Filter Cap Question

[dpm309]

I am getting ready to re-cap an early '70's Fender Deluxe Reverb and was wondering if I can use 22uf/450V instead of the 16uf/500V caps?  I can get Nichicon 22ufs for a fraction of the cost of a Sprague Atom 16uf or 20uf.  I have been using the Nichicons in nearly all of my builds and repairs with no problems.  They are smaller and are great for putting them in small chassis.  Nichicon does not make a 16uf or 20uf/450V cap.  From what I read on Doug's parts page, it shouldn't make a difference but I just want to be sure.  What does everyone else use on these amps?

Thanks,

Dan

[Backwoods Joe]

Sure. Sounds like a good plan.

[tubeswell]

The difference in capacitance between an electrolytic cap that has '16uF'  printed on the side and an electrolytic cap that has '22uF'  printed on the side is much of a muchness. Chances are they could both actually be capable of 19uF.

The main thing you need to watch is the voltage rating of the caps versus the idle/standby voltage of the respective filter cap node, especially where you run the amp without tubes plugged in and there is no load on the B+. In this situation, all the supply nodes may sit at the same B+ voltage. If this unloaded B+ voltage is under 450V, you will be okay, if you only intend to run the amp on standby for relatively short periods of time in the overall scheme of things.

In all likelihood, if the factory is keen on product reliability, the voltage rating printed on the side of the caps is most probably a wee bit on the conservative side. But this doesn't mean that you should run the e-caps above the rated voltage. What's more, if you are running them at the rated voltage, its probably better to do so only for relatively short periods of time. In fact its probably optimal to run e-caps at around 15% less than their rated peak voltage. JM2CW

[sluckey]

wondering if I can use 22uf/450V instead of the 16uf/500V caps?

I'd say it depends on the B+ voltage applied to the caps. If the B+ voltage is less than 450, I'd say go for it, but if the B+ is greater than 450 I'd use the 500V caps. The difference in microfarads is not an issue. BTW, there are plenty of 22µF/500Vdc caps available that are much smaller than the Atoms.

[dpm309]

The original caps in the amp are 16uf/450V so I should be OK.  On another note, the owner wants me to look into a blackface conversion on this amp.  I know there has been a lot of discussion on this forum but I wonder if this conversion would really make that much of a difference in sound/tone.

Dan

[jim]

You could use a totem pole filter (like the Super Reverb) with two 220 mf 350V caps in series. making the voltage rating 700V before
the standby switch.  Leave it in standby till the cows come home....Jim

[ThermionicEngine]

The original caps in the amp are 16uf/450V so I should be OK.  On another note, the owner wants me to look into a blackface conversion on this amp.  I know there has been a lot of discussion on this forum but I wonder if this conversion would really make that much of a difference in sound/tone.

Dan

The original values are not necessarily adequate now.  I concur with Slucky, I would measure the actual B+ in your amp.  Your line voltage could be 5 or six volts above the line voltage of 117VAC,which could translate to an additional 20VDC.  Other factors could also create higher (Bias settings e.g.) B+ voltages. So an amp that was producing B+ 435 VDC could hit 455 VDC or more.  I look for high temp caps, and higher voltage ratings.  F&T (Germany) makes some very good caps IMHO.  Their 22 uF @500 VDC cap is much more compact than the Atoms, are half the price, and have an excellent track record for me. 

All my customers have been happy with the BF mod. (of course they asked for it)  I think there is something to it.  If you re-cap, it will probably perform much better.  The conversion mod will be the icing on the cake.
All the best

[stratele52]

Why do you want to use cheap caps in a high quality amp ? Use Sprague or F & T caps. Don,t do this mistake , you'll hear de difference.

[FYL]

Why do you want to use cheap caps in a high quality amp ? Use Sprague or F & T caps. Don,t do this mistake , you'll hear de difference.

Electrons don't know if the components are cheap or overpriced.

Nichicon VX axials are much better than Sprague Atoms and nearly on par with F&Ts A series.

[ThermionicEngine]

I have found that discussing the merits of various cap brands is like discussing politics or religion.  A lot of techs have very strong opinions on this topic and no one is about to change their point of view.  There was a big thread about this topic about 6 months ago.  The use of higher voltage ratings generally does not evoke much argument.  There is a point of diminishing returns- I would not use a 25 uF @ 150VDC as a 12AX7 cathode bypass cap. I do use the 25 uF @ 50VDC, though. A lot of amps out there with IC's, Nichicon and Xicon.  Then there is Sozo... Lots of strong opinions there.  I use F&T's and Atoms most of the time.   I have never had an issue with either of them, that I can recall over the last 17 years.  But that's just my experience and I respect the views of others.  Much debate on this topic fo' sho'...

[FYL]

I have found that discussing the merits of various cap brands is like discussing politics or religion.

Caps, resistors et al. are passive components. They can be fully characterized objectively as well as subjectively.

Yes, there can be some differences between them - some ceramics are vastly non linear. Yes, some may impart a sonic signature of their own - ie carbon comps are noisier than metal films. But no, there no magick caps, resistors or wires, as demonstrated a zillion times by proper double blind tests.

And yes, most self-proclaimed tone experts or golden ears are unable to find any sonic difference when doing a proper ABX test. A Sozo will sound just like a Mallory. But in a sighted test, they'll wax lyrically about the component of the month and it's tone-enhancing properties.

The bottom line is: if you feel better using fancy components, use them.

[Willabe]

Nichicon VX axials are much better than Sprague Atoms and nearly on par with F&Ts A series.

I bought a bunch of Nichicon VZ(M) radials, 105 degrees.C. FYL have you tried these? Any thoughts on them?

Nice $, nice size, can't hurt to have the extra temp. rating. After sluckey posted pics of a couple of e-caps he had cut the can open on to show the real size of whats inside, that was enough for me to try them.


               Brad      

[dpm309]

The B+ voltage is 438V so I am going to go with the Nichicons as I am told by another amp tech that these do have a safety margin.  As I mentioned above, I have had no issues or returns using these.  On the blackface conversion, I compared the 2 schematics and will remove the 1200pf caps on the power tubes, replace the 2 330K resistors on the PI with 1M, change one of the cathode caps from 5/25 to 25/25 on V5, and change the cap in the bias circuit from 50/70 to 25/50.  Also, I will change the 2 resistor in the power supply with 10K, 2W resistors.  I don't know if I am going to change the rectifier tube from a 5U4 to a GZ34.  Do these mods look reasonable?

Thanks,

Dan

[FYL]

I bought a bunch of Nichicon VZ(M) radials, 105 degrees.C. FYL have you tried these? Any thoughts on them?

No direct experience with the 450V version, but the 63/100V perform well, are affordable and very reliable.

[Willabe]

Good enough, thanks FYL.


               Brad     

[ThermionicEngine]

The B+ voltage is 438V so I am going to go with the Nichicons as I am told by another amp tech that these do have a safety margin.  As I mentioned above, I have had no issues or returns using these.  On the blackface conversion, I compared the 2 schematics and will remove the 1200pf caps on the power tubes, replace the 2 330K resistors on the PI with 1M, change one of the cathode caps from 5/25 to 25/25 on V5, and change the cap in the bias circuit from 50/70 to 25/50.  Also, I will change the 2 resistor in the power supply with 10K, 2W resistors.  I don't know if I am going to change the rectifier tube from a 5U4 to a GZ34.  Do these mods look reasonable?

Thanks,

Dan

Sounds okay to me.  Rectifier tubes both work. I would keep the higher value cap on the bias supply. You might have to address layout when you cut the caps across pins 5 and 8 on the power tubes but more often than not, you won't.  They put them in to stop high frequency oscillation, which many folks feel came from poor layout or cheeper components.  good luck!

[dpm309]

Thanks for all of the input.  I am now headed in the right direction.  I tested the tubes and found that I have 4 bad/marginal preamp tubes that should be replaced.  For some reason, all 6 of the preamp tubes are 12AX7s.  I will install the correct 12AT7s in V3 and V6.  The amp does not sound bad with this preamp tube configuration but I think I should install the correct preamp tubes.  Can anyone think of a reason that 12AX7s were substituted for the 12AT7s other than that was all they could get at that time.

Dan

[HotBluePlates]

That or the owner/retuber thought "12AX7 = More Gain!" Even though that wouldn't be true for the reverb driver, and only marginally true for the phase inverter.

... an early '70's Fender Deluxe Reverb ...

... On another note, the owner wants me to look into a blackface conversion on this amp.  I know there has been a lot of discussion on this forum but I wonder if this conversion would really make that much of a difference in sound/tone.

I don't know how the amp in question compares to the pull-boost Deluxe Reverb schematic posted in the schematic library, but...

My opinion is the greatest single benefit might be to correct some generally-lousy silverface wiring. I dunno if it's the case in your amp (some models had almost zero change during the silverface era) but sometimes the layout is changed somewhat, and wire runs become very much longer. You then have a greater risk of oscillation, which then necessitated the caps from output tube grid to ground, yada, yada...

The owner may like the idea of a bias adjust instead of a bias balance, if your amp is wired as a balance control. I personally feel the issue of 330k grid reference resistors in the phase inverter is a non-issue, because the apparent input resistance is very much higher than 1M due to bootstrapping. However, the switch downward to 330k is also why the input cap to the phase inverter was raised to 0.01uF, to maintain the same bass response.

So, your call on whether it's a good idea. It is more bench time for you to charge the owner... Not usually a good reason to mod an amp though.

[dpm309]

Well, I changed the filter caps, 4 bad preamp tubes, and did the blackface conversion and I am getting decent sound out of the amp.  Checked the bias (I installed 1 ohm, 3 watt resistors from pin 8 to ground on the power tubes) and I can't get more than 2.1 mA.  When I set my multi-meter for DC volts, I get about 22mV.  My plate voltages are 431 VDC and according to Weber's bias calculator, I should bias it to ~22mA.  Could there be problem with my multi-meter?  I used it recently on the Twin Reverb conversion and I got good readings on both mA and mV settings.  I am confident that I have it biased properly since if I remember correctly, 1mA equals roughly 1mV.  As I said before, the amp sounds good and is performing correctly.  I checked it with another multi-meter and got 22mV and 5.2mA.  What could cause this?

Thanks,

Dan

[sluckey]

(I installed 1 ohm, 3 watt resistors from pin 8 to ground on the power tubes) and I can't get more than 2.1 mA.  When I set my multi-meter for DC volts, I get about 22mV.

So, you're getting 22mV across a 1 ohm resistor.

Using Ohm's Law, I = E/R = 22mV/1 = 22mA. Sounds just right to me.

When using 1 ohm resistors on the cathode you always set your meter to measure voltage across the resistor. Then you use Ohm's Law to calculate the current through the resistor. Since you will be dividing the voltage by 1 you could say the measured mV is equal to the calculated mA.

To measure actual current you would have to break the cathode ground and put your meter (switched to measure current) in series between the cathode and ground. Then you would measure the actual current which should also be 22mA.

[dpm309]

Thanks Sluckey, I should have know that.  Must have been having a senior moment.  Anyway, I feel better about shipping this amp out now.  By the way, the blackface mods I made really brought this amp alive but a lot of that could have been due to new filter caps.

Dan