Is there a way to test the quad cap in the Mk III?
I have searched but found no discussion
I have searched but found no discussion
Testing the Quad Cap ?
Greg_M- Posts: 40
Join date: 2010-09-18
Location: Poulsbo, WA
anbitet66- Posts: 89
Join date: 2009-12-23
Age: 47
Location: Valley Stream, NY
Check out this site:
http://www.capacitorlab.com/anatek-blue-esr-meter/index.htm
I built this tester about 18 months ago, and have used it many times from computer circuit boards, to my Pioneer SX-34b, to an old AC unit in my parent's home when the compressor quit (it wasn't the S/R cap). It tests capacitors in circuit, but you don't get a "capacitance reading". Instead it reads the capacitors impedance in circuit, usually ignoring resistors and other components. It cost me $79 and shipping then and is well worth the cost in saving time and effort. Use it with a regular capacitor tester and you'll see for yourself that sometimes a cap tester can lie about the condition of a capacitor.
http://www.capacitorlab.com/anatek-blue-esr-meter/index.htm
I built this tester about 18 months ago, and have used it many times from computer circuit boards, to my Pioneer SX-34b, to an old AC unit in my parent's home when the compressor quit (it wasn't the S/R cap). It tests capacitors in circuit, but you don't get a "capacitance reading". Instead it reads the capacitors impedance in circuit, usually ignoring resistors and other components. It cost me $79 and shipping then and is well worth the cost in saving time and effort. Use it with a regular capacitor tester and you'll see for yourself that sometimes a cap tester can lie about the condition of a capacitor.
View the profile
Sal- Posts: 189
Join date: 2009-02-05
Location: Central New Jersey Dynaco-ST70.com
Greg,
The best way to check the quad cap is to connect a high voltage power supply set at or slightly below the rated voltage to the capacitor (disconnected from the circuit in the amp) and with a milliamp meter in series to measure the current at the capacitor rated voltage. If the capacitor is good it should pull a few milliamps maximum per section and not get warm. Once that is done, you must, I repeat must discharge the capacitor and measure the capacitance of each section.
Actually, you can measure the capacitance of each section before you put B+ to it to measure current.
And lastly, if you have a ESR meter, you can check that also.
Hope this helps.
Sal
PS: If you don't have a high voltage power supply you can use the power supply in your amp. Be very careful as there are lethal voltages under the chassis.
The best way to check the quad cap is to connect a high voltage power supply set at or slightly below the rated voltage to the capacitor (disconnected from the circuit in the amp) and with a milliamp meter in series to measure the current at the capacitor rated voltage. If the capacitor is good it should pull a few milliamps maximum per section and not get warm. Once that is done, you must, I repeat must discharge the capacitor and measure the capacitance of each section.
Actually, you can measure the capacitance of each section before you put B+ to it to measure current.
And lastly, if you have a ESR meter, you can check that also.
Hope this helps.
Sal
PS: If you don't have a high voltage power supply you can use the power supply in your amp. Be very careful as there are lethal voltages under the chassis.
j beede- Posts: 209
Join date: 2011-02-07
Location: California
I am not sure that it makes sense to test the quad cap in a Mark III... they are always bad 
Both Fry's and Harbor Freight sell a DMM for $40-45 that includes (among other things) a capacitance function that I have used to measure 9pF up to 99µF. I used it on a "good" Mark III quad cap that had passed the analog meter "visual" test. Using my DMM It measured 0/62/32/29µF versus the expected 30/20/20/20µF. I bypassed the 0µF and 62µF sections to see if that would get the amp running on the bench again. It did. I haven't investigated the tester's operation, but I assume the test is done at low voltage and low frequency.
...j
Both Fry's and Harbor Freight sell a DMM for $40-45 that includes (among other things) a capacitance function that I have used to measure 9pF up to 99µF. I used it on a "good" Mark III quad cap that had passed the analog meter "visual" test. Using my DMM It measured 0/62/32/29µF versus the expected 30/20/20/20µF. I bypassed the 0µF and 62µF sections to see if that would get the amp running on the bench again. It did. I haven't investigated the tester's operation, but I assume the test is done at low voltage and low frequency....j
Bob Latino- Admin
- Posts: 1525
Join date: 2008-11-26
Location: Massachusetts
J,
That 62 section on your cap is probably GOOD. You can't measure the capacitance of a quad cap in circuit and get an accurate result. Here's why. On a stock Mark III or ST-70 you have a 30, 20, 20, 20 cap. But those cap sections are connected together by the choke and resistors. The choke and resistors allow bleed over capacitance from one section to the other. This will give you the impression that the cap's rating is higher than it should be. The only true way to get an accurate reading of one section of a quad cap is to remove all the wiring and resistors between each section and then measure each section by itself to chassis ground.
Bob
That 62 section on your cap is probably GOOD. You can't measure the capacitance of a quad cap in circuit and get an accurate result. Here's why. On a stock Mark III or ST-70 you have a 30, 20, 20, 20 cap. But those cap sections are connected together by the choke and resistors. The choke and resistors allow bleed over capacitance from one section to the other. This will give you the impression that the cap's rating is higher than it should be. The only true way to get an accurate reading of one section of a quad cap is to remove all the wiring and resistors between each section and then measure each section by itself to chassis ground.
Bob
j beede- Posts: 209
Join date: 2011-02-07
Location: California
Thanks for the remonder Bob. I would only measure a cap out of the circuit. I suspect that the higher than expected reading could be due to an internal short between segments. As I noted the cap function in my meter runs at low frequency and low voltage. Things might be quite different at high voltage!
...j
...j
Bob Latino- Admin
- Posts: 1525
Join date: 2008-11-26
Location: Massachusetts
J,
A quick test of a quad cap (but not the best test) is to do a resistance test on the 4 sections. Set your meter at its highest resistance setting. Next place the BLACK probe on the chassis and the RED probe on one of the sections of the quad cap. If the cap section is good it should show a rising resistance as the cap charges from the battery in your meter. Eventually the meter should rise to over 1,000,000 ohms (1 meg ohm). When you get to the next cap section it may show a resistance over 1 meg already because it may have also slightly charged through either the choke or one of the resistors on the quad cap. If the resistance stops rising at say 5K or 10K, the cap is probably leaking and should be replaced.
Now here is the catch ... A resistance test is not the best test because that cap is only being hit with 9 volts from your meter and may test fine only looking at the 9 volts. In circuit, when hit by 440 VDC, it may crap out completely. Solution - next time you buy a multitester spend a little extra and get one that also has a capacitance scale.
While we are on the subject of meters sometimes it pays to have two meters - one to back up the other. Why? Sometimes you get a reading and you look and say > "This can't be right?" OR the reading is way off what it should be AND you believe it. "hey - there is a problem with my amp". If you suspect something is wrong you pull out the OTHER meter and make the same measurement. If you get the same reading on the second meter there is a problem. I had someone once send me an amp that "the bias readings on all four tubes are WAY off". I checked it out on BOTH my meters and the bias could easily be set to the proper setting. It turns out that his "auto-ranging" meter would not auto-range any more. His meter was defective.
I use a B&K meter and back that up with a Fluke meter if necessary. The B&K is off just a little ... A 1% metal film resitor will read about 1/2% to 1% high on the B&K. The Fluke is dead accurate. I use the B&K daily and back it up with the Fluke. (It's a Fluke model 115 > cost about $130). For just measuring bias on your ST-70 most meters are fairly accurate but if you want a dead accurate meter spend the extra money and get a Fluke. The Fluke 115 will also measure capacitance.
Bob
A quick test of a quad cap (but not the best test) is to do a resistance test on the 4 sections. Set your meter at its highest resistance setting. Next place the BLACK probe on the chassis and the RED probe on one of the sections of the quad cap. If the cap section is good it should show a rising resistance as the cap charges from the battery in your meter. Eventually the meter should rise to over 1,000,000 ohms (1 meg ohm). When you get to the next cap section it may show a resistance over 1 meg already because it may have also slightly charged through either the choke or one of the resistors on the quad cap. If the resistance stops rising at say 5K or 10K, the cap is probably leaking and should be replaced.
Now here is the catch ... A resistance test is not the best test because that cap is only being hit with 9 volts from your meter and may test fine only looking at the 9 volts. In circuit, when hit by 440 VDC, it may crap out completely. Solution - next time you buy a multitester spend a little extra and get one that also has a capacitance scale.
While we are on the subject of meters sometimes it pays to have two meters - one to back up the other. Why? Sometimes you get a reading and you look and say > "This can't be right?" OR the reading is way off what it should be AND you believe it. "hey - there is a problem with my amp". If you suspect something is wrong you pull out the OTHER meter and make the same measurement. If you get the same reading on the second meter there is a problem. I had someone once send me an amp that "the bias readings on all four tubes are WAY off". I checked it out on BOTH my meters and the bias could easily be set to the proper setting. It turns out that his "auto-ranging" meter would not auto-range any more. His meter was defective.
I use a B&K meter and back that up with a Fluke meter if necessary. The B&K is off just a little ... A 1% metal film resitor will read about 1/2% to 1% high on the B&K. The Fluke is dead accurate. I use the B&K daily and back it up with the Fluke. (It's a Fluke model 115 > cost about $130). For just measuring bias on your ST-70 most meters are fairly accurate but if you want a dead accurate meter spend the extra money and get a Fluke. The Fluke 115 will also measure capacitance.
Bob
Similar topics
