Hi every one. I need some help with a scott 299c. I have the amp working with a variac but I can't apply full line voltage without blowing the fuse on the amp. I have replaced the power supply caps the rectifier for the bias and changed resistor R214 to bring voltage close that on schematic. Voltages at the plates of the output tubes is higher than expected. I know line voltages are higher than when the amp was made so how do I compensate for this. Doesn't this change how the output tubes are biased. Please note I have limited tube experience.
Harry
Harry
I'm probably not going to be much help, but i did wake up one morning with acrid smoke thru-out the house from a 299 power trafo that had shorted....
dave
dave
Harry, it's time for some step by step tests to determine where the problem is. I assume you've done the basics like testing all the tubes and are using the correct fuse of 2½ amp slo-blo. (even 3A is OK) And also made sure the new filter capacitors aren't getting warm. Here's where we start.
First, remove the 5AR4 rectifier tube and see if you can go to full line power. If not pull all the tubes and disconnect the bias rectifier AC leads and try again. If the fuse still blows, it's the power transformer. Also check the small .01 cap across the primary.
If the fuse stays with no 5AR4, the PT is OK. Something is drawing excess current and you must narrow it down. The output tubes may be pulling excessive current because of improper (low) negative bias voltage. The coupling capacitors feeding the output tubes (7591s) may be leaking positive DC to the grids and upsetting the bias voltage. It's always a good idea to replace these, especially if they're original. You'll need a voltmeter to check this. Excessivly higher screen voltage will also make the tubes run harder. Plate voltage not as much. The amplifier should tolerate today's line voltage.
Last edited:
Thanks for the reply. I will check the .01 cap at primary and all of the coupling caps have been replaced. If the output tubes are not installed I can go to full power and can trace a signal through all the stages. This makes me think the output tube bias may be my problem. I do get operation of the amp with the output tubes installed and it sounds fine but I can't go to full line voltage without the fuse blowing. This makes me think the output tubes may be fine. How do I adjust the bias and what voltage should I have there?
According to the schematic (http://hhscott.com/pdf/299C.pdf) Scott uses the negative bias supply to power four of the small signal tube filaments. Apparently they are run a little under rated voltage as well. (the -40V point) You said you changed the value of R214. I will assume you increased it to provide the same -40 volts. This may be part of the problem since the plate and screen voltage is now higher, the negative bias voltage needs to be increased as well. I would first try putting the original value of R214 back and see how it runs. The tube filaments will probably run closer to rated voltage, but that won't hurt them.
If the amplifier now works ok, make note of the negative voltage present on the 7591s control grid (pin 5). The schematic originally called for -18.5V. With an increased B+ level, the negative bias voltage will need to be a little higher to keep quiescent current through the output tubes to around 60mA for class AB1 operation. (The RCA tube manual calls for -21V w/450 plate & 400 screen @ 66mA)
One handy modification I would make is to install 10 ohm resistors in the 7591A's cathode to ground. This will provide an easy way to measure tube current and 10 ohms makes the math simple.
Should you wish to keep the original "starved" filament condition, you will need to increase the bias voltage to the grids by other means. Since the circuit voltage is essentially fixed, increasing the value of R210 (?) 1.5K ohms should raise it up. Some experimenting will be necessary. Or you could keep the 1.5K and tap voltage off further back. Say from C208 or C209 if needed.
Dumb things I have done, installed electrolytics with the arrow as +; rebuilt a power supply and tested the voltage in it with no load on it. Assumed I could simply not install the two 12AX7s for the phono section in a Scott 299. Start measuring AC going into a power supply and then wonder what I could have done to get such a strange low reading out of the rectifier--I've still got my meter set to measure AC!
Best from Tucson
Bob
Best from Tucson
Bob
OK I have returned to the original resistor for R214 this has helped 21 volts. it was suggested I change it due to the fact that I changed the bridge rectifier and there is less loses with the new one. You also suggested a cathode resistor to check quiescent current do I need to install one for each tube.
Harry
Harry
It's also entirely possible that once full cathode temperature is reached in the output tubes, one of the tubes is running away due to gas. The close grid cathode spacing of the 7591 would help that along a bit too.
Yes, one for each tube is desireable. ½ or 1 watt carbon comps. Measuring the voltage across the resistor will give you current. (I=E÷R) Normally the advice to change the resistor you did is valid. However, you have extenuating circumstances. Putting it back was a test to see if the amp would take full line voltage. Perhaps a though check of the resistor values in the bias circuit is in order. And if you are using todays "modern" tubes (chicom garbage) performance and specifications can be way off base. Like mt66 says, one (or more) of the 7591As may be acting up when hot. Try looking closely at them in a darkened room for signs of redness on the plates or a glowing screen grid inside the plate structure.
Ok looked at tubes in the dark two of them had a blue arura seen around outer of the plates. one I could see blue inside the plate by the grids and one nothing at all. the tube that was clear the current was 48 mil amps the 3 others were @ 100mil amps. Sounds like I have some bad tubes. Would you agree? If so any suggestions about replacements? JJ's Harmonix
A blue color outside of the plate is pretty normal for a power output tube. Blue inside the plate near the grid is less healthy but not necessarily a death sentence depending on how severe it is. But 100 mA is too much current. With 400 volts on the plate, 100 mills represents 40 watts of plate dissapation. The tube is rated for 19 watts. (per the RCA manual)
Since you cannot adjust each tube's current individually, you will need to get pairs that work with the provided bias voltage. You want them to run around 60mA each. Unfortunately I have no real experience with todays imported tubes. I don't like them and never use them. (I have a lifetime supply [thousands] of NOS tubes for my personal use.)
I don't know him but I would suggest contacting Jim McShane for tubes. I have only heard/read good things about him and truely believe he is a knowledgeable and trustworthy seller. Explain your situation to him and he'll surely be able to help you with the best tubes. Here's the link.
Home Page
Victor
BTW, I just looked at Jim's site and found he has Electro-Harmonix 7591A in stock. He's also aware of some problems with Scott and Fisher amps that use these. Copied from his site:
Important note! There have been some cases where the new production 7591A tubes (EH and JJ) have had trouble
in circuits using a very high value grid resistor - as vintage Scotts, Fishers, and others do. The best solution is to change
all the grid resistors to 180K-200K (from the old value over 300K - sometimes they even exceed the old RCA tube manual
recommendation!),
Important note! There have been some cases where the new production 7591A tubes (EH and JJ) have had trouble
in circuits using a very high value grid resistor - as vintage Scotts, Fishers, and others do. The best solution is to change
all the grid resistors to 180K-200K (from the old value over 300K - sometimes they even exceed the old RCA tube manual
recommendation!),
Ok after Hollwstate said the dissipation was 40watts and the tube was rated at 19 watts I checked my math with a calculator this time not my head. my father like to say "no sense being Supt unless you can prove it" My voltages at the 10 ohm resister were from .115volts to .167volts. .167/10=.0167 amps or 16milliamps. It is clear my bias voltage is to high now and I should be able ramp up my line voltage. I am going to construct a bias circuit to make adjustment easy. I was thinking of copying the bias circuit for the 299d what do you think?
Harry
Harry
I think adding two seperate bias level adjustments (one for each channel) would be a good idea. It's something I would do myself using quality pots of course . And if you will be running today's tubes, consider Jim's suggestion of lowering the grid return resistors as well. (the ones that go to ground) I don't know why Scott elimimated these adjustments from the C version. Perhaps to make it simpler for the user by having fewer adjustments, or maybe the bean counters just got in the way. Back then tubes were built to a more ridged standard. Today it's make them close and "good luck Charlie".
- Status
- Not open for further replies.