'm playing something she's not too crazy about. She'll listen to Delbert McClinton on a Fisher-Price record player, but to get her to listen to stuff she doesn't like for any period of time is a trick most hi-fi gear can't turn.

We'd just come back from visiting her family in Kerrville, and I had a stack of old records that her older brother had planned to get rid of. Rather than lecture him about the superiority of the LP medium and the joys of rediscovering your old records with a topnotch analog rig, I grabbed 'em like an LA looter and ran for the car.

One of the records was the long-out-of print The Best of Donny Hathaway ( Atco SD 38-107). Side 2 includes the great, great, GREAT, GREAT live version of "The Ghetto" off the amazing, amazing, AMAZING, AMAZING

Donny Hathaway Live LP; I'd long since lost my copy of this treasure, JA keeps his in a vault at the bank, and finding the Best Ofrecord in a pile of throwaways in Kerrville, TX, not exactly a bastion of hot buttered soul, was a mind-blower! As much as I love Donny Hathaway, though, Dara's not really into his brand of slickified, highly arranged, expertly played urban cool; the rawer, simpler Stax/Volt Singles are more her thang. I worship Stax/Volt too, but Donny Hathaway Live, to me, is one of the high-water marks of soul music, and a musician's Lexicon O'Lix. I cleaned the thick layer of dust off the Best Of record, put it on the Well-Tempered, and cranked up "The Ghetto." As soon as Willie Weeks's fatback Fender bass line came swaggering out of the Forté Model 4 and the ProAcs, Dara came right in and sat down next to me.

"I thought you didn't like Donny Hath away." "I don't. But this song sounds good. So hush." Test passed.

Thomas N. MEASURES

A full set of measurements of both Forté amplifiers was made in their unbalanced mode, with selected measurements repeated for the balanced modes.

Following its 1/3-power preconditioning, the Forté 4 was warm to the touch but not as hot as I would have expected considering its class-A operation. (Unlike class-B amplifiers, a class-A amplifier runs coolest at its maximum output.) Its input impedance measured just over 49k ohms unbalanced. In balanced mode, the input impedance measured a very low 635 ohms. The Fortés be long to that small family of amplifiers which provides a balanced input impedance matching the typical impedance of a balanced pre amp output. By matching impedances, out put to input, in a line-level to line-level connection, the interconnecting cable is treated as a transmission line, allowing for maximum power transfer. There may be advantages to this approach, but the driving preamplifier (or other source) must be capable of providing significant current. In the more typical balanced arrangement-in non professional applications, at least-a high input impedance at the power amplifier limits the current flow. At least one other manufacturer, however (Jeff Rowland Design Group), is a strong advocate of matching input and output impedances in a balanced, line-level connection. The more expensive Rowland amplifiers provide a choice of low or high input impedances in the balanced mode; this is not an option in the less expensive Forté 4.

The gain of the Forté 4(8 ohm load) was 27.4dB unbalanced. In balanced mode, with a 50 ohm source impedance, the gain was 25.9dB (and thus did not exhibit the 6dB gain advantage over the unbalanced mode normally expected of a balanced link). With a 600 ohm source impedance-such as you might typically experience with many balanced output sources-the gain of the amplifier was 20.5dB. The output impedance measured under 0.05 ohms at any frequency. DC offset was negligible at 9.8mV in the left channel, 4.7mV in the right. S/N ratio (un weighted, wide-band at 1W into 8 ohms) measured 76.5dB in either mode. The 4 was noninverting, with pin 2 of the balanced input configured as the positive leg, pin 3 the negative, the AES standard.

The Forté 4's frequency response into 8 ohms via its unbalanced inputs (fig.1; the response through the balanced inputs and into other loads was a virtual overlay) was flat within the audio band, with a gradual ultrasonic rolloff reflected in the rounded leading edges of the 10kHz squarewave response (fig.2). The unbalanced crosstalk shown in fig.3 indicates some asymmetry above 1 kHz, which I would not expect to be audible given the low levels overall. There was less asymmetry through the balanced inputs. While the crosstalk signal below 1-khz in fig 3 is probably noise, the increase in crosstalk with increasing frequency is, as always, the result of capacitive coupling between channels.

Fig.4 shows the small-signal THD + noise vs frequency for the unbalanced mode These are very fine results, with no rise in the distortion at higher frequencies, or other strange anomalies The results for the balanced inputs (not shown) were only marginally different and slightly lower-all of the curves lying below 0.02% THD +noise. The waveform of the THD +noise appeared to be random noise, with a hint of third-harmonic distortion just visible Under some circumstances a 60Hz diode switching noise (visible as sharp, alternating positive and negative spikes or impulses) could be observed in the out put (but at a very low level). The spectrum of the Forté 4 reproducing 50Hz at 67W output into a 4 ohm load ( 2 / 3 rated power at that load) is shown in fig.5.

The distortion components here, well down in level, with minor exceptions decrease with increasing frequency-the second, third, and fifth harmonics (100Hz, 150Hz, and 250Hz) lie at -93dB (about 0.0025%), -87dB (about 0.0045%), and -89dB (about 0.0035%) respectively. All other harmonics are at or below -98dB (about 0.0012%). Fig.6 shows the response to a combined 19 + 20kHz signal-the intermodulation of these two frequencies-at 28W ELMS into 8 ohms (just below visible signs of clipping with this signal). The highest artifact here lies at -68dB at 21kHz (0.04%), with a similar if just slightly lower level at 18kHz. The level of the 1-khz difference frequency is at -85dB, or about 0.006%. Into a 4 ohm load at 52.5W (again, just prior to visible clipping), the distortion levels were slightly higher but still very low in level: -65.7dB at 21kHz (0.05%), marginally lower at 18kHz, and at or below -79.4dB (about 0.01%-the 17kHz figure) at all other frequencies.

The 1-khz THD +noise vs level curves for the Forté 4 are shown in fig.7. The curves are self-explanatory, with low distortion up to the curve break-point or knee and a rapid rise above that. The Forté 4, despite its modest power into 8 ohms, appears to be a solid per former into low-impedance loads, its out put almost doubling as the load impedance halves. Its discrete clipping measurements were 56.9W (17.55dBW, left) and 56W (17.5dBW, right, 113-115V line), both channels driven, and 573W (17.6cIBW, 114V line), one channel driven, into 8 ohms; 96.5W (16.8dBW, left) and 97.6W (16:9dBW, right, 112V line), both channels driven, and 108.2W (17.3dBW, 114V line), one channel driven, into 4 ohms; and 1843W (16.6dBW) into 2 ohms, one channel driven (113V line). Altogether, while I found nothing specific

Fig.1 Forté 4, frequency response at 1W into 8 ohms (right channel dashed, 0.5dB/vertical div.).

Fig. 2 Forté 4, 10kHz squarewave at 1.414V peak into 8 ohms.

Fig. 3 Forté 4, crosstalk: R-L, top at 10kHz; L-R, bottom (10d8/vertical div.). Fig.4 Forté 4, THD+noise vs frequency at, from bottom to top: 1 W into 8 ohms, 2W into 4 ohms, 4W into 2 ohms (right channel dashed). Measurement dominated by noise.

Fig. 5 Forté 4, spectrum of 50Hz waveform, DC-1 kHz, at 67W into 4 ohms. Note that the third and fifth harmonics at 150Hz and 250Hz are the highest in level at -87dB (0.0045%) and -89dB (0.0035%).

Fig. 6 Forté 4, HF intermodulation spectrum, DC-22kHz, 19+20kHz at 28W into 8 ohms (linear frequency scale)

... in the Forté 4's measurements which might explain CG's surprise at its gutsy sound (aside from its fine performance into low impedance loads), neither did I find anything that would preclude it. This is a solid per former on the test bench and, apparently, equally at home in the listening room.

The Forté 6, following its 1/3-power pre conditioning (which subjects the class-A/B output stage to maximal thermal stress), was, as expected, hotter to the touch than the Forté 4 had been. The measured input impedances, frequency responses, and voltage gains of the 6 were effectively identical to those of the Forté 4. Also like the 4, the 6 was noninverting, with pin 2 of the balanced input con figured as the positive leg, pin 3 the negative.

The Forté 6's output impedance was under 0.07 ohms at any frequency. DC offset fluctuated from 24-37 mV in the left channel and 46-50 mV in the right. These levels are quite a bit higher than normally experienced with […]

Fig. 7 Forté 4, distortion vs output power, from bottom to top at 10 W, into 8, 4, and 2 ohms.

Fig. 8 Forté 6, THD+noise vs frequency at, from bottom to top: 1W into 8 ohms, 2W into 4 ohms, 4W into 2 ohms (right channel dashed).

Fig. 9 Forte 6, 1- kHz waveform at 2W into 4 ohms (top), noise and distortion waveform with fundamental notched out (bottom).

Fig. 10 Forté 6, spectrum of 50Hz waveform, DC-1 kHz, at 147W into 4 ohms. Note that the second and third harmonics at 100Hz and 150Hz are the highest in level at -84dB (about 0.006%).

Fig. 11 Forté 6, HF intermodulation spectrum, DC-22-kHz, 19+20kHz at 87W into 8 ohms (linear frequency scale).

Fig.12 Forté 6, distortion vs output power, from bottom to top at 10W, into 8, 4, and 2 ohms.

[adapted from Nov-1992 issue]

Also see:

AUDIO RESEARCH PH1 PHONO PREAMPLIFIER (review)