Reverb doesn't work

[DeepEnd]

Some of you may recall that I built a volume attenuator box for my Fender FM65R that goes in the effects loop between the "Pre Out" and Power In" jacks. It works fine but here's the problem: When the attenuator box is in the circuit, the reverb, a great old-style spring unit, doesn't work. No reverb at any setting. I'm baffled but here's a link to the schematic if anyone is so inclined: http://support.fender.com/schematics/guitar_amplifiers/FM_65R_schematic.pdf.

[onelife]

looking at the diagram, i don't see any reason the loop would cut off the reverb under normal circumstances so it appears something is amiss with your amp - some fender amps are notorious for flakey solder connections

 

what happens if you plug the preamp out directly into the poweramp in?

 

maybe there is some kind of printed circuit board issue that shows up when you use the jack(s) for the loop - try moving the plugs a bit to see if the reverb comes back on

[WRGKMC]

The send Taps the signal through a resistor and cap just before the op amp that drives the reverb send and returns. It also taps the reverb pot through a resistor and cap.

 

The return is switched and ground to the input jacks yet the jack sleeve have their own ground.

 

The diagram is too small to read the ground lug numbering. It may be the ground potentials to the reverb circuit are removed or changed when the return jack is used. This would disable the reverb pot from working properly to mix with the send signal.

 

Have you tried a straight cable without an attenuator between the send and return to see if you have reverb?

 

If you don't the ground config is designed to disable the reverb circuit.

If you do the attenuator may be causing a ground bias issue with the reverb tank circuit.

 

This is an oddball setup with the grounds. The send is buffered with an op amp so using the attenuator shouldn't hurt anything but It may be set up this way to prevent reverb tank oscillation and why its got a tap before and after the reverb tank top the send.

 

Some amps have the reverb tank after the preamp send, some before.

 

It does appear that if the send is used to send a signal to the PA the reverb would be present. With both the send and return used, either the impedance needs to match or the reverb is disabled.

 

Try jumping the input and output with a straight cable. If the reverb disappears, the amp was engineered to do that.

If it does work, then you may want to use an L pad to attenuate your speaker instead or using line level attenuation.

 

 

[gardo]

 

Have you tried a straight cable without an attenuator between the send and return to see if you have reverb?

Try jumping the input and output with a straight cable. If the reverb disappears, the amp was engineered to do that.

If it does work, then you may want to use an L pad to attenuate your speaker instead or using line level attenuation.

 

 

Good starting point

[WRGKMC]

I was able to look at the schematic in more detail at work. Nothing strange about the grounds I see.

 

The dry feed to the Send uses a C32 = .33us cap and R 21 = 22K resistor. The signal from these will be stronger then the wet reverb feed from feed from C38 = .22uf cap and R58 = 330K resistor (plus the 850K pot). These two sets of caps and resistors set the maximum reverb level. The pot adjusts the level from a shorted reverb output to the maximum fed through C38 & R58 and the pot.

 

There should be reverb coming from the Send. It will be low in volume but it should be there.

 

You're using a line level attenuator and cranking the preamp up for drive tones but attenuating the feed to the power amp.

The two caps, C32 and C38 high pass the wet and dry signals. The wet signal has more lows removed by the .22 cap then the dry signal does with the .33 cap. You want less lows on the wet reverb signal.

 

The attenuator pot is likely throwing the two ratios dry vs wet off. The power amp is wanting to see the full line level signal from both legs. The Ratios coming from the wet feed have no gain compensation when the attenuator is being used.

 

My guess is the reverb is there, but since the preamp is cranked up its dominating the feed to the power amp. This is an analog transistor amplification curve issue that would need to be analyzed on an oscilloscope or have the op amp removed and placed on a load line analyzer.

 

The output at different voltages vs amplification curves aren't linear. The voltage may increase in even steps but the amplification rises like a curved hill. At the low voltage end, there may be nothing till a certain voltage is reached then the amplification rises steeply and after that it levels off and changes very little with voltage changes.

 

By using an attenuator and cranking the preamp, the dry signal is passing on the high end of the amplification curve and the Wet reverb signal may be on the low end of the gain curve, so the ratio between Wet vs Dry is no longer mixed together in the right amounts to be heard.

 

Fender chose the balance of wet vs dry feeding the power amp without any additional attenuation.

 

 

Believe it of not I have this same thing happen with my Little 15W Marshall practice amp. With preamps set low and the power amp cranked I get a normal reverb with a clean amp tone. Its not an overly strong reverb but comes out about 50% wet vs dry with those settings. When I crank the preamp up for drive and turn the power amp down, The reverb practically disappears in comparison to the driven signal.

 

Reverb send and receive elements in the tank can only be driven a certain amount. You cant saturate the driver or increase the receivers sensitivity without modifying the circuit. You can gain the dry signal up but the reverb send/receive levels remain constant, especially the receiver which returns the wet signal to the mix.

 

This is normal and expected in a simple reverb circuit because there is no change in ratio between the wet and dry feeds. That compensation just isn't there on these inexpensive designs

 

Your reverb is probably there but the dry signal is gained up high so the reverb level is being masked by the dry signal. Checking it with a straight cord should reveal if there isn't something else going on.

 

You can modify the reverb strength to run in attenuated mode if you want. It may be a but too much reverb when you use the amp normally. Adding a switch and adding a parallel resistor to R58 should be all you need where you can switch between attenuated or normal modes. Another 330K resistor connected via a switch should double the reverb strength. Maybe something higher for less and something lower for more wet reverb. I'd say a resistor between 150~500K would do the trick.

 

Or you could just buy a reverb pedal and run it in the effects loop before the attenuator if you want to continue using the attenuator and don't want to modify the amp.

[DeepEnd]

Thanks. I'll try a straight cable, no attenuation, first chance I get and report back.

[DeepEnd]

. . . The attenuator pot is likely throwing the two ratios dry vs wet off. The power amp is wanting to see the full line level signal from both legs. The Ratios coming from the wet feed have no gain compensation when the attenuator is being used.

 

My guess is the reverb is there, but since the preamp is cranked up its dominating the feed to the power amp. This is an analog transistor amplification curve issue that would need to be analyzed on an oscilloscope or have the op amp removed and placed on a load line analyzer. . . .

Well, I finally had a chance to check it out and it looks like you got it. I had the knob on the attenuator at 12 o'clock and there was no reverb. I tuned it up 45 degrees to about one thirty and the reverb was back. A straight cable between the "Pre Out" and "Power In" produced normal behavior. If this graph is accurate, I've gone from about 10dB attenuation to about 7dB (http://e2e.ti.com/blogs_/archives/b/...potentiometers), which means I probably could have simply plugged my guitar into Input 2, which is for guitars with active pickups and has less gain. I swear, I'm starting to hate this amp.

[WRGKMC]

Well at least you figured it out. Apparently attenuating too much shuts the wet feed from the reverb down completely.

There may be a simple way to mitigate the problem. If you to add a cap with a value between .1uf ~1uf inside the attenuator box between the leg on the pot that's grounded, and ground (put it in series with ground). What this will do is work as a high pass filter.

 

Your signal which is AC will go to ground like it does now but the circuitry in the amp that supplies DC bias to the reverb circuit wont be loaded down and remain normal. I'm not 100% sure it will work but Its a common sense item to try which in theory it should work.

 

In any case its an inexpensive and simple fix and in if it removes the loading effects of the attenuator box on the amp its should restore the reverbs normal operation no matter where the knob is set.

 

If the problem is a ratio issue between dry and wet reverb feeds with the attenuator added, then you may want to think about using an Lpad attenuator on the speaker instead of attenuating the line level between the pre and power amp. Its more expensive but you can save allot if you build it yourself.

[DeepEnd]

I could try an L-Pad but I don't really want to sink any more money into this amp. I have a stray .047 uf capacitor lying around from when I replaced the pots in my bass and I might try that.

[WRGKMC]

045 might not be high enough, it will wind up acting like a tone control instead of an attenuator. You need at least .1uf ~1uf

 

I added this to your diagram.

 

 

 

If it wasn't for the need of the bypass out remaining at line level, you could swap the two red wires on the pot around and attenuate the input instead of the output to the power amp. This would bring the wet and dry reverb levels down evenly.

 

 

Another possibility might be to use a higher value pot of 50~100K which would have less load on the reverb circuit. The attenuation would be less linear when turning the knob but 100K is a common master volume pot in amp circuits. It may go from full off to full on within 1/2 the knobs turn instead of a full turn but you'd have less load across the reverb circuit. I'd try the cap first though. It only costs a buck.

 

We live and learn on these things. I'm sure you'll look at having those options available when you're thinking about buying your next amp. Having a normal line out, effects sends and master volume lets you crank the preamp the way you like, and the master volume has no effect on the line send. You just set the master volume for what you need and the line send does what its supposed to do.

 

We don't always know what we need when we purchase amps until we run into situations like yours. Others can learn of their importance and the limitations of work arounds and maybe decide to spend a few extra bucks when deciding to buy one amp over another. Many buy on impulse and neglect to even think about the options. Many of these options used to only came with premium gear. Luckily even inexpensive amps are beginning to include them now.

 

I bought allot of vintage gear throughout the years and many of the amps had no line outs, master volumes or effects loops. My last New purchase was the Marshall Valvestate which had all of them plus the line out was speaker emulated which is even better. A normal line out can be harsh and spikey because the signal is missing the coloration of the power amp and speakers. With speaker emulation it nails the miced cab coloration and dynamics and works fantastic for recording direct or live PA feeds.

[DeepEnd]

Okay, check me on this: http://www.instructables.com/id/Pass...h-Pass-Filter/. If I'm interpreting this formula correctly, R would be 10,000 for the value of the pot and (2πRC)/1 with a 1uF (0.000001F) capacitor would give a cutoff frequency of about 15.9Hz while the 0.047uF capacitor would have had a cutoff frequency of about 339 Hz. I can see why you said it would be a tone control! OTOH, 15.9Hz is inaudible for most people and well below the lowest note on a guitar in Standard tuning (about 82.4Hz) anyway. I'll go shopping for a capacitor the first chance I get. It shouldn't be hard to add it to the circuit.

[DeepEnd]

Update: I stopped by the nearest Radio Shack and bought a 1uF metalized film capacitor. $1.99+tax. I soldered it in and lo! It worked. I turned the knob all the way down to 9 o'clock, which is a fair amount of attenuation, easily 15dB, and there was still sweet, springy reverb. Thanks again.

[WRGKMC]

Update: I stopped by the nearest Radio Shack and bought a 1uF metalized film capacitor. $1.99+tax. I soldered it in and lo! It worked. I turned the knob all the way down to 9 o'clock' date=' which is a fair amount of attenuation, easily 15dB, and there was still sweet, springy reverb. Thanks again.[/quote']

 

Hey well how bout that. You see your symptom was similar to many things I've done in the past. The hard ground on that pot was probably draining the DC bias on that reverb chip and shutting it down. The cap cant pass DC so just as long as its high enough to pass all the frequencies and act like a straight wire for AC only I figured it should work.

 

As a note, I've worked with simple filters allot in guitars and had the values pretty much memorized. I Have a chart someplace I did a long time ago that I calculated all the RLC combinations for passive filters. It was a grid where I could just select the frequency I wanted to either pass or cut and it listed the components I'd need to do that. The only problem with it is you'd run into some oddball component values and would have to connect say caps of different values in series/parallel to get those oddball values because manufacturers only make specific standard values. You then have to mix and match the standard values to get the actual values you want.

 

Later when I started working in the field, you'd learn some easy tricks. One is to have some decade boxes for Caps coils and resistors. They come in real handy doing repairs because you cant always make out the part numbers on components especially if they burnt up. You'd simply connect a decade box and give it your best guess based on the circuit, and ramp the voltage up slowly with a variac.

 

Given the fact many components have a 10~20% tolerance, ball park guesses are often good enough and so long as the repair passes a good stress test it will often work fine.

 

I used to use these decade boxes with guitars using different coils and caps to use in various tone circuits. I can pretty much use my ears and detect what frequencies I want to pass of cut, then its just easier for me to go grab a couple of parts out of my parts bins and just do it vs trying to remember how to calculate it out down to the nearest value.

 

I'd do that stuff when I was young and wanted to fix complex issues where values were strict. You'd run into that allot of times balancing high quality stereo channels. Even with the correct values calculated, it rarely worked however. All the components in a long amplifier chain all contribute to the final output so as a tech you're always tweaking with things to get the best results tweaking things to make them work, much like in your case.

 

Anyway I'm glad it worked. Now you have a useful tool and it didn't require buying a whole new amp, just a $2 cap.

[DeepEnd]

They had 2.2uF capacitors on clearance for 75¢, which would have probably worked and given me a cutoff of about 7.23Hz, but I'd already decided to go with 1uF based on our discussion. Anyway $1.25 isn't that much. Thanks again. I've also updated my thread in DIY.

 

That said, I'd still dearly love to try out a Vox VT40+, 60 Watts SS 1X10. I'm not crazy about the modelling features but it's another lightweight reasonably powered amp that won't have the issues my Fender does.

[WRGKMC]

I'm not sure what you're using for a formula but there should be no high frequency cutoff. High pas caps only have a low frequency cutoff, a low pass cap has a high frequency cutoff. A cap conducts better the higher the frequency goes up because its the change in polarity that allows it to emulate conductance.

 

Your .047 cap would have conducted all mids and highs, but there wouldn't have been enough capacitance for the lower frequencies to excite the cap and pass them. (the reverse when used in a tone circuits, the cap rolls off all the highs and leaves the bass.)

 

A 1uf or higher should pass all your frequencies down to your lower bass notes and everything above it, with no high end roll off. (In a revere configuration as a low pass tone cap it will roll off nearly all frequencies)

 

 

You can look up coupling caps in circuits for more on this. Gain stages in amp designs are connected with higher value caps which pass the entire audio signal but they separate the two stages DC bias so the bias of one stage doesn't affect another. The coupling caps are the signal feed that pass 20~20Khz audio signal just like your High pass is taking all frequencies to ground.

 

As you can see in this diagram, C1 separates the guitar pickup from the transistor. This allows only AC to feed the transistor. The other resistors around the transistor, set up the working DC bias voltages. C4 is another Coupling cap which extracts the signal from the transistor circuit and prevents the output from loading the transistor down by robbing its DC voltage. Both are 1uf caps which pass nearly all the frequency spectrum. (at least all that's worth hearing. there may be some sub lows that are important to a bass guitar so you're likely to see higher value coupling caps in a bass circuit to get better lows)

 

As far as Vox goes, don't underestimate their amp modeling. They do have some well designed circuits and produce some fantastic tones. They aren't overly complex to set up either. I think that have allot of other manufacturers beat in that field but you know how competitive it is. Like all manufacturers today, they often have presets that showcase the possibilities. You have to know where to take those presets and mod them for even better tones and set up your own presets to get the best results. A beginner may not have developed an ear for targeting the results he wants and relies strictly on the presets and therefore never explores the full range of the gear.

 

Vox has some wide ranges in their gear if you get past the typical ear candy and get into setting them up for specific results. I was quite impresses with both their stomp boxes and their higher end SS guitar amps. My recording buddy has one of those big AD100VTH heads and super Beatle cabs that really does a fantastic job.

[DeepEnd]

I'm not sure what you're using for a formula but there should be no high frequency cutoff. High pas caps only have a low frequency cutoff' date=' a low pass cap has a high frequency cutoff. . . .[/quote']

I didn't say anything about a high frequency cutoff. I was referring to a low frequency cutoff. I already posted it three days ago and asked if I was on the right track but here's the formula for a simple high pass filter again: (2πRC)/1. You can find it here, just like I did: http://www.instructables.com/id/Passive-Filter-Circuits/step1/The-High-Pass-Filter/. If the resistance is 10,000 Ohms from the pot and the capacitance is 0.000001 Farads from the capacitor, that gives (2π X 10000 X 0.000001 = 0.06283)/1 = 15.9154943 Hz. Lower frequencies are progressively more attenuated. Substitute 0.047uF in the equation and you get (2π X 10000 X 0.000000047= 0.002953)/1 = 338.6275 Hz, which definitely wouldn't work well because it covers all the fundamentals of a guitar (82.406889-329.6275569 Hz) and a bit more so I would've ended up with a weak signal dominated by highs. Thanks for talking me out of it.

 

Thanks too for the comments about Vox amps. A local mom and pop has a used AD30VT for $129 but I want something a bit louder than 30 Watts SS. Otherwise I'd check it out.

[WRGKMC]

Cool. Maybe I misunderstood. You posted something there about a 6 or 7K cutoff that had me scratching my head. Maybe you were referring to how a cap worked as a low pass or something.

 

Anyway, it works now and you're good to go. I'm not sure how well the small Voxes work. some of the ones with small speakers are probably fine for practice. I's probably want something with at least a 12" for doing anything live. The amp emulations and effects in the vox gear is pretty darn good. I'd just want to be sure I could match a drummer playing live. A single 10 at any volume just isn't quite enough for me size wise. A pair will work, but I use mostly big stuff, 2 and 4X12's, 4X10's etc. I like a wall in back of me for plush sounding clean chords that match the kick and bass. Leads can came from a smaller speaker but I love that clean chord big cab resonance even at low volumes. Guess it dated back to my acoustic days when you can feel the acoustic resonate against your chest. I like that same feel coming from a guitar cab.

[DeepEnd]

Sorry. I wrote 7.23Hz not KHz. Apparently you misread it; I can see where that would have been confusing. I'm an acoustic guy at heart so I like to feel the notes too but given that a 65 Watt 1X12 is too loud if anything, I think a 1X10 with a similar amount of Wattage (the Vox VT40+ I mentioned is 60 Watts) would probably be fine. Anyway, my Fender is doing the trick for now and I'm keeping my eyes open but I haven't found anything yet.