Comparing Electrolytic Capacitor Specs for Tube Radio Repair

Thread Starter

kmpres

Joined Jan 24, 2016
12
While replacing a 5.0uf electrolytic cap in a 1961 Japanese AM-FM-SW 7-tube hot-chassis radio I decided to measure and compare some of the specs of the four different 4.7uf caps I had in stock. The numbers varied more than I was expecting and choosing the best replacement became that much more difficult. Mind you, this was hardly an exhaustive test. I just wanted to see the relative merits of the four caps I had on hand so basically just measured the out-of-circuit capacitance and ESR values, as well as those of the original cap, at the five different frequencies offered by my DER EE DE-5000 LCR Meter. I know, the DE-5000 is not really a lab quality meter, but one does have to start somewhere.

All caps (except the original) are new and recently purchased. The original cap is a 5.0uf 50V electrolytic cap made by Japanese manufacturer National Radio and the four new ones are; a Topmay 4.7uf 50V cap; two MIEC 4.7uf caps (250V and 450V); and a Rubycon 4.7uf 450V cap. The Topmay was the cheapest and the MIECs the most expensive. Interestingly, however, the Topmay performed the best, especially at high frequencies, and the Rubycon the worst, despite the opposing opinions I sometimes see expressed by the electronics-cognoscenti in some of the online forums (see data in attached spreadsheet). The Rubycon was also the most unstable in that at 100 kHz its capacitance dropped and ESR rose noticeably during a single meter time-out cycle.

In the radio, this cap is placed in parallel with a 120pf cap between the cathode and plates of the 12AL5 FM Detector tube. I don't suppose the capacitance value is particularly critical here, I'm just curious as to the wide variance in readings picked up by my LCR meter and whether you think that any one of them would be better than the others in this location.

The radio was dead when I acquired it so I have no original symptoms to use for comparison.
So, based on the above data, which one of the five caps would you use in this radio?

Thanks.
 

Attachments

MrChips

Joined Oct 2, 2009
19,273
You are overthinking this problem.

How critical are the capacitor specs will depend on the specific circuit application. There is no generalized preference for all applications.

The tolerance on capacitance value can be as high as ±20% for electrolytic capacitors. It is not uncommon to find new electrolytic capacitors showing +20% on its nominal value.

You want the voltage rating to be 20-50% higher than the actual circuit voltage. For example, if the circuit is running at 30V, use a capacitor rated at 50V or 63V.

In your specific case, install the 4.7μF/50V capacitor. Measure the voltage across the capacitor. If the measured voltage is less than 40V then you're good to go.

Did replacing the capacitor get the radio working?
 

Thread Starter

kmpres

Joined Jan 24, 2016
12
It'll be a while before I get this radio working. Still have all the paper caps to replace, then figure out how to test and replace any bad tubes, then fix a shorted selenium rectifier, then figure out how to align a Japanese radio without instructions, all while relying on mail-order for all parts purchases. Radio work is challenging from an expatriate's point of view but also a lot of fun, so I'm not complaining.

I concur on the Topmay cap. I was just surprised that it tested so much better than the mainstay Rubycon and MIEC caps. Also, so far, all the electrolytic caps I have replaced in this radio were pre-measured to be 10% to 20% less than their rated value, but I replaced them with capacitance values slightly higher than the originals and voltages much higher so they should work fine. One triple-cap can (all three 70uf/150V) were 25% lower than their own rating and the one 5.0uf/50V cap was 90% higher, so there's no consistency. However, the new ones are smaller in size which makes soldering them in place a bit easier.

Thanks for the advice.
 

Thread Starter

kmpres

Joined Jan 24, 2016
12
Actually, I'm making better progress than I suggested above. An 8uf cap from Vishay/Sprague turned out to be a much better match than all the other electrolytic caps I tested. I was not aware of these inconsistencies before because I never had a decent enough LCR meter to measure them. Conversely, the Illinois Capacitor polypropylene caps I'm using to replace the paper caps have even better specs and are much more consistent. Pity they can't make them small like the electrolytics. The lesson here for me is to watch out for the wide variance in small electrolytics and you don't always get what you pay for.
 
Top