The note changes interval to interval. The same interval has the same 3rd note on different guitars.

Okay, but which one should I buy? None of the pedals I found is so short on general compressor artifacts than the Keeley. And it is very player friendly, apart from the intermodulation. I have tried the Distressor Plugin as well, couldn't set it so well like the Keeley. So this is both about challenge to use and what features it offers. I have ideas which one should I buy, some are a bit pricey, but why I am not rushing to buy them is because the ones I bought so far had also been praised by many professional musicians and users (the Demeter or the Carl Martin), or the maker advertised itself as a group of geniuses. And then when you got the product the story was no more that happy.

 

Use a Different Compressor Topology (Most Effective)
If you need clean interval playing high on the neck, OTA compressors are simply not ideal.

Better options: - Optical Compressors -> Much smoother, less IMD. Examples:

- Diamond Compressor --> www.thomann.de/de/diamond_guitar_compressor_eq.htm
Diamond Guitar Compressor EQ - $239.00 - --> www.diamondpedals.com/products/comp-eq

Add a narrow EQ notch at the offending IMD frequency after the compressor.

A parametric or digital EQ in a loop can put a narrow cut around ~540–550 Hz just after the compressor. If the offending product stays roughly in that area when you use those intervals, this can make it much less audible without affecting the main notes too much.

 

Last edited: Feb 24, 2026 at 9:47 AM

Having 2 optical compressors they are not the first that come to my mind. The settings on the ones I have are either problematic due to internal "presets", or the rate of compression is insufficient. Others may behave differently though. Right now I am thinking about trying the Strymon compadre. I watch its video before, the presenter mentions the same problem with compressors and its usual artifacts (not the intermodulation) that I also noticed, and he says they made an effort to diminish those as much as possible. So that makes it interesting for me, and the report of Grok AI about the intermodulation problem mentions the Strymon might be a suitable choice for me.

The parametric EQ: the problem is that the 3rd notes cover a longer range, and depending on various factors that may arise on smaller fine tunings of the compressor that are required for different recordings, the range can even get wider. This photo show it, this is at one of my recent settings:



This is well in the range of the notes I also play sometimes. I can't tell whether they are conflicting or not, because right now the 3rd notes are a bit off the chromatic scale, but my intonation is also off at the stem of the neck, so I couldn't draw a proper conclusion about this at the moment.

However, I have experimented with the Carl Martin compressor that I have. I don't like it because of the available settings, but I could set it in a way that is not as good as the Keelely, but could be a compromise, that I can use until I find a better solution. The only big problem is that there is no tone knob. And this way the signal coming out of it is very bass dominant. I have been thinking about grabbing the Tone circuit from the Keeley circuit, build it and add that after the Carl Martin. I am just not sure what parts I need. Can you help in that to determine which elements should be copied and if I need to add something what to add?

 

I know you from previous posts. Resolving this issue could take weeks; they've now focused on it and are delving into the nuances that guitarists in bands like Deep Purple don't even notice. I would check out what the big names use. If you then buy the same product, you can't go far wrong.

Steve Morse --> https://equipboard.com/pros/steve-morse
(By the way, he also uses the "Keeley Compressor 2-Knob")

P.S. Trust your ears and don't fall into the abyss of digital displays.

 

This is not the effective way in my case. And I can tell you exactly why. I have been thinking about why noone seems to have noticed or heard of the problems I presented. And I think the answer is because most of the users don't play the same music that I or my friends play. In the style we play, most of the time we play between the 12-24th frets, and sometimes even higher, so we need to tune the guitar up with 1 whole note. And we use clean signal, and mostly only delay and chorus, but sometimes only the clean signal. The style involves playing lots of intervals and chords. This approach is not very common among heavy metal guitar users. But me and friends and their predecessors often do it, one reason is because the superstrat and some other shapes like the Steinberger or the B.C Rich allows for easy reach of strings for intervals and chords near the 24th fret.

So I think you see now how the two things come together. Near the 24th fret this problem appears as a serious disturbing factor.

Some players use single coil guitars. Now that story is still need to be investigated, because when I tried single coil guitars with the problem pedals, they did not seem to have any distortion or 3rd note. Now as the source of problem is known, I will borrow some and recheck, because the intermodulation should appear on those as well. So maybe I missed something or the are more to this problem that is still need to be investigated... And, if the single coils don't have this issue, those players who use those guitar will never have this problem. One more reason for the problem stay undetected. To further complicate this, I once tried an Alnico 5 guitar as well, it sounded clean. The only problem is that I don't like that kind of sound that pickup makes.

So I think it is understandable from this why I don't really look for info on pedal choice from other bands outside of the style. From inside I do. The problem is many of them don't share this info with you, many otherd do, but don't use compressors. Another side of the story though is that I had this problem with other types of pedals, too, like delay and chorus. The only problem is I sold them about 10 years ago, and as it turned out the distortion problem is 2-fold (clipping on input and intermodulation), I cannot verify now which one was present on those.

Then there are the multieffects, which lack the intermodulation problem. Every player who use them won't have the problem I have. One more reason for the undetection.

On the other hand, when you ask have I heard about the problem or not, now as we know what's happening in the pedal, I have to say yes. Because by knowing the problem know, I can interpret correctly what was reported to me earlier or what I experienced. For example I noticed during recordings that amps do a strange distortion that they should not have, and when I asked the engineers why that distortion is there when the input gain is okay and the speaker cone is not torn, they couldn't answer it. Now I know why. That was exactly this intermodulation sound. I can realize that know when I recall that sound from my memories. Also, players talked about sometimes that there are distortions between the B-E strings when playing intervals, and to diminish that, they recommended to use +1 gauge string for the E string. So if you have a .10 set, use .011 for the high E string. Etc. I tried it, when I tired it I only had the multieffect, and I had thought before that I need to listen to the harmonic distortion or the beating, so, as I was not informed about the real problem I don't know if it helped or not. Maybe not, because I think even now the E1 string is a.12 on my .011 set. Or not. I will test it, but i don't think it won't help, as the problem also happens between the G-B strings, and sometimes between the D-G as well.

 

Somehow, they haven't really analyzed the problem properly yet.

As usual with you, all problems suddenly appear. It could be this, or that. Please find a real pro like Gary Moore who can help you. Write to the guitar or effects manufacturers; I've seen that you've already discussed this problem in other forums.

Believe me, what you're playing isn't anything extraordinary, and with guitars, almost everything is known.
After all, it's the most played instrument in the world.

What kind of guitar do you have? Count the frets again. I recommend you have your guitar professionally serviced. What strings do you use and what gauge? What amplifier do you use? How do you store your guitar, on a guitar stand? How long have you been playing guitar?

 

Last edited: Feb 24, 2026 at 6:32 PM

I have been playing for 27 years and believe me I did research on this problem empirically extensively. It is not a matter of setting tweaks whether it appears or not, especially if the settings you go for tend to strenghten it. It just disguises the problem, but doesn't solve it.

And it is not like the problem suddenly appears. If it seems like that, it is because before I did not know what this problem is. Then, when me and my helpers started to investigate it we thought it is input signal clipping. Experienced technicians thought that. When that was solved, for a while it seemed the problem is gone, then I found not. Nobody knew why. 5mV input signal shouldn't cause problems in any pedals. Yet it was there. So yes, until you are not aware of the real problem, especially when you are doing the research as a total beginner in electronics, the problem may seem to come and go. Especially when the problem challenges pros as well. When we were in the solve-the-clipping stage, some pros were even agitating against it, trying to make it seem if I was stupid, insisting on setting tweaks instead saying guitars have no too strong signals. Then I had a chance to measure the output Vpp of guitars, and VOILA, the humbuckers I tested had a signal way over the pedal limits. Before that everybody came with what you come with now. Check this, check that.They did not consider that I have been checking that for years. And now as it revealed that even guitars like the Schecter has too strong signal, they just don't say anything, just behave as if they knew it all the time.

And even now when we know the real problem, it reveals that at a certain EQ settings the problem is stronger, at others it diminishes, and there are still questions arising. It is not because of me but because the type of the problem. When you play solo it doesn't appear. When you play certain intervals it appears very audibly, other times hides in the background as noise. It does appear differently even with the guitars I have here. On my guitar the 3rd note is always distorted, on the Schecter it is very often clean, sometime it goes upper than the 20th fret, sometimes not. So you see this is not an easy to catch problem, but thanks to Spectralayers and the fact that the Carl Martin doesn't do it, it could be found what causes the real problem, and this helped to understand everything in this come and go mess.

If you blame the approach of mine you need to blame the pros as well because you can see for a year they pushed me into the direction to consider it a clipping issue, and tackle that, even when I shared lots of samples, and the intermodulation was there in the samples. Yet even experienced people either had no idea what that was or thought it is clipping. So if the things with me seem to suddenly appear and suddenly not, this is because this is a challenging problem even for pros. You have explained the problem this time on the previous page, for that I am very thankful. But if you blame me why I am not doing this or that, or that when it comes to me the problems suddely appear and go, so then, let me ask you: where were you with the explanations when I asked about this problem here previously? Even then lots of comments arrived, everybody taking the stand of smart and pro, but the answers were unrelated or trolling. Where were the not-appearing-suddenly-people then?

And think of that fact as well, that I asked lot of pros about it during the years, were in studios, tried many quality guitars, and noone could tell what it is. No engineers, no makers of the pedals. So what do you expect from a beginner then? When the maker doesn't know it. And doesn't even inform you. Because until I asked them about the pedal input Vpp, no maker had the idea that it should be matched. Others didn't even reply. Things to not appear suddenly? Come on....

Also keep in mind, my approach may not be perfect, but it was my approach and endurance that lead to the answer in the end, and it will be thanks to this approach from now on, that if I talk about this problem with pros who were not aware of this, they will know. Because there are many people like that in the music industry. And I am not blaming them, but it is a fact. So don't come with this suddenly stuff. If you don't want things appear suddenly, then consider the title of this discussion, and inform me how to tweak the Keeley to remove the problem from it. If you don't know I don't mind if you comment but don come with this suddenly stuff.

 

Thanks for your detailed report!

Normally, you have to reproduce and document a fault, since anything electronic can be affected by all sorts of interference. That's why you work with the process of elimination.

From a neighbor's baby monitor to electromagnetic interference from space, you often end up checking or replacing all the components to narrow down the problem.

If problems occur, I would test the guitar for intonation, adjust the neck, check the fret spacing, see if the frets are too worn, replace the saddle, test the bridge, test the nut, and test the tuning pegs. Try different strings. That's how I would test my guitar.

If I understand correctly, you also tested your pedals with another guitar.

The amp and effects pedals should be measured to see if there are any fluctuations. If I remember correctly, we talked about writing down the specifications of the equipment and doing a visual inspection. For electronic components, only a radio and television technician or a music store with a repair shop can help.

Sometimes you search for a long time before you either give up or simply buy different equipment.
I can't say much more, but I wish you all the best on your journey to trouble-free music.

Intermodulation (IM) is a non-harmonic distortion of at least two tones transmitted simultaneously.

This creates sum and difference tones which, because they are not in an even ratio to the desired signal, are perceived as dissonant and therefore highly disturbing. The IM factor is the ratio of these distortions to the desired signal. IM distortion can occur, for example, in a loudspeaker driver that has left the linear range of its motor. If a transducer has to transmit a 50 Hz and a 500 Hz tone simultaneously, and the bass is "cut off" because the diaphragm excursion becomes too large, the superposition of the two waves can produce new, disturbing harmonics (at frequencies of 450/550 Hz).

 

Last edited: Feb 24, 2026 at 9:14 PM

TIM or intermodulation distortion

Transient Intermodulation (TIM) - Intermodulation distortion

The question of whether negative feedback is detrimental is frequently discussed in connection with transistor amplifiers. Various publications, particularly certain hi-fi magazines, occasionally make negative feedback the scapegoat for all sonic shortcomings.
The topic of negative feedback has already been discussed in detail in previous articles ("Audio Academy" Topics 1 to 4). This article will explore the question of whether and why negative feedback is harmful in hi-fi amplifiers from a different perspective.
If negative feedback is bad, why isn't it in operational amplifier circuits?

And what exactly is "TIM"? And what does TIM have to do with negative feedback?

TIM stands for "Transient Intermodulation," also commonly referred to as "intermodulation distortion" in German. No one doubts that this distortion exists, but hardly anyone knows how it arises or whether and how it can be prevented.

Intermodulation distortion (TIM) is interestingly only observed in transistor amplifiers, and rarely in tube circuits; however, only when these are designed as emitter followers or cathode followers. In transistor amplifiers, the emitter follower is standard, while in tube amplifiers – where it is typically a cathode follower – it is rather the exception. In tube technology, the cathode follower generally serves as an impedance converter. The voltage gain is always less than 1, while the current gain is often more than 100.

If an emitter follower is used as an impedance converter, it works without problems. The crucial point is that the load resistance must not be a complex resistance, but rather a purely resistive load across the required frequency spectrum. This requirement cannot be met when an emitter follower amplifier is driven by a loudspeaker. However, transistor amplifiers almost exclusively use this circuit design to drive loudspeakers.

Structure and function of an issuer follower

In an emitter follower, the emitter follows the base. Hence the name. But why does it do that?

Let's consider an NPN transistor. In an existing circuit, the transistor's conductivity increases as the base becomes more positive. This increased conductivity results in a higher emitter current. This current flows back to the generator and collector via the load resistor, causing a voltage drop across that resistor. This means that the emitter side of the connected resistor follows the base. Hence this designation.

To explain it figuratively: the electrons at the input of the resistor accumulate until, due to the backflow into the transistor, the transistor's conductivity decreases and the current through the resistor is balanced by the resulting emitter voltage. This corresponds exactly to Ohm's law.

But what if the emitter current is phase-shifted relative to the base due to a complex load resistance?

If this phase shift is inductive, initially no current flows into the resistor, and the transistor does not appear to improve its conductivity, even though the base becomes more positive. In fact, the resistance now appears higher, and with a higher resistance, a smaller current is sufficient for the same voltage drop. This is indeed referred to as impedance. When the correct current finally wants to flow, the base becomes less positive again, and the phase-shifted current essentially draws current from the emitter. This also makes the emitter less positive. Therefore, under complex loads (such as loudspeakers), the emitter only partially follows the base. The output transistors of an amplifier in an emitter follower configuration thus have current-controlled negative feedback. This current is usually phase-shifted relative to the control voltage at the base for a certain period during a given amplitude.

The emitter voltage, i.e., the output voltage, is directly dependent on the emitter current. This is the principle of the emitter follower. The voltage gain is always less than 1, but it decreases further with increasing current over a given amplitude due to the increasing base-emitter voltage. And because this current gain curve is not a straight line, but a curve of arbitrary shape, non-linear deviations from the target value also occur. These deviations depend on the instantaneous value of the current during a given amplitude. Additionally, the current and the phase shift dependent on it can change several times during a given amplitude.

This circuit doesn't really live up to its name, because the emitter only follows the base with a deviation. And this deviation is by no means constant. If this voltage, which depends on the complex load, is used for negative feedback, the output voltage of such an amplifier cannot be amplified linearly relative to the input voltage.

It's important to remember that the phase-shifted emitter current introduces a time delay between the input and feedback voltages. Between the emitters and bases of the output transistors, voltages with opposite directions of change are present for a limited time. This inevitably leads to distortion.

These distortions are temporary within an amplitude; they are therefore “transient”.

In principle, a transistor can be controlled not only at its base but also at its emitter. This is precisely what is exploited in emitter followers.

There are many theories about TIM.
And here, TIM is most likely explained accurately…

The distortions known as TIM are caused by the phase-shifted current feedback of the output transistors in an emitter follower circuit. These distortions have no effect as long as such an amplifier is operated without overall negative feedback.

Because transistors are inherently nonlinear components, it is virtually impossible to construct an amplifier that operates linearly without negative feedback. Without negative feedback, an acceptable damping factor cannot be achieved. While each individual amplifier stage can be subjected to its own negative feedback loop, TIM (transient interference) inevitably reappears in the output stage.

To improve all parameters of an amplifier, the installation of strong overall negative feedback is recommended. This also improves the damping factor.

The overall negative feedback circuit attempts to keep the amplifier's output voltage proportional to the input voltage and controls the output transistors accordingly. However, these transistors are already subject to negative feedback through the phase-shifted emitter current. Only now is TIM (transistor interference) generated and has an effect, because a mixture of two phase-inverted negative feedback loops disrupts the amplifier's operation. This circuit produces the often-lamented "transistor sound," and indeed, commercially available transistor amplifiers—albeit to varying degrees—always exhibit this characteristic sound.

Is it possible to reduce TIM…?

Often, attempts are made to design an amplifier without negative feedback. Such amplifiers exhibit unacceptable distortion and a very low damping factor.

Or the amplifier is made particularly "fast" so that it could also function as a radio transmitter. It is then equipped with strong negative feedback. This negative feedback could correct the distortion caused by the emitter follower. TIM only becomes noticeable when this distortion is not corrected and lies within the audio frequency spectrum. Such an amplifier often exhibits low total harmonic distortion (THD) and a high damping factor. Whether and to what extent TIM is still noticeable in this case cannot be generally assessed. However, building such amplifiers is very complex and correspondingly expensive.

… or avoid TIM?

To avoid TIM, an amplifier must be designed so that emitter follower distortion doesn't occur at its output in the first place, because it isn't an emitter follower at all. Such a circuit can have arbitrarily strong negative feedback. This results in very low total harmonic distortion (THD) and eliminates TIM entirely. Furthermore, this amplifier has a very high damping factor.

The hi-fi world is constantly amazed by impressive measurements. Amplifiers boast top values for distortion and TIM – often below the measurement threshold. However, measurements are still generally taken using purely resistive loads. TIM and distortion only arise with complex loads, such as when driving a loudspeaker.

TIM, or intermodulation distortion, was the topic here. And we keep getting asked why on earth negative feedback is supposed to be bad. Designers and journalists apparently accept this without question. But it isn't! That should have been cleared up by now.

Matti Otala, a professor in Finland, made a name for himself as an authority on hi-fi in the 1970s. Like everyone at the time, he favored transistors and worked in the world of emitter follower amplifiers. It's quite possible that the demonization of negative feedback originated with him. He was very actively involved with TIM (Transient Intermodulation) at the time – he may even have "invented" it. However, Otala wasn't just a theoretical thinker; he also built amplifiers. Very good ones for their time, in fact – and, consequently, (almost) without negative feedback.

ABACUS took its inspiration from vacuum tubes, but, like Otala, prefers transistors. According to "Der Spiegel" in 1985, ABACUS then turns these transistors on their head. It's that simple! For those who find that too simplistic, let us reveal that the ABACUS has no TIM (transistor induction loop), virtually undetectable distortion, and 100% negative feedback – all measured with loudspeakers connected.

September 2009 / May 2017 - khs

Source: www.abacus-electronics.de/info-service/informationen/audio-akademie/tim-oder-intermodulationsverzerrungen.html