DOA Dan Echo

[Cruton]

Hello everyone,

I'm working on reviving a Dan Echo (schematic here: https://www.dropbox.com/s/mnp42vknn9wb61a/Dan%20Echo%20Schematic.pdf?dl=0). The pedal doesn't produce any output, bypassed or effected.

Working backward from the output, I've confirmed that the signal is present at IC4B pin 6 (the inverting input of the JRC4558 being used to sum the output. The switching seems to work fine, as the signal at pin 6 is either dry or effected depending on the footswitch. I can manipulate the tone, repeats, hi/lo, and speed and hear it at pin 6. However, there is no output at pin 7 of the opamp.

Voltages on IC4 (JRC4558):

1: 2.53V                  8: 5.05V
2: 2.53V                  7: 1.94V
3: 2.50V                  6: 1.94V
4: 0.02V                  5: 2.52V

Pins 6 and 7 seem a little low. I've confirmed that there's continuity from pin 7 to the C201 output cap, and from the cap to the output jack.

My first thought is to replace the JRC4558, but I'm not sure whether I'm missing something else that could explain the audio loss and wonky voltages at pins 6/7. I'd rather not deal with wrestling that chip out of there if it's not the problem.

[willienillie]

My first thought is to replace the JRC4558

Mine too.

wonky voltages at pins 6/7

Probably meter loading.

[anotherjim]

C3 could be bad. Do pins 6,7 volts wobble when you twiddle the mix control?

Ah, "does yo volts wobble when you twiddle" should be a song lyric.

[Cruton]

I just checked, and I get 1.94-1.96 on pins 6/7 regardless of the position of the mix control. So these volts don't appear to wobble when I twiddle (which is indeed a great song lyric).

I always forget about meter loading, that's gotten me in trouble before once with a cheap meter... anyway, would it be possible for the IC to be bad if all the voltages are correct (within meter-loading range, at least)? The only other time I've run into a bad IC, the voltages were waaaayyyyy out of whack.

[willienillie]

If you have sound going in, but none coming out, and the voltage conditions are all correct, seems like it almost has to be a bad IC.  If it's a through-hole 4558, cheap and easy to try.

[Cruton]

I have some 4558's around, so I'll try to put a new one in it tonight and see if that solves it.

[ElectricDruid]

If the 4558 is wired up as a standard inverting mixer, you shouldn't be able to hear anything at pin 6. That inverting input is a virtual ground mixing node, and for that reason there's nothing to hear there - it's a ground, virtual or not.

So that smells wrong, right from the off.

Plus 5V isn't a valid supply voltage for the 4558. The datasheet I found suggested +/-4V as the minimum.

https://datasheetspdf.com/pdf-file/489253/NewJapanRadio/JRC4558/1

Maybe someone's made a 5V variant though, I dunno.

HTH,
Tom

[Cruton]

I may have misspoken earlier. From what I can see on the schematic, pin 6 (the (-) input) takes both the wet and dry signals, and pin 5 (the (+) input) is tied to Vref through a 10k resistor (R19). There's no audio at pin 5, but there is at pin 6.

[Cruton]

And I agree, the 5V supply voltage is weird. It appears to be the case for these pedals though: http://mirosol.kapsi.fi/2015/03/danelectro-de-1-dan-echo/

[PRR]

> 5V isn't a valid supply voltage for the 4558.

It isn't sure to "meet specs". Gain may be down, distortion high, etc.

It may "work", for someone's idea of "working". Especially in pedals where "specs" are unimportant.

[Cruton]

Wise words!

I replaced the 4558 and it fired right up. It seems like 5v "works" in this application. The article I linked earlier said that the 5v makes it easy to start clipping the signal, which made sense to me. After playing it for a bit, I feel like I could hear that. Sounds good enough for a cheaper pedal though, I'm enjoying it.

Thanks for the sanity check everybody, and the fun lesson about "specs".

[PRR]

5V isn't a valid supply voltage for the 4558. The datasheet I found suggested +/-4V as the minimum.

http://www.ti.com/lit/ds/symlink/rc4558.pdf   page 9

The lone diode is really a "Zener", really a B-E junction of around 7V breakdown. This against the resistor to the left of it sets all the current sources.

So +/-4V (8V) will turn-on the 7V Zener.

And 5V will put about 4V where they planned to have 7V, so all the currents will be "half".

Gain, GBW, and slew will be about half of nominal spec. But 4558 is really fast enough for clean 100mV signals, not fast enough for 10V cymbals (over-zzzz-ed TASCAM gear), but not-wrong for even 1V of dirted guitar.

[Cruton]

Given the application that the 4558 is used in here, would there be any benefit from the usual op amp swaps (TL072, NE5532)? I socketed it so it's easy enough to try, but I was curious if knowing it's running off of 5V would itself negate the value of such a swap.

I guess first step would be to check specs to make sure they would "work" under the 5V like the 4558 does.

[Mark Hammer]

The Dan-Echo uses a PT2395 delay chip, which relies on an external DRAM chip, rather than incorporating RAM inside the chip, like the PT2399 does.  Do we know the external DRAM is OK?  If it was compromised in some way, there would be no delay signal.

[Cruton]

Thanks for pointing out what the symptoms of a compromised DRAM would be, that's good to know. The DRAM did catch my attention in the schematic. I've never dealt with a pedal where there's RAM to troubleshoot, so seeing it made me a little antsier going into the repair.

But luckily for me the delay section seems just fine, it was just the output mixer that had kicked the bucket.

[Govmnt_Lacky]

More than likely, the use of +5V as the Vcc is due to the use of the PT2395 chip.