Hiwatt midrange and frequency response...

[RandomPrecision]

I stumbled across this Tone Stack Calculator site. It's a pretty cool tool, if you haven't seen it. If I understand it correctly, it basically models the frequency response of the preamps of common guitar amplifiers, including Hiwatt. It also allows you to virtually adjust tone control pots, and even change the values of the RC networks that make up the tone stack, and see how that influences the frequency response.

Now, I know the sound and frequency response of a guitar amplifier is more than just the preamp. Obvious other major influences include the speaker(s) power amp, transformers; there are clearly numerous other factors as well. But for the purpose of this discussion, I'll assert the classic all other things being equal qualifier.

So my question is, do Hiwatt amps truly have a midrange dip, as suggested by the TSC website above? I feel like a lot of what I've read about Hiwatt suggests that they have a particularly strong midrange, and no midrange dip, like Fender. In particular, much of this impression is based on what I've read on Gilmourish.com, e.g. the Amps page. I don't have links handy, but I've read a lot of random forum commentary (not just here) where folks suggest the Hiwatt circuit is more "hi fi", which would imply a flat frequency response. (Maybe I should be careful with that statement, given that some hi fi enthusiasts champion equipment that decidedly does not have a flat frequency response!)

Either way, the TMB controls allow you to mostly dial out the midrange dip.

Is anyone aware of any work done where people have measured the frequency response of guitar amplifiers? Parts Express sells a speaker measurement kit that I've been tempted to buy for a while now (I also do DIY speakers and audio electronics). But I'd be really interested in seeing the actual measured frequency response of popular amplifiers, along with all the relevant details (speakers, settings, room, etc).

 

[Les H]

At what frequency range in Hz/KHz is at the bottom of the dip?

 

[telemnemonics]

Interesting question.

A point to consider when multiple parties discuss "midrange" is that AFAIK there are multiple definitions of what frequencies midrange includes.

Then when an amp is defined as having a mid dip as you say the site says the Hiwatt preamp has, is that with all knobs at noon?
Some amps like the BF Fender don't seem to add much mids with the knob while others take your head off with a twist.

The mids knob on a Marshall adds a higher frequency range than the mids knob on a SUF Muff, which adds what sounds to me like a lower mids range.

AFAIK which isn't a whole lot, the bypass cap I think it's called that limits the lowest frequencies passed to the power section has a lot to do with what we hear for low mids when we turn up the bass knob. I may have this part wrong though.

It's been a while since I played Hiwatts but I think of them as more lower mids forward and Marshall as being more upper mids forward.
But the low mids forward thing I think I recall from Hiwatt is not low mids like closer to bass, it's more low mids in the true midrange family.

I notice on some amps the mids knob adds a nasal character while other amps the mids knob adds a more open clear character.
Is that due to different frequency?
Or due to different preamp circuitry clipping creating the nasal character?
Some preamps do seem to add clipping while other do not.

 

[telemnemonics]

Not sure how to interpret these two charts that suggest Hiwatt has way less bass and more treble than Marshall, and while Marshall may have more gain it only has one Marshall chart when Marshall has many preamp circuits.

 

[joebloggs13]

Hmmmm. Interesting. I will be following this thread closely. I own a Reeves Custom 50, and agree to the HiFi/flat response statement. But i thought that was because the amp's headroom(also assumed other large headroom amps responded similarly). Cranked, it's like a hammer hitting you in the chest. I don't really know now.

 

[MuddyWolf]

Frequency response on a chart isn't sound. A guitar into a Hiwatt head into a Hiwatt cab with Fane speakers is sound. Hiwatt without Fane is like Marshall without Celestions.

 

[Obelisk]

As someone who used to play Marshall amps but switched over to Hiwatt amps, the frequency response seems relatively flat in comparison to a Marshall based on the 5F6A preamp. Since the guitar is mostly mids, having a slight dip in the mids makes some sense because that is where the guitar frequency range really lies. If an amp has too much midrange response, the result can be a really boxy guitar sound. What I like about the Hiwatt is the lack of mushy bass. That frequency chart shows that a Marshall(assuming Super Lead/2203 100 watt heads or Model 1987/2204 50 watt heads) are based on a bass preamp. When turned up loud, a Marshall will begin to lose some definition whereas a Hiwatt keeps getting louder all the way up to 10 because a tone stack of the Hiwatt isn't accentuating all of that mushy bass that a Marshall does. The tone controls actually do something on a Hiwatt that other guitar amps just don't do(like actually work at shaping & enhancing a guitar sound rather than amplifying 3 humps of 2-4 frequency ranges).

I had a friend who brought over a 72 Super Bass. Those early metal panel Marshall amps get really loud, but we were astonished that my 50 watt SA112 combo seemed louder while pushing the same 4x12 of late 70's Celestion 80. I didn't have a meter, but I would guess that the Marshall read higher in Decibels while the clarity of the Hiwatt made it seem perceptibly louder. Whatever Dave Reeves did differently with his tone stack works great for me. The problem is a Hiwatt exposes the flaws of a player. If you're a bad player, don't play Hiwatt amps.

Frequency response on a chart isn't sound. A guitar into a Hiwatt head into a Hiwatt cab with Fane speakers is sound. Hiwatt without Fane is like Marshall without Celestions.

Probably true though I would point to the Partridge transformers vs Drake & Dagnall transformers being as equally important details to the final sound. The Hiwatt also employs a 12AT7 phase inverter. Amps that use a 12AT7 in that position tend to be more hi-fi than those that use a 12AX7 phase inverter. I think every decision Dave Reeves ever made was to create the loudest & clearest guitar sound ever devised at that time. He probably chose Fane over Celestion for a number of reasons. Clarity would be certainly be #1 or 2 on that list(price being the other)

Of speakers I generally play 1x12 Thiele cabs with EVM-12L. My "Marshall" amp of today is a 68 Traynor YBA-1 that sounds like god through one of those Mesa Thiele cabs, as do my Hiwatt amps. My 79 SA112 came with a factory Celestion G12H-80. I have some Fane speakers around, but I am really hard on speakers & prefer to not to chance my love of octave dividers & ring modulators busting a set of vintage Fane speakers. I do have a 2x12 cab with a pair of mid 70's cast frame Fanes which sounds great. I use a 62 Deluxe or a WEM 30 watt PA amp to push that cab. I actually prefer to use late 70's to mid 80's Celestion G12H-80 or EVM-12L because I know them really well & they don't seem to bust.

Another point to ponder: unlike a Marshall, Hiwatt amps sound good with an array of different speakers. Marshalls definitely sound better with Celestion speakers.

 

[Mexitele Blues]

Frequency response on a chart isn't sound. A guitar into a Hiwatt head into a Hiwatt cab with Fane speakers is sound. Hiwatt without Fane is like Marshall without Celestions.

Very true. These charts are only the response of the tone filters and not the final voice of the amp.

 

[RandomPrecision]

At what frequency range in Hz/KHz is at the bottom of the dip?

About 310 Hz.

A point to consider when multiple parties discuss "midrange" is that AFAIK there are multiple definitions of what frequencies midrange includes.

Indeed! I once had some notes on what was "generally accepted" as lows, mids, and highs --- from all the authority that comes from a random post on the internet.

Then when an amp is defined as having a mid dip as you say the site says the Hiwatt preamp has, is that with all knobs at noon?

For that Tone Stack Calculator (TSC) I linked, yes, that is the default, all tone pots at the 50% position. The TSC app actually lets you adjust those knobs to see how the frequency response changes. (In fact, you can actually change fixed resistor/capacitor values to see how that affects the modeled preamp response.)

I notice on some amps the mids knob adds a nasal character while other amps the mids knob adds a more open clear character.
Is that due to different frequency?
Or due to different preamp circuitry clipping creating the nasal character?
Some preamps do seem to add clipping while other do not.

(Emphasis added). I don't know the answers to those questions, but that was kind of my hope in posting, to see if we could find some connection between theory and practice. As others pointed out, the preamp is but one piece of the system.

I wish that TSC site had a model of an EF86-based preamp, like the one in one channel of my Ceriatone Dizzy30 (Matchess DC-30 clone). To me, at low bedroom/family-friendly volumes, most amps don't sound that different. At least I thought that until I got this DC-30 clone; even at conversation-level volume, preamp at 10:00 (or even less), that ef86 channel has a really unique sound, and I haven't heard anything like it before. And this case is particularly interesting, because I can switch between two preamps (I have an A/B/Y box), which allows me to keep all other variables consistent (i.e. guitar, power amp, speaker, etc).

I had a friend who brought over a 72 Super Bass. Those early metal panel Marshall amps get really loud, but we were astonished that my 50 watt SA112 combo seemed louder while pushing the same 4x12 of late 70's Celestion 80. I didn't have a meter, but I would guess that the Marshall read higher in Decibels while the clarity of the Hiwatt made it seem perceptibly louder. Whatever Dave Reeves did differently with his tone stack works great for me.

Interesting. I was once told an anecdote with the opposite implication. This was in the context of shopping for speakers (hifi speakers, not guitar speakers). The seller told a story about how he sold one of his higher-end pairs to a customer. That customer typically liked fairly low/modest volumes. But when he got the new speakers, he turned them up so loud, he couldn't hear his phone ringing. But that volume was not intentional, he didn't even realize the volume was dramatically higher than normal until his wife came home and complained about him not answering the phone! The seller's point was that lower distortion implies higher volumes don't seem as loud. (But this is in the context of hifi sound reproduction, which is a different game than electric guitar amplification.)

The problem is a Hiwatt exposes the flaws of a player. If you're a bad player, don't play Hiwatt amps.

Repeating interenet hearsay here, but I've often seen that attributed to the power supply. Hiwatts use solid-state (diode) rectifiers, and the older Fenders and Marshalls used tube rectifiers. The tube rectifier is often associated with the "sag", which makes sense based on the increased resistance (resulting in greater voltage drop) as current demands increase. Also, bigger filtering caps after rectification can reduce the sag effect as well. You often see these massive, overbuilt power supplies in boutique hi-fi products, the idea being that you want a power supply that is as close to perfect DC as possible (i.e. 100% stable, no noise, near zero impedance, etc). But we (usually) have different goals with a guitar amplifier, as we are creating sound, rather than reproducing sound. There are no rules for creating (i.e. creativity); but for sound reproduction, perfect accuracy to the source is generally preferred.

Of speakers I generally play 1x12 Thiele cabs with EVM-12L. My "Marshall" amp of today is a 68 Traynor YBA-1 that sounds like god through one of those Mesa Thiele cabs, as do my Hiwatt amps. My 79 SA112 came with a factory Celestion G12H-80. I have some Fane speakers around, but I am really hard on speakers & prefer to not to chance my love of octave dividers & ring modulators busting a set of vintage Fane speakers. I do have a 2x12 cab with a pair of mid 70's cast frame Fanes which sounds great. I use a 62 Deluxe or a WEM 30 watt PA amp to push that cab. I actually prefer to use late 70's to mid 80's Celestion G12H-80 or EVM-12L because I know them really well & they don't seem to bust.

I've never heard (at least not knowingly) an EVM-12L, but having read about them, they are on my list to someday explore. I used to have a Traynor YBA-2 many years ago, I wish I hadn't sold it!

My "Hiwatt" is a Trinity Triwatt, that I built from the kit Trinity offers. It could be just confirmation bias, but my experience with it is consistent with most of what I've read about Hiwatts: very "punchy", immediate tone. I can see where the "hifi" description comes from (although some people use that as a ding). Very usable tone controls, great pedal platform. Definitely need to add some reverb or light delay to mask my mediocre playing!

I put mine in a 1x12 cab, and for a speaker I used the Reeves Vintage Purple, which is supposed to be a reproduction of a vintage Fane. I didn't do any comparisons, and haven't even tried this amp through other speakers. But I'm happy with the end result. (Not happy with the fact that my Triwatt is stuck over at my buddy's house (rehearsal space), and with the social distancing stuff, I can't play it!) But as you own and have experience with real Fanes, you might be interested in seeing how the Reeves compares. FWIW, I think Hi-Tone also makes vintage Fane reproduction speakers, but they were all sold out when I needed a speaker for my Triwatt.

 

[telemnemonics]

WRT the point about nasal mids being frequency based vs preamp gain based, one major point I see on those charts is that the Hiwatt preamp has seriously less gain than the Marshall preamp (whichever of 100 Marshall preamps are modeled), and I do know that the classic late Plexi early metal panel SS rectifier non master four input Marshall preamp does indeed add some but not much distortion in the gain stages.
Even among those near the same Marshall circuits though, say after '67 when they "tripled" the power section filtering, there is a lot of variation in final sound. Much of the variation is IMO power section voltages.
Where @Obelisk mentioned his friends '72 Super Bass being less loud than his Hiwatt 50, I would not call a 72 an "early metal panel", because I've had many '69-'74 metal panel Marshalls and the '69 and '70 are brutally loud and clean all the way up, while the '71 and later got more gain and are much more distorted starting at lower numbers on the vol knob. So early metal panel is distinctly '69 & '70, after which the middle years of metal panel JMP was different.
I didn't take voltage measurements on my Marshalls but I'd guess that the voltages may have been a little lower in the middle era metal panel JMPs.

A later ('73?) metal panel 50 I had that started distorting at like 2 on the volume was still louder than my friends BFTR to a point where his amp was inaudible before I turned up all the way. Then I had a '69 small box 50 that stayed clean all the way up and was a good deal louder than the dirty '73 50w. Same cab usually.

WRT sag, rectifiers and power supply filtering, Marshall scrapped the tube rec by maybe '65/ '66, but didn't increase filtering much until '68.
I had a '67 100 with the less filtered power supply but no more gain than the later preamps, which was drastically more liquidy sustaining and distorted than the several '69 & '70 100s I had. I presume that in addition to the power supply sag from little filtering (yet SS rectifier) it must have had lower voltages too. Extreme pick response where it had the classic "clean distortion" I think of the Plexi as delivering, or pick harder and it would snarl.
A player that only bashes the strings will not get any clean sounds, so those '67 and earlier Marshalls really sound different depending on the player.
Then when circa '71 Marshall "fixed" the problem of too clean sound with more preamp gain and then lower voltages, those metal panels did not have as much of a range from clean to snarling just by pick attack.

What they did have though was less of the problems with the earlier inadequate power supply filtering like ripple and ghosting, and maybe other problems like I suppose hum.
Way tighter bass too, depending on the chosen speakers.
The earliest too clean metal panels were more speaker sensitive IMO than the '71 and later, being so clean it helped to use mushy toned greenbacks to soften the brutally bright loud clean nature of those couple of years production.

Conversely, the '72- '74 sounded classic dirty Marshall even through and EV.
I'm not very fond of the 200w EVM12L but if you want to keep your amp as clean as possible it sure works with lots of power.
The earlier alnico "coffee can" EV SRO was a 70w speaker and a whole nother beast, fat and tactile.

One thing that's a mystery to me is the feel of the Hiwatts, where youmight think it would be so lacking in compression for all the clean headroom, that it would be harder to play. I find Hiwatt very comfy to play, and while not really compressed, it has just enough compression to FEEL comfortable. No clue where that mild compression comes from, certainly not the power section!
We know a tube can create compression without creating distortion, and the best tube compressors do not sound "compressed", they just sound right, not fatiguing and not harsh in the transients.

Conversely the circa 1970 Super Lead and Super Bass 100s were really fatiguing to play and to hear, with super punchy attack and ear splitting high end.
I sold a '70 Super Bass to a shop that had the truly awesome Harry Joyce Hiwatt 100, and those two amps were just night and day different, with the Marshall more like what those who think Hiwatt is too hi fi complain about.
I'd bet in a double blind test, most players would get them backwards and think the punchy clean Super Bass was the Hiwatt, and the fat sweet comfy Hiwatt must the the "less hi fi" Marshall.

 

[joebloggs13]

Interesting comments about speaker choices on the Hiwatt. When I ordered my Reeves Custom 50, Bill was trying out a new speaker to replace the Purples he was using. I decided to go with Celestion Creambacks in my 2x12 cab. While they sound great I am curious as to how the Purples, or any other Fane clone speaker would compare. Would it be a drastic difference?

 

[Mike Eskimo]

 

[kLyon]

Interesting, related: about 35 years ago, Howard Dumble (now Alex) told me that Hiwatt was the only amp he respected.

 

[clydethecat]

As someone who used to play Marshall amps but switched over to Hiwatt amps, the frequency response seems relatively flat in comparison to a Marshall based on the 5F6A preamp. Since the guitar is mostly mids, having a slight dip in the mids makes some sense because that is where the guitar frequency range really lies. If an amp has too much midrange response, the result can be a really boxy guitar sound. ...

Also, since tone controls are passive and lossy and cannot actually boost, a mid dip can give the impression of more bass and more treble.

 

[igor5]

WRT the point about nasal mids being frequency based vs preamp gain based, one major point I see on those charts is that the Hiwatt preamp has seriously less gain than the Marshall preamp (whichever of 100 Marshall preamps are modeled), and I do know that the classic late Plexi early metal panel SS rectifier non master four input Marshall preamp does indeed add some but not much distortion in the gain stages.
Even among those near the same Marshall circuits though, say after '67 when they "tripled" the power section filtering, there is a lot of variation in final sound. Much of the variation is IMO power section voltages.
Where @Obelisk mentioned his friends '72 Super Bass being less loud than his Hiwatt 50, I would not call a 72 an "early metal panel", because I've had many '69-'74 metal panel Marshalls and the '69 and '70 are brutally loud and clean all the way up, while the '71 and later got more gain and are much more distorted starting at lower numbers on the vol knob. So early metal panel is distinctly '69 & '70, after which the middle years of metal panel JMP was different.
I didn't take voltage measurements on my Marshalls but I'd guess that the voltages may have been a little lower in the middle era metal panel JMPs.

A later ('73?) metal panel 50 I had that started distorting at like 2 on the volume was still louder than my friends BFTR to a point where his amp was inaudible before I turned up all the way. Then I had a '69 small box 50 that stayed clean all the way up and was a good deal louder than the dirty '73 50w. Same cab usually.

WRT sag, rectifiers and power supply filtering, Marshall scrapped the tube rec by maybe '65/ '66, but didn't increase filtering much until '68.
I had a '67 100 with the less filtered power supply but no more gain than the later preamps, which was drastically more liquidy sustaining and distorted than the several '69 & '70 100s I had. I presume that in addition to the power supply sag from little filtering (yet SS rectifier) it must have had lower voltages too. Extreme pick response where it had the classic "clean distortion" I think of the Plexi as delivering, or pick harder and it would snarl.
A player that only bashes the strings will not get any clean sounds, so those '67 and earlier Marshalls really sound different depending on the player.
Then when circa '71 Marshall "fixed" the problem of too clean sound with more preamp gain and then lower voltages, those metal panels did not have as much of a range from clean to snarling just by pick attack.

What they did have though was less of the problems with the earlier inadequate power supply filtering like ripple and ghosting, and maybe other problems like I suppose hum.
Way tighter bass too, depending on the chosen speakers.
The earliest too clean metal panels were more speaker sensitive IMO than the '71 and later, being so clean it helped to use mushy toned greenbacks to soften the brutally bright loud clean nature of those couple of years production.

Conversely, the '72- '74 sounded classic dirty Marshall even through and EV.
I'm not very fond of the 200w EVM12L but if you want to keep your amp as clean as possible it sure works with lots of power.
The earlier alnico "coffee can" EV SRO was a 70w speaker and a whole nother beast, fat and tactile.

One thing that's a mystery to me is the feel of the Hiwatts, where youmight think it would be so lacking in compression for all the clean headroom, that it would be harder to play. I find Hiwatt very comfy to play, and while not really compressed, it has just enough compression to FEEL comfortable. No clue where that mild compression comes from, certainly not the power section!
We know a tube can create compression without creating distortion, and the best tube compressors do not sound "compressed", they just sound right, not fatiguing and not harsh in the transients.

Conversely the circa 1970 Super Lead and Super Bass 100s were really fatiguing to play and to hear, with super punchy attack and ear splitting high end.
I sold a '70 Super Bass to a shop that had the truly awesome Harry Joyce Hiwatt 100, and those two amps were just night and day different, with the Marshall more like what those who think Hiwatt is too hi fi complain about.
I'd bet in a double blind test, most players would get them backwards and think the punchy clean Super Bass was the Hiwatt, and the fat sweet comfy Hiwatt must the the "less hi fi" Marshall.

Great points

 

[erratick]

Maybe I should be careful with that statement, given that some hi fi enthusiasts champion equipment that decidedly does not have a flat frequency response!)

It's a cool site. Really like that kind of thing. Measuring guitar amps/tone stacks and sounds has come a long with modeling and IR type computer analysis.


Although I think you said relatively flat response, but I want to caution anyone thinking guitar amps have "flat response" and even what flat response is... because flat frequency response is not tube guitar amps generally.

Even a layman can tell if they play some music they know well (even from a phone or whatever) through a guitar amp or a bass amp. It doesn't sound like playing it through speakers or headphones- it's very coloured.

But also there is the fletcher munson curve- so how loud you play affects the way your ears/brain perceive the volume at certain frequencies.

So sounding flat is not the same as flat response-

What is the Fletcher Munson Curve? Using Equal Loudness Curves in Mixing and Mastering

What are the Fletcher Munson curves, or equal-loudness contours, and how do they apply to audio mixing or mastering? This article will cover that all in detail.

www.izotope.com

But if you think after playing music through a tube amp for guitar or bass that it's flat, then your hearing is probably suspect.

 

[dom ray mustang]

Marshalls had mostly bright caps over the bright volume channel pot, as in comparison Hiwatts had rarely bright caps. A big influence, which makes a direct comparison solely based on the TSC graphs of both tone stacks not really possible (as depending on volume and the value of the bright cap) to draw qny conclusion (beyond all other differences in the circuit and parts such as transformers, speakers, as mentioned in posts earlier).

 

[edvard]

In general, all FMV tone stacks have an inherent mid-dip, it just depends on where it is that creates the toanz.
@telemnemonics is correct that the final judgment is "it depends on the circuit", and from what I've been able to gather, the Hi-Fi world seems to center on 1000Hz being the ideal midrange, while the guy from Elliott Sound Products thinks true midrange is 630Hz, and the classic Tube Screamer mid-bump is right around 450Hz (IIRC).

YMMV, contents may have settled during shipping, take as directed.

 

[hamerfan]

@telemnemonics thanks for the charts

Funny that the Marshall has it dip where the Tubescreamer has it bump: at 700Hz

 

[Dacious]

And remember - the preamp may have a peak here or a dip there but it's nothing like as dramatic as a speaker response curve and that is even before considering how that might be altered by the speaker cabinet/enclosure