I think some theory , gives an insight on what ''should'' work. When combined with experimenting to find out if any theory is proven or is disproved, you can go back with that experience to try and figure out why and what went right, or what went wrong. That is how you can build your knowledge base and experience. Amps sometimes had a low and high Z input, audio mixing boards still do. Variable Z is not needed, too expensive and basically impractical. You don't really need to dial in a magic impedance number because it would change from plugging one instrument then to another instrument. Guitarist plugs guitar in, then they boost and cut on the tone and preamp levels to compensate. But that is also why some techies play around with overwinding pickups to get a hotter sound for a given amp, it may sound thicker or thinner depending on where the Q goes. So there you have the pickup impedance being ''adjusted'' to get a particular sound on a fixed input amp impedance. Yes it is complex, but it is not ''unobtainable'' understanding to get results.
Yeh, I should have said, "no amount of electronic theory alone, without an understanding of what musicians are trying to achieve, will help..."
Sometimes it's a lack of theory that gets in the way. Apparently they don't teach much 'audio' in school these days. My father's Physics textbook from the '30s had a whole section on sound. Mine didn't. They needed the space to talk about radios and such. These days, texts are probably full of digital stuff.
Take the tone control, for instance. Way too many think it's an RC filter. It's really sort of a Zobel network. You can't understand the tone control by isolating it from the pickup. I often suggest people try to understand guitar controls without the concept of "ground". After all, if you unplug the guitar from the amp, there's no connection to ground yet there's still a voltage across the output if you strum the strings. Yet many seem to have this idea that "ground" is like a wastebasket, a black hole where things go never to return. The tone control is connected across the pickup's coil. It doesn't "bleed highs to ground" as is often stated - it shorts out the coil so there's no voltage difference across it, hence no current flows. In other words, no highs are produced when the tone control is on '0'.
Of course, the RC filter idea doesn't explain why there's more output with the tone control on '0' than when it's on '2', but you wouldn't know that happens if you don't play guitar.
Last edited:
Almost none, sadly. They mostly give you a general grasp of the basics.Very little practical theory! Maybe I was lucky because my musical interest preceded and was a big factor in leading me into electronics. Lucky again because my basic electronics education got me working in industry and kept hobbying in musical stuff. There are some very good practical reference books , once the basics are understood. The articles by NP and others appearing here on DIY Audio are really helpful and insightful too. You are right about the 30's , the golden age of radio..sound was THE entertainment of the day. If you find anyone left who was a tech or an engineer from that era, they know about sound! tubes, transformers, speakers, mics and maybe instruments too. They are a goldmine of practical information.What is obvious to them, would probably have a modern engineer's head spinning. We are going to miss them.
As the input impedance required is directly related to the output impedance of the source, then there's no 'stretch' required at all.
Active guitars/basses have a low output impedance, and thus are happy with a fairly low input impedance - so why would you be saying "I'd double check that"?.
Passive instruments require a high input impedance amp (at least 1M and preferably more). passive ones only require a far lower impedance, the 'standard' line level input impedance of 47K is more than enough.
An active Tele pickup from EMG has an output impedance of 10K ohms.
A typical passive Tele pickup is about 7K ohms.
Right off the bat, your statement that active instruments have a "low output impedance" comes in to question. Low compared to what? Obviously, not low compared to a passive pickup.
As you suggest, the 10K active pickup won't notice a 47K ohm input impedance, but the 7K passive pickup will lose it's "twang" with less than the standard Fender 1 Meg input. So, about this "directly related" impedance requirement thing you're talking about.....
You're comparing apples to oranges, really. In one case we have to worry about properly loading the pickup's coil, in the other case it's an amp to amp connection, which is much more forgiving. Nothing to do with the actual impedance, everything to do with the type of device under load.
A typical passive Tele pickup is about 7K ohms.
Right off the bat, your statement that active instruments have a "low output impedance" comes in to question. Low compared to what? Obviously, not low compared to a passive pickup.
As you suggest, the 10K active pickup won't notice a 47K ohm input impedance, but the 7K passive pickup will lose it's "twang" with less than the standard Fender 1 Meg input. So, about this "directly related" impedance requirement thing you're talking about.....
You're comparing apples to oranges, really. In one case we have to worry about properly loading the pickup's coil, in the other case it's an amp to amp connection, which is much more forgiving. Nothing to do with the actual impedance, everything to do with the type of device under load.
Hi Fezzle, As you had mentioned you are thinking of doing this for a bass guitar..I guess you may not be concerned about loss of twang. You are looking for a particular response from the DI transformer. One thing under discussion is amp input impedance which is a resistive/capacitive load. The DI transformer input is inductive/resistive (and bit of Cap).
I got to thinking that for a bass guitar this could be actually quite good for tone response with a reactive load of the DI transformer. In this case with passive pickup to transformer you are actually reinforcing the response at the low end frequencies.
When I said that those types of transformers typically are at max 100K range here is a trick that you can look at to play around with varying the impedance and response to your taste. If you have a 5 way pickup switch on your guitar, you know that you can play around with selecting 2 pickup coils in parallel, or series (phased in or out) for different sounds. You can do the exactly the same with 2 transformers. Series connected primaries will give you double the input impedance, parallel will drop it to half, and same goes for the secondary sides.
The picture above is showing a typical 5 position rotary pickup switch ( like on a PRS style)
I will just make a quick drawing to show what I am getting at.
I got to thinking that for a bass guitar this could be actually quite good for tone response with a reactive load of the DI transformer. In this case with passive pickup to transformer you are actually reinforcing the response at the low end frequencies.
When I said that those types of transformers typically are at max 100K range here is a trick that you can look at to play around with varying the impedance and response to your taste. If you have a 5 way pickup switch on your guitar, you know that you can play around with selecting 2 pickup coils in parallel, or series (phased in or out) for different sounds. You can do the exactly the same with 2 transformers. Series connected primaries will give you double the input impedance, parallel will drop it to half, and same goes for the secondary sides.
An externally hosted image should be here but it was not working when we last tested it.
The picture above is showing a typical 5 position rotary pickup switch ( like on a PRS style)
I will just make a quick drawing to show what I am getting at.
I don't believe a passive pickup is a simple 7K only, if it was (and assuming no controls) then a 47K input impedance would be pretty well fine. In any case, as I mentioned long ago - the controls mean the amp has to provide a high impedance in any case.
Active guitars are also likely to be considerably lower than 10K.
The quoted text pretty much describes most every amp on Earth. Hardly an absurd statement. Guitar/bass amp input impedance is pretty much just the grid leak resistor on a preamp tube. Typically it is 1M. Try using a smaller one and it really does mess with the tone/volume controls on the guitar.
Right. And thus the values for the control pots in guitars and basses were chosen to work with a 1 Meg input. You can't turn it around and claim the amp's input depends on the control values. That's like trying to buy just the right sized steak to work with a pre-determined pinch of salt.
I don't think you guys have wound your own pickups. I don't think you've built guitar amps. I don't think you've even tried different pot values in guitars.
I do think you're presuming and postulating based on inexperience.
Actually isnt it really more akin to a parallel LCR 'tank' with variable filter Q? After all the C used on the tone pot, effects things just as much.
A little aside:
Nigel and Keriiwena you both need to cool off.
I have had a similar problem, and I think the both of you have valid points.
I AM an electrical engineer, but to be honest I cant be botherd to calc the math to sort MY impedance issues right now(baby due, full time work and a degree in progress, plus DM amp refurb, speaker builds the list goes on)
I use a pedal board(Vox tone lab) o/p Z is 22k, connected to DM 120W valve head (i/p Z 1M).
According to Nigels posts, this should work fine.
IT DOESNT. Thin, no bass, and harsh tone. DI'd the pedalboard sounds very very good, straight in with NOFX, the amp sounds amazing.
I have played about with potential dividers to no avail, (and which I thought SHOULD work) and I believe matching t/x IS required in this case.
That is, unless someone can tell me im wrong AND why.
P.S. the typical line output Z is more like 10k - 20k in my experience, and either 1V p-p or 2V p-p depending upon app(home/PA) 47k is higher than anything ive ever come across for a true line output. Also since the typical pickup is somewhere between 5k and 15k and at its 'open' position the output of my Gibson at does not see anything but the minimum pot R... i.e. ~0 Ohm
Active is another thing altogether, and I dont use it so I know nothing of the impedances involved in that case.
Last edited:
unless in Hifi, IME at least, where it *used* to be recommended that i/p and o/p should be matched fairly well, i.e.2:1 inputZ : outputZ and IIRC anything above 10:1 gets a bit hairy, especially is there is decoupling caps at the input/output
I'm an electronics service engineer, and I can absolutely assure you that HiFi has NEVER used impedance matching. The 47K value I mentioned above is the pretty usual line input impedance on HiFi gear - with line output impedances probably 1K-2k - PA gear of course is most commonly nominally 600 ohm output impedance.
Interestingly, I happen to have to hand a PDF of the manual for the old Leak Stereo 70 HiFi amp (which I used to own years ago). The line inputs are 47K, but the line output isn't listed in the spec - however, it's fed directly from the emitter of an emitter follower (via an electrolytic) so it's going to be something in the hundreds of ohms region.
I haven't wound pickups, but I have built guitar amps - and while I've designed controls for guitars (for an old friend who makes his own), I've never tried low values in one - it would be pretty pointless.
Your error in the matter is 'believing' a pickup is 7K impedance - assuming it IS specified as that, that will only be at one specific frequency (usually 1000Hz is used for measurements). At higher frequencies the impedance will greatly increase (look up inductive reactance), if it's 7K at 1KHz, then it will be approximately 70K at 10KHz. Hence the need for high input impedances, and the 'dull' sound if you use too low a one. Some bass amps have adjustable input impedance, to dull the sound for specific effects.
never used Z matching? no correct. not in the sense we're talking about, BUT most separates Ive come across, just 'conveniently' have o/p's of around 20k. I believe it harks back to the golden days of radio, (and even still in the 70s) where there was no continuity in Impedance between separates from different manufacturers. In those days it REALLY mattered exactly WHAT radio/turntable one paired with a particular amplifier, since the sound varied so much. These days its a standardized 'rule of thumb' that in most cases produces an optimal coupling of devices.
Im not into arguing but I can tell you right now that NONE of my hifi equipment has 47k input Z, my Alesis PA mixer DOES however.
Unless im doing something wrong with the DMM, Cambridge Audio A1, CD input Z 250k; CD player output Z 22k
I think the real problem here is the miss-use of the term 'LINE input'. A true line input may well be 47k for a 2V p-p device, however for a 1V p-p( NOT a true 'line' input), Im pretty sure its 22k. then in order to use the same preamps/opamps for either PA or HIFI all you do is change the grounding R from 22 to 47k or vice versa(plus a small amount of circuit adjustment.
aside: my late father also made his own electric guitars, and wound his own pickups(bless him), not to mention working for Goodmans in the early 70s, so I have a little handed down experience of my own lol PLUS being a qualified electrical Beng, and working in power generation and conversion, I understand all about inductive reactance.............stereo repair really isnt alot different, no offence, but I dont need partronising.
Besides, who said a pickup has a constant L? its operation isnt that of a simple coil inductor, particularly when you have a tone control with a C attached........
Last edited:
It's something which would interesting to investigate
Particularly as the pedal board is DESIGNED to feed in to the input of a guitar amp, which is high impedance.
As a simple 'guess' (based on previous weird happenings over many decades), I would suggest the input to the valve amp either isn't AC coupled, and it may be outputting a DC voltage upsetting the pedal board. Or, it IS AC coupled, and the pedal board requires a DC path to earth, to bias the output coupling capacitor.
I don't suppose you have the circuit diagrams for both units?. If you do I could examine them for you, and see if I can spot what might be the problem - but I would definitely suspect incorrect biasing of a coupling capacitor.
Line output from a PA is normally 600 ohm, although from an old valve amp it could easily be 10K-20K, valves don't do low impedances well (and of course were designed to feed to high impedance valve inputs).
Not outputs - these are always much lower, few hundred ohms to a few kilohms. IN/OUT are never impedance matched, it's a VERY bad idea.
Optimum meaning low output impedance feeding high input impedance, with a minimum ratio of at least 5 to 1, and more commonly at least 10 to 1.
Interconnects between different makes have never really been a problem (the low to high fixes the impedance question) - the only problem was signal levels, not impedances, there were no fixed signal levels.
I'm presuming you're familiar with DIN sockets?, they used a different impedance standard and voltage standard to phono (RCA in the USA) connections - yet were close enough to be interchangeable.
47K is an 'average' figure - it was just coincidence that the specific manual I had to hand used that value.
Just checked my PA mixer/amp, mono line inputs are 20K balanced, stereo ones are 100K unbalanced, preamp and line outputs are 150 ohms (balanced and unbalanced).
For reference, those two figures (20K and 100K) can be assumed to be 'identical', as they both more than swamp the output impedance they are designed to be fed from (150 ohms in that specific case).
Thats a bit more like it lol
back to civility
my amp has no decoupling cap at the input, but 10k in series, followed by 1M to ground, bypassed with 47pF. maybe this bypass C is the source of the problem....
As for the pedalboard, I actually believe its intended for Modern amps(relatively), also contains both digital and Valve technologies, offers a stereo output (L/mono and R & 'phones). I think really it is optimised for use DI'd into a mixer, so maybe thats the problem I have.
although, I also think every hifi amp kit/design Ive ever build has used a 22k ground R(only 2 amps though lol)
back to civility
my amp has no decoupling cap at the input, but 10k in series, followed by 1M to ground, bypassed with 47pF. maybe this bypass C is the source of the problem....
As for the pedalboard, I actually believe its intended for Modern amps(relatively), also contains both digital and Valve technologies, offers a stereo output (L/mono and R & 'phones). I think really it is optimised for use DI'd into a mixer, so maybe thats the problem I have.
although, I also think every hifi amp kit/design Ive ever build has used a 22k ground R(only 2 amps though lol)
Last edited:
No, it makes no difference - low in to high causes no issues.
If you have a meter?, check if there's a DC voltage on the output of the pedal board - because there's no blocking capacitor on the valve amp, so any DC from the pedal board will upset the bias on the first valve.
22K is fine, it's well within the normal range. I can't even remember what values I've used in the many amps I've built - basically it's pretty unimportant for line levels. If you're building a phono preamp though it's pretty important, and amps like the Leak I mentioned (and even more so the Quad amps) had configurable phono stages, including the input impedance.
I compare it to a Zobel because in the 'middle' of the tone control's range (actually, about 70K IIRC) it damps the resonance and there's a typical, if slightly droopy, -12dB rolloff. Above and below that the tone control merely raises or lowers a resonant peak, one based on the internal C of the pickup going towards '10' and another based on the tone C as you approach '0'. A good description is, "a tone pot acts as a pan control between two resonances, with a null in the middle".
But yeh, seeing it as a tank circuit is a big step away from the RC filter concept. (Not suggesting anyone here said that, it's just a very common misconception on guitar-based fora.)
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.