|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
64.203.19.193
In Reply to: A NEW Dynaco ST-70 posted by Bob Latino on April 5, 2007 at 14:34:59:
Hi Bob,Checked out your link; those are very good prices with all the work and parts you throw in. It's very rare for me to see a stereo 70 with such nice chassis - on the island here everything rusts in a few years....
Indeed, the st70 was designed to be a super-tight budget amp, and has tons of faults, which need to be corrected. However, as it is, it has a very unique sound that many like. With mods that correct these faults, we get better sound, but with different character. However, considering the excellent opts, and a couple of the parts that can be reused, one can rebuild it with sane amount of money to an amp that you'd like.
I think there was never an amp that was so popular for tweaking, and rightly so. I have found that one of its (many) weakest links is the bias supply. The bias voltage of the EL34s is super critical, yet in the st70 it is very crudely filtered. Using a high voltage schottky there, with beefed up caps (CRC filter with 2000uF - R- 100 uF instead of the original 50 uFs), it will sound way better. I have done a lot of PSUD simultions for the bias supply, and the first cap value determines the ripple rejection by far. There are rapid gains up to 2000 uF or so, above that you have to go way higher (10.0000uF) to have noticeable improvement, but that size cap would not fit under the chassis. Increasing the second cap size above 100 uF hardly lowers the ripple at all. I try to keep the cap values as low as possible, as doubling the cap size slows the cap down to half speed: it will take twice as long to recharge after a charge drain. With 100 uF second cap, the sound is very fast, and powerful. With 2000uF second cap the slound is bloated and sluggish.
I was trying to change the filter to RCRC, but any kind of resistance before the first cap slowed the sound of the amp tremendously.
Cheap tweak, done with a few $, and makes a huge difference. :))Keep on the good work!
Follow Ups:
Janos,Watch the size of those caps, especially the second one. The bias supply might take longer to come up to full voltage than the warmup time of the output tubes, especially if you use solid state diodes on the HV supply.
Hello Kyle,Thank you for the warning. It takes 1sec to reach 75% of the bias voltage with the 100uF as the second cap. (2sec - 90%)
I'm using tube rectification, so it's not a problem, but 2 sec should be safe with ss HV. I don't know how much the bias can limit the sudden hv rush on the cold cathode... but I'd guess it would be protetive.
Janos,You are right that the original bias supply is not a tight circuit. With the VTA driver board that I use on the new amps that I build the bias supply is ON THE BOARD and not centered around the 7 lug terminal strip. The original bias supply and those 10K potentiometers didn't really give enough range and as the tubes wore they didn't have enough adjustability. On the old amp with the old bias system the trick was to RAISE the value of R1 and LOWER the value of R2. Joe has some bias expanding instructions on his web site but what I have found is that if I parallel a 27K ohm resistor with the 10K on R1 you get about 7300 ohms and if you parallel a 7.5 K resistor with the original 10K on R2 you get about 4285 ohms. With an "effective" 7300 ohms for R1 and 4285 ohms for R2 you get a wider bias range and can more easily bias tubes that have some time on them. This is a little more range than the instructions on Joe's web site but works well for me on a customer's amp especially if the selenium rectifier has been replaced with a diode.
This post is made possible by the generous support of people like you and our sponsors: