I was thinking about safety issues and given the limited BW of guitar amps I was surprised that I have not seen any consideration of transformer input. Has anyone tried/considered this? Other than cost is there any reason why it would be undesirable?
This would be the safest input method would it not?
mike
This would be the safest input method would it not?
mike
Hi Mike,
The problem is the high impedance of guitars, they like to see inputs with 100k ohms or so. And transformers work best when at least one side is low impedance, which would not be the case with a regular tube amp input. There are not many xformers around which qualify for this task, maybe the Jensen JT-112P-2HPC would work but it is a little bit too specialized for something else, and has quite a bit of capacitance, not good.
One can use an active buffer circuit before the xformer, then the primary works on a low impedance. I did this, with a low noise JFET-Follower (search for "Curl follower" for the topology), running off an extra isolated (floating) supply, and driving a Lundahl LL1531. There was no degradation in sound and only little additional noise.
The reason was not safety issues but galvanic isolation of the input (no earth GND connection), to avoid ground loop hum. It also helped against RF problems (no more radio station in the background). A problem is danger of ESD damage, which tends to kill semiconductors (or degrades them, each time a little more), while tubes don't bother at all. ESD protection is absolutely necessary.
- Klaus
The problem is the high impedance of guitars, they like to see inputs with 100k ohms or so. And transformers work best when at least one side is low impedance, which would not be the case with a regular tube amp input. There are not many xformers around which qualify for this task, maybe the Jensen JT-112P-2HPC would work but it is a little bit too specialized for something else, and has quite a bit of capacitance, not good.
One can use an active buffer circuit before the xformer, then the primary works on a low impedance. I did this, with a low noise JFET-Follower (search for "Curl follower" for the topology), running off an extra isolated (floating) supply, and driving a Lundahl LL1531. There was no degradation in sound and only little additional noise.
The reason was not safety issues but galvanic isolation of the input (no earth GND connection), to avoid ground loop hum. It also helped against RF problems (no more radio station in the background). A problem is danger of ESD damage, which tends to kill semiconductors (or degrades them, each time a little more), while tubes don't bother at all. ESD protection is absolutely necessary.
- Klaus
I see the problem. Would a cap (or cap with small resistance in series) from the input jack ground to the signal ground accomplish much the same thing safety wise (and take out the microphone bite threat) just as well?
mike
mike
There's no need for transformer coupling, the amplifier chassis, and thus the guitar lead, should be earthed anyway.
If you really are concerned?, use a radio pack on your guitar - no connection at all then! - but correct use and maintainence should keep a normal guitar amp perfectly safe.
As suggested, the required input impedance for a passive guitar is too high for a transformer, with the suggested 100K being really on the low side. Most guitar inputs are between 1M and 10M - but for an active guitar this isn't needed, as an active guitar has a much lower output impedance from the amplifier inside.
If you really are concerned?, use a radio pack on your guitar - no connection at all then! - but correct use and maintainence should keep a normal guitar amp perfectly safe.
As suggested, the required input impedance for a passive guitar is too high for a transformer, with the suggested 100K being really on the low side. Most guitar inputs are between 1M and 10M - but for an active guitar this isn't needed, as an active guitar has a much lower output impedance from the amplifier inside.
If you really want a transformer guitar input, Jensen makes the JT-DB-E direct input transformer. It is a 12:1 ratio step down that is made for guitar pickups. Guitar pickups run in the range of 5K impedance, which is not particularly high. I have measured 10V P-P from my Les Paul whilst doing Townshend style roundhouse power chords ;-)
The Direct Input transformers are meant to be used to input a guitar into a balanced microphone preamp or recording console input with a nominal 600-1600 ohm input impedance, hence the name "Direct Input" or D.I. but will work fine for a built in input circuit. You will need to make up for the voltage ratio of 12:1 (-21db) with more gain in the front stage which may lead to more noise. It might be the first guitar amp with a parallel 6922 input stage...
Michael
The Direct Input transformers are meant to be used to input a guitar into a balanced microphone preamp or recording console input with a nominal 600-1600 ohm input impedance, hence the name "Direct Input" or D.I. but will work fine for a built in input circuit. You will need to make up for the voltage ratio of 12:1 (-21db) with more gain in the front stage which may lead to more noise. It might be the first guitar amp with a parallel 6922 input stage...
Michael
Michael Koster said:
Check any guitar amp, you need MUCH more than 5K for a passive guitar - considering the pots are usually 250K the output impedance is obviously in the 100's of kilo-ohms.
Aim for 1M minimum, and preferably more - if you really want to be 'flash' (and you can buy expensive preamps that do this), then have a nice high impedance and place a pot (wired as a variable resistor) across the input, this will allow you to adjust the input impedance.
I'm missing something here; apart from galvanic isolation, what is the point of using a Tx to step down the level into a guitar amp? Doesn't make a lot of sense to me.Michael Koster said:
I think he meant that the source impedance from a passive guitar is around 5k.Nigel Goodwin said:
Nigel,
If I am understanding you correctly the ground of the input jack should go directly to the chassis safety ground not just to the signal ground. Is that correct? Sometimes signal and safety ground (earth) are separated by an RC string.
mike
If I am understanding you correctly the ground of the input jack should go directly to the chassis safety ground not just to the signal ground. Is that correct? Sometimes signal and safety ground (earth) are separated by an RC string.
mike
- Status
- Not open for further replies.