So dude brings a Sivertone 1474 to fix. It's all stock. It does kinda work except for 2 burnt resistors and the reverb. It has a piezo reverb that they don't make any more. I've been researching and researching, and everyone just says to match your old reverb with a new one. Since I can't, I decided to rewire the circuit for either a 6CG7 or 6N1P driving a 8FB2C1B since it had the highest impedance (1925Ohms in). The Ampeg Gemini 1 used a 6CG7 to drive a reverb tank, but they didn't say what the impedance was.
I don't want to add another tube, just remove the 2nd 12AX7 and put in a 6CG7 or 6N1P, both can get down to 6.5KOhms internal resistance, with the 6N1P giving me a lot more gain.
Will the MOD 8FB2C1B that I already have
mate up with a 6CG7?
Top picture is the Ampeg, bottom is the Silvertone.
I don't want to add another tube, just remove the 2nd 12AX7 and put in a 6CG7 or 6N1P, both can get down to 6.5KOhms internal resistance, with the 6N1P giving me a lot more gain.
Will the MOD 8FB2C1B that I already have
mate up with a 6CG7?Top picture is the Ampeg, bottom is the Silvertone.
SilverTone did the piezo because it saved a tube. Both Gain and Drive. I don't think you will get there, well, without a tube or equivalent amplifier.
A 6N23P seems to be a closer impedance match. With it I could get a gain of 31 for the gain stage and 27 for the drive stage. With the original 12AX7, Sears got a gain of 64 for the drive stage and 31 for the gain stage, so I do believe my signal is just going to be halved.
I don't know how much is lost in the reverb, and I don't know if there is a difference in loss between the transducer and piezo reverbs.
I don't know how much is lost in the reverb, and I don't know if there is a difference in loss between the transducer and piezo reverbs.
If you want a tube drive for a high-Z pan input check out the Ampeg circuits. Like the VT-40.
That piezo reverb is utter junk. Even when working right, it has all the tone of a screen door spring.
I have in the past successfully repaired those by harvesting a piezo disc from a cheap tweeter, and modified the little clamps to suit the disc. (One side versus two sides connections) But for all my efforts I just wind up with the same junky reverb.
That piezo reverb is utter junk. Even when working right, it has all the tone of a screen door spring.
I have in the past successfully repaired those by harvesting a piezo disc from a cheap tweeter, and modified the little clamps to suit the disc. (One side versus two sides connections) But for all my efforts I just wind up with the same junky reverb.
Well I got the resistors and caps replaced. I replaced the reverb driver and gain with a 6N23P. The 6N23P has a Ri of about 3K at 9.34ma. It all does work, except I'm getting a small bandwidth through the reverb. From 1khz to 2.3khz I'm getting about 4 times more signal out than at 600hz or 6000hz. In that band I'm getting more signal than through the non reverb channel. Why is that?
I got the reverb replaced with a MOD 8FB2C1B. Going by the Ampeg schematic I used a 0.47uf after the 6N23P driver, and am wondering if it's too small. I've been testing it to day and I found a 0.22uf cap on the input or output of the reverb does wonders, it increases 600hz while lowering 1Khz. But it also lowers 2.3Khz a lot. I've tried putting a pot in series with the 0.22uf cap and adjusting it, it helps, but very little. I've also tried a 6N1P, 6CG7, 6DT8, and 6BZ8, but it doesn't change the bandwidth.
I've tried 3 different circuits (Gemini II with a 6GC7, 6N1P, 6DT8, 6N23P, and 6BQ8, a pentode driver using a 6GH8A, from a online schematic, and just a cathode follower using a 6U8) all of them gave me the same variations. I even tried a Zune HD music player, and got the same results, a dip at 600 Hz and a peak at 1KHz. So I did a test and put in a series of audio test tones at a fixed input and recorded the output of the MOD reverb tank.
With 5 volts in:
60Hz - 50mv out
200Hz - 30mv out
600Hz - 12mv out
1KHz - 60mv out
1.5KHz - 30mv out
2KHz - 20mv out
5KHz - 5mv out
All into a 100Kohm resistive load. This is not what I was expecting from a good reverb tank.
With 5 volts in:
60Hz - 50mv out
200Hz - 30mv out
600Hz - 12mv out
1KHz - 60mv out
1.5KHz - 30mv out
2KHz - 20mv out
5KHz - 5mv out
All into a 100Kohm resistive load. This is not what I was expecting from a good reverb tank.
Reverb tanks are generally driven with a high impedance, semi constant current. Use a series resistor 10X the tank's nominal input impedance. That will slant the whole spectrum for better highs.
However the 600/1K dip/peak sure sounds like a poorly damped mechanical resonance. The "good" springs often have bits of foam, perhaps to damp such rings. I'd never thought the MOD series was supposed to be fine wine; more a lower price spread.
I did put in a 10K pot in series with the output a couple of weeks ago. Played around with it, but it didn't seem to do anything. Since I tested the reverb with the pot at 0ohms, I turned it up to 10K and retested and used a audio oscillator that put out 20 volts to compensate.
Series resistance 0Ohms 10K
50Hz 3mv 2mv
200Hz 15mv 2.5mv
600Hz 17mv 2mv
1KHz 100mv 17mv
2KHz 10mv 3mv
5KHz 6mv 5mv
I did verify that the oscillator was putting out 20 volts on each frequency.
Series resistance 0Ohms 10K
50Hz 3mv 2mv
200Hz 15mv 2.5mv
600Hz 17mv 2mv
1KHz 100mv 17mv
2KHz 10mv 3mv
5KHz 6mv 5mv
I did verify that the oscillator was putting out 20 volts on each frequency.
OK, so what happened was, I found a Accutronics datasheet that shows the 8 series and it's frequency response. As you can see it's all over the place, and indeed my peek at 1KHz fell flat at 980Hz, a good 10-15 db lower. It's not a concern with a guitar since it doesn't produce a single frequency in a sine wave. If I had a guitar, I would have been done a month ago.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.