Channel switching - Blues Deluxe

[MarkusBass]

Lowell,

I still suggest to measure the U3A input voltages. If the difference between these two voltages is positive, the first channel is switched on. If it is negative, the second channel is on. It looks like the difference is very close to 0, hence the problems. Can you measure the voltages with DMM?

Mark

[lowell]

Enzo and MB,
I am modding per RGs suggestion, but also wanted to be clear on those hot dropping resistors. I will measure those voltages on U3 once the mod is done.

thanks!

[Dave Curtis, dB AudioTech]

Check the solder joint on the indicator LED itself. I went through this nightmare with my HR Deluxe until I noticed that a physical jolt to the amp chassis would cause it to switch.

Ditto that. I've fixed a few deluxes/devilles with the same problem by resoldering the LED.

[lowell]

Hey guys... so I performed the voltage regulator mod and the amp is now stable as far a switching goes.... HOWEVER, there is now some weird noise happening. You know that interference from a cell phone that makes a speaker go crazy? THAT type of noise is what this sounds like. The noise comes and goes and is not in any rhythmic pattern... very chaotic. The noise level remains the same regardless of volumes being up or down, I believe it's in the power amp somewhere. I will open it back up and poke around but if anyone has an idea what this is let me know.

[lowell]

ok guys... this noise is driving me mad and I feel as though chasing a ghost. I do however know that the voltage regulator mod I performed has caused it. When the drive switch is engaged the noise stops. When clean channel is engaged the noise comes and goes. It is similar to that cell phone intermittence noise but not as loud. I measured some voltages and wonder what you all think might be the issue... maybe the issue is using 15v regulators w/ the diode in series w/ ground that is causing this problem but I'll let more experienced techs be the judge. Markus Bass - You said the voltage on clean channel at TP30 should be roughly 2.4v.... how did you get that? When I look at that I see the 28VAC, then the CR6 shorting the negative portion of the wave to ground leaving 14v right? Then R59, R60, and R61 creating a voltage divider leaving me w/ a .1x14v calculation equalling 1.4V at anode of CR7 and .8v at cathode of CR7. Tell me what I'm miscalculating here.

otherwise here are voltages I'm getting:

U3 (clean channel)
1 -14.55
2 3.41
3 .44
4 -15.86
5 0
6 11.95
7 14.34
8 15.67

U3 (dirty channel)
1 13.27
2 .06
3 .72
4 -15.86
5 0
6 10.77
7 14.56
8 15.67

[MarkusBass]

Markus Bass - You said the voltage on clean channel at TP30 should be roughly 2.4v.... how did you get that? When I look at that I see the 28VAC, then the CR6 shorting the negative portion of the wave to ground leaving 14v right? Then R59, R60, and R61 creating a voltage divider leaving me w/ a .1x14v calculation equalling 1.4V at anode of CR7 and .8v at cathode of CR7. Tell me what I'm miscalculating here.
Otherwise here are voltages I'm getting:
U3 (clean channel)
1 -14.55
2 3.41
3 .44
...
U3 (dirty channel)
1 13.27
2 .06
...

Lowell,

I just simulated the switch in SPICE program. I'm sorry but your first asumption that on cathode of CR6 diode there is 14V is already wrong. First of all there is R59 in front of it so the voltage is divided with some ratio. Then, the voltage is not a sinusoid signal any more and you cannot say that it is 28V/2. I attach a printout from SPICE. Looking from the top there is signal on cathode of CR6 and signal on cathode of CR7. The first signal has an average of 11V and RMS of 17.6V. The second one has average of 2.29V and RMS 2.32V. Please note that signal on cathode of CR7 is the same as signal on pin 2 of U2. Since this not a sinusoid signal you cannot simply measure it with a DMM.
Regarding the noise, I think that you get high frequency oscilations. Did you add capacitors as in 7815/7915 datasheets? I would fix the amp without those ICs, and only later add them to the amp.

Mark

[MarkusBass]

You should also remember that when converting VAC to DC the voltage does not stay the same. In case of full wave rectifier you have to multiply it by 1.41. Full wave and half wave rectifiers are explained quite well here: http://www.allaboutcircuits.com/vol_3/chpt_3/4.html

Mark

[Mars Amp Repair]

resolution to the HD Deluxe 16 volt regulators

Hey guys,
there is already another thread dealing with this issue. I have begun to replace those 470 resistors with 750ohm 7watt pc mount resisotors. I have about 5 of em out in the field now with no issues.
These zener supplies do not need to be nearly as 'stiff' as they've been designed. the 750ohm creates plenty of current to keep the zeners in hard forward bias and the tiny relays & couple of op-amps don't draw anywhere near enough current to cause the zener's to fall out of regulation.
find pix attached. It works great & resolved the issue of heat to the pcb & intermittents. I think I'll make a new thread for this...glen

[lowell]

Markus thank you for your expertise. I'll try my best to understand the spice simulation... and will read that recification page. fyi THIS noise did not start until I put the regulators in, previous to regulator mod the amp was suffering from intermittent channel switching, not noise.

[lowell]

Markus,
I've mulled the #s over in my head and to be honest I'm not grasping it. How do you get 11v average and 17.6vRMS from the first spice pic of 36v half rectified? I looked up that your 1.41 is the square root of 2. To get RMS voltage you divide peak voltage by that right?

thanks!

[MarkusBass]

Markus,
I've mulled the #s over in my head and to be honest I'm not grasping it. How do you get 11v average and 17.6vRMS from the first spice pic of 36v half rectified? I looked up that your 1.41 is the square root of 2. To get RMS voltage you divide peak voltage by that right?
thanks!

The values are just calculated by the SPICE program . I think that the average is calculated as arithmetic average of voltages and the RMS value is calculated as calculus of the signal. This is explained here:
http://en.wikipedia.org/wiki/Integral
and here:
http://en.wikipedia.org/wiki/Audio_power
When a signal is not a clean sinusoid usually there is not a simple way to calculate RMS value. In this case the signal is half-sinusoid (with values going down to -0.7V) . So in thes case your calculation would be a "rough" estimate of a correct value.
PS: the SPICE program is LTSpice from Linear: http://www.linear.com/designtools/software/ltspice.jsp . It's free so you can easily try it yourself. With the program you can easily verify problems like the one you have now. You can also simulate a failure of one of components and see whether the symptoms match the one that you encounter. In this way you can "virtually" repair amps and later verify whether your diagnosis is correct.

Mark

[lowell]

Markus,
On the datasheet for the 7915 regulator it has the two caps on input and output for the "standard app." Do they have this wrong or am I missing something. Shouldn't the output cap be reversed for the negative voltage?

http://www.onsemi.com/pub_link/Collateral/MC7900-D.PDF

[Enzo]

Although it usually is -, the curved line on a cap indicates outer foil, NOT polarity. The standard app drawing on the first page doesn;t specify polarity, but the caps should show + to ground as on the drawings further down th sheet.

[MarkusBass]

Markus,
On the datasheet for the 7915 regulator it has the two caps on input and output for the "standard app." Do they have this wrong or am I missing something. Shouldn't the output cap be reversed for the negative voltage?
http://www.onsemi.com/pub_link/Collateral/MC7900-D.PDF

Capacitors for 7815 and 7915 are shown (including polarity) on Picture 8 in this datasheet.

Mark

[lowell]

Markus,
thanks for the help with this and the SPICE info. I've played around w/ the zip file in the other thread that you uploaded.
The issue now is not the amp switching anymore. It's a noise... a very low volume noise. It is only there on the clean channel and is there regardless of volume/tone control settings. Is it possible that U3 is bad? How can I test that op-amp?