HR DeVille number two.

[Bluebark]

I tested the next project yesterday now that the first DeVille is fixed (probably sell). This one is a 410 and it seems to work but there is a crackle coming from the speakers. When I plug into the clean channel, all the pots seem to work as they should but volume is very weak. The sound coming out of the speakers sounds like it is going through a whoopi cushion and pushing the More Drive button produces a high pitched tone. I checked the bias before shutting it off and got a reading in the mid 80mV. I was able to get it down into the 60s and shut it down. Also, I noticed the fact that the filter caps had virtually no voltage stored when I went to drain the caps. Looking in the chassis I see no hot spots but noticed some kind of crystalized shrapnel around a resistor. Whatever it is, it’s caustic, looking at the diode below.




Looking into the high bias thing, I see two large carbon screen resistors. I think that’s what you call them. They are 47 Ω ±0.02% instead of 470 Ω 1W resistors listed in the schematic. So it looks like someone setup the box to run hot?? The work looks Pro so I'm doubting it was an ordering mistake. What kind of damage, if any, can this do to the electronics? I have read that running amps in the high 60s-70s will chew through tubes but not much on individual components. I did find a complete matched set of new Sovtek 5881/7025s from Groove Tube in a bag sitting on top of the reverb tank.



I believe my first move after inspecting the back of the MBs should be changing out all the electrolytic caps, replace the screen resistors and the weird resistor plus diode thing. Any thoughts before I start ordering parts?

As always, thank you for taking the time to read my post...Bob

[AlNewman]

80 mV compared to 60 mV at the bias doesn't mean anything without all the other voltages.
E-caps are a good idea, including the radial caps for bias and low voltage supply.

If the switching and reverb works on the clean channel as is, you likely don't have an issue with your low voltage supply yet.

Normally the 1st thing to do besides taking voltages is check all connections, clean pots and jacks, etc.  Beyond that, with a squealing noise when switching, my 1st thought would be a microphonic tube.  Beyond that there's ribbon cables, bad solder joints and traces, etc.

[tubeswell]

https://m.youtube.com/watch?v=3259fb4w37I

[stratomaster]

Plug into the Power Amp in of the amp and play through it.  If it sounds ok, then the noise preceeds the phase inverter.  If the noise continues I would begin to suspect the plate resistors in the phase inverter. For some reason these tend to die.

This is in addition to all the other known failure points like the low voltage supply and low quality axial caps.

[Bluebark]

Thanks Guys for the fast reply. I know these are like the amp from hell, watched all the vids on these things. But it's what I have to learn on. The first one had bad channel 2 and the switching didn't work. I swapped out all the electrolytic caps, upgraded some resistors and replace three bad ones. Applied a few mods that I found here and on the net and plugged it in using a bulb limiter like everyone said. Nothing smoked and the bulb acted the way it should. So since the caps are basically end of life I am going to replace them then re-flow the solder joints like I did on the first and go from there.

Al, I believe the 80mV at the bias test point is due to there being 47 Ω ±0.02% resistors to the power tube screens instead of 470 Ω 1W resistors listed in the schematic.

Stratomaster, I don't fully understand your fist suggestion, but I'll do my research on it for sure while waiting for the parts to come. I will also go over all the big resistors to make sure they are in spec.

Thanks everyone...I'll update after the cap refresh and inspection.

[AlNewman]

Al, I believe the 80mV at the bias test point is due to there being 47 Ω ±0.02% resistors to the power tube screens instead of 470 Ω 1W resistors listed in the schematic.

I don't think that's the issue.  There were amps built without screen resistors that work fine.  You can test the current draw of the screens by measuring both sides of the resistor, which at idle should be about 10% of total current draw per tube.  You can also test the plate current across the transformer and figure out the actual plate dissipation per tube.  It's more likely that either somebody increased the plate dissipation at the adjustment, or you're losing current through the bias capacitors, lowering the bias voltage.

It's also possible you have one tube drawing more current than the other, which can also be determined by measuring both tubes through the transformer and screen resistors.

As I mentioned earlier, the most accurate way to approach a problem like this, besides cleaning/tightening all pots, jacks, and connections, is to record voltages at all test points, and then you at least know if the tubes are supposed to be performing as designed.

Stratomaster had a good suggestion by plugging into the effects loop in to see if the volume and crackling problems were an issue with either the pre amp or power amp side.

[Bluebark]

Al, just made a spreadsheet for Delux/DeVille amp testing. I will throw some tubes in the amp and go over all the test point for voltages. I just figured it was moot since the caps do not seem to hold voltage.

[AlNewman]

The 2 balance resistors on the series reservoir caps act as bleeding resistors.  If you fire up with the LBL and it doesn't stay lit, it should be safe to remove the LBL to test and check voltages.  You should likely replace the E-caps either way, but you can at least maybe discover other parts you need before ordering.

[Bluebark]

Here are the results of my probing. Some of the readings are so far out of spec that I may go over them again using a different meter. Once again, I tested caps for voltage right after shutting off amp and nothing...0.0V. I think it might be wise at this point to just do the cap refresh and then take a second set of reading.

[stratomaster]

Here are the results of my probing. Some of the readings are so far out of spec that I may go over them again using a different meter. Once again, I tested caps for voltage right after shutting off amp and nothing...0.0V. I think it might be wise at this point to just do the cap refresh and then take a second set of reading.

Keep in mind many of these voltages are +/- 20%. Don't go chasing down non -problems.

I'm not following your statement regarding the caps. Are you under the impression that they should store charge after powering off, and them not doing so is indicative of a problem?  If so, that is woefully incorrect.  Many amps, this one included, will drain at power off for a few reasons.  The biggest one is the tubes remain warm and conducting for a short time after power is removed. This will drain down the stored charge.  The second reason is that the series caps in the reservoir have balancing resistors that are meant to equalize the voltage across each cap.  This will also gently discharge or 'bleed' the caps when power is removed.  Measuring low or no voltage on the HT line after powering down is normal behavior.

[Bluebark]

I just finished going over the amp with an auto ranging meter and did get some different reading and when I powered off and check the cap they retained some voltage.

Stratomaster, to clarify what I was seeing on the caps was absolutely No voltage retention what so ever...0.0V after shut down. That happened both times I had the amp on yesterday. Today, I did find voltage in the caps after shut down, mostly around 20V/50V. I have done nothing to the amp since yesterday accept let it idle out of standby for a couple hours while I went over the test points. Why did take 2 hours for me to record all the voltages you might ask? My OCD kicked in and every time I got a strange reading, I had to measure it over again. Sometimes up to three of four times...it sucks.

The parts should be in by the end of the week, so I am going to pull the boards and prep them for the cap refresh and reflow the solder joints and have a good look at that resistor that looks fried.

Thank you the follow up!

[AlNewman]

So, I guess by test points, I would mean DC voltage at all the pins of the tubes, the test points on the schematic are designated as signal points by ovals, and DC points as rectangles.

Nevertheless, there is some good information in your spreadsheet, as there's some weird stuff going on at PI.  Judging by your voltages, it looks like R57 is either open, or you have a bad ribbon cable or trace in that area.  TP 26 should be about double the measured DC voltage, and TP 24 should be around 215V DC.

[Bluebark]

Al, thanks for the info. I went over the CBs with an auto-ranging meter and have included the updated spreadsheet. I don't have a signal generator so all the "signal test point" voltages on the sheet are garbage. I pulled the boards last night and gave the board a good look'n. Someone has deffinetly in there mucking around...old flux all over the tube pins so they were reflowed at one time. The 470Ω screen resistors at R61 & R62 have been swapped out with non FP 47Ω carbon resistors.



Checked the 82KΩ 1/4W at R57, and it is indeed open but "looks" fine.



I also looked at the 100Ω resistors at R80 & R81 and they read around 51Ω. Looking at them you can see they are burning out, so will be swapping them out also.



I am going to start installing the new caps (F&T/Nichicon) and those bad resistors.

[SEL49]

I also looked at the 100Ω resistors at R80 & R81 and they read around 51Ω. Looking at them you can see they are burning out, so will be swapping them out also.

50Ω is exactly what they should read in circuit. You must unsolder one end of those resistors to get a true reading.

[stratomaster]

The 82k should be at least a half watt. I usually put a 1 watt in and do the other PI plate plate resistor (100k) as well. I think the voltage rating of the original resistor gets exceeded--as opposed to over dissipation, so if you've got the specs for your replacements it would be good to check the working voltage rating and compare to the circuit. The 1watts tend to carry a higher working voltage.

[Bluebark]

I also looked at the 100Ω resistors at R80 & R81 and they read around 51Ω. Looking at them you can see they are burning out, so will be swapping them out also.

50Ω is exactly what they should read in circuit. You must unsolder one end of those resistors to get a true reading.

I would like to discus this a little if you don't mind, because direct observation was telling me otherwise. When I read a resistor in situ with probes on either side of the resistor I get the correct reading per resistor specs...until now. lol When I measure point to point (Y) I get 50.1+- which is half of the 100Ω resistor. When I measure both resistors (R) I get 6.7Ω. Now my OCD kicks in. To my logic based mind, measuring between both resistors should give me a reading around 200Ω...clearly not what I am seeing. So to brake down down my logic, I always thought the meter reading was for whatever was between the two probes. Looking at the schematic, the only place to me where they connect is the light bulb filament. The board is out of the chassis so not connected to transformer or ground and not tubes.Having a hard time wrapping my mind around this.





I guess I am going to have to start a library...any recommendations on a good starter book? I ask you all because this place is full of well seasoned electronics guys. And I am sure you all have read books that were both garbage and little gems worth their weight in silver (Au is overrated). I am a tactile learner so reading, without doing is easily forgotten...and getting older is not improving things at all.

[AlNewman]

Try pulling the lightbulb and measure the filament. 

[Bluebark]

Measuring 3.9Ω. This is just my OCD peaking...I wouldn't give it much thought. But I do appreciate it though Al. Thank you.

Got a box from Doug yesterday, so most of my parts are in for the refresh. I'll install what I have and go from there.

Thanks again.

[passaloutre]

Remember you're also measuring the DC resistance of the filament winding on the PT (likely single-digit ohms) when you measure those resistors.

[Bluebark]

If I get the lingo correct...by PT, you are talking about Power Transformer? The board is out of the chassis, no connection to any transformer.

It's just my ocd kicking off...lol.

[Bluebark]

The amp is up and running. Sounds like crap! LOL Not like the first DeVille At All. But the first one I applied the Fromel mods to. The input jacks are jacked, so will be replacing them. Have not tested the fx loop yet. The pots are all scratchy, so I will need to do something about that. Channels and switches all seem to be functioning as expected.

Going to print out another spreadsheet and go over the voltages I see where the amp stands according to specs.