• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

never say never - 300B amps...

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
OK OK OK now I know I always said that indirectly heated tubes were the way to go

BUT...

I seem to have got me a couple of Shuguang 300B-98's and some 3.5K OPTS - oops! - I already have a range of HT transformers.

This amp is supposedly something simple, whilst I mull over the Gargantuan P-P amp that will eventually take over the world... Heck, I'm even considering cathode bias.

So my first question (& this is a SE amp). The common (Standard)bias point seems to be around -80V with 400v B+ (ok maybe not exactly there but in that ballpark)

WHY? What am I missing?

looking at this from the point of view of a total noob to DHT and certainly the 300B, whats 'wrong' with at most 300v on the plate at 90-100Ma, OK that's not max Pout but the tube is then operating in a linear area...

At the moment I'm looking at say 260 - 270v on the plate, Just -50v (fixed) bias, and 90-100ma But that seems really unconventional.

So what gives.

As always any guidance / pointers gratefully received - & I did search honest!

Andy
 
analog_sa said:
Just curious: how well does your output transformer like 100mA?

a fair point, I have a selection of 2 types at the moment, and I have little idea about either :xeye: still the favourites weigh in at >3 kg a piece. so I'm hopeful.

I understand a little of what you mean though, having just dumped KT88's at 100MA with some piddly little transformers in favour of EL34's because the reduction in standing current paid big dividends. However, in that case the tube still has plenty of beans. Biasing an 300b at -80v when my chart runs out of curves at -120 seems optimistic.

I'm on thin ice here and I know it, unknown OPT's and an unknown tube type. heh heh, well I guess this'll be what they call "a learnin' experience"

Andy
 
Andy,

A chart of 300B operating points gives: 350 V. on the plate, -76 V. on the grid, and Ib = 50 mA. into a 3.6 KOhm load as yielding 7.8 W. That should get you going.

You don't see many "fixed" bias 300B amps due to the possibility of runaway. There are a number of ways to prevent runaway. Combination bias is the method I favor. You'll have to bypass the smallish cathode resistor with your O/P trafos. Driving the 300B with a voltage follower allows for a small grid leak resistor and little chance for runaway. A 3rd technique is the use of a grid choke, instead of a grid leak resistor. Low DCR is combined with a high AC impedance.
 
andrew_whitham said:
whats 'wrong' with at most 300v on the plate at 90-100Ma, OK that's not max Pout but the tube is then operating in a linear area...

At the moment I'm looking at say 260 - 270v on the plate, Just -50v (fixed) bias, and 90-100ma But that seems really unconventional.

I have not tried the 300B, so I can't comment from experience, but I suspect that *some* users are trying to squeeze as much power out of them as possible. 8W might not seem like a lot, but it's more than the 3 to 4W you'll get from a 2A3 or the 1.5 to 2W that you might get from a 45. I think those are the popular contenders to the 300B. Some people love the 300B on its own merits, but some just want a DHT with more power....

Anyway, I think that running at a lower operating point offers the very real advantage that you could build a two stage amp. Avoiding a third stage is worthwhile IMHO. A 5687 or its cousins can drive a 300B grid with 50V peak directly from a CDP output.

-- Dave
 
Re: Re: never say never - 300B amps...

dsavitsk said:
My experience with this particular tube is that is can sound hard, but running the heaters low (mine are at about 4.3V) improves things considerably.

Now thats worth knowing. The power supply for the plates will be based on an LM317K (since I've seen it done and have a copy of the schematic) So adjustable is fine. Whilst we're at it what does reducing the heater power do for tube life? It must be good right? up to a point anyway.

When [and if] the thing works sucessfully then different tubes can be tried. I didnt want to shell out on the expensive ones only to find out that running closer to the 100Ma limit would effectively deal them hot electric death...

Andy
 
Re: Re: Re: never say never - 300B amps...

andrew_whitham said:
Whilst we're at it what does reducing the heater power do for tube life? It must be good right? up to a point anyway.

Mine have been fine for a few years. I guess too low could lead to stripping, but so far so good for me. There is an article somewhere where someone claims that lowering heater voltage on DHT's (but not IDHT's) lowers distortion. I have done no measurements myself, so I have no idea, but I do know that these particular tubes sounded better.

Oh, you also might want a higher current VReg instead of the LM317 -- the heaters draw 1.2A (1.22A or so for mine IIRC) which is pretty close to the 1.5A limit o nthe part.
 
I've made a couple of 300B amps, from 450V, 400V, 350V plate voltage and I finally settled on 350V. I didn't care for all the power I could get since my speakers can be driven to party level by even a type 10. To me the 350V operating point posted by Eli has the better SQ based on my equipment. Never used DC heaters and has always used AC on my DHT amps.

Having said that, I was able to convince my friend who's suffering from a type 26 hum issues and for the longest time does not want to use DC as it will "affect" the sound. So after much encouragement, he finally went DC with his 26 but never liked the sound. He tinkered around and said, if fed 80% of the filament voltage, he's getting "close to AC" performance. He's using current regulation (did I say that right?)
 
Eli Duttman said:
Andy,

A chart of 300B operating points gives: 350 V. on the plate, -76 V. on the grid, and Ib = 50 mA. into a 3.6 KOhm load as yielding 7.8 W. That should get you going.

Eli, yep I've seen that chart, but looking at the curves I was using it looked like one half of the wave form was going to spend its time in the scrunched up bits... basically I dont follow WHY thats the recommended setting. I wondered if there was a simple explanation.

Actually I wonder If I'm actually confused over the ratings, since my data sheet is a JJ one, or the STC one (all 3 pages) that you pointed out. Its clearly saying that 100Ma is max for quiescent.

Is there a decent datasheet anywhere I could be pointed at?

I clearly need to do more maths...

For the sake of discussion I've attached the curves I was using, from the electro harmonix datasheet, (I'm allowed to do this? apologies if not, but it WAS in the public domain.)

Andy
 

Attachments

  • 300b_with operating points.jpg
    300b_with operating points.jpg
    88.6 KB · Views: 1,125
Re: Re: Re: Re: never say never - 300B amps...

dsavitsk said:

Oh, you also might want a higher current VReg instead of the LM317 -- the heaters draw 1.2A (1.22A or so for mine IIRC) which is pretty close to the 1.5A limit on the part.

as it turns out I got distracted mid-post...

You have some suggestions? I just have a "couple" of these parts, I guess I could always add a pass transistor. (Actually I think I read somewhere that this also reduces noise)

Arnoldc,

To be honest I havent considered AC heaters (and OK, maybe I should)
Now, I know everyone has the most silent amp in the world but really? The danger of humm is what put me off this tube in the first place. (You have to consider that I'm the kind of muppet that DC heats indirectly heated tubes.) and I know that has NO effect I can hear...

So principally what are the differences? more distortion or what, is it quantifiable/qualitative?

Cheers.

Andy
 
Administrator
Joined 2004
Paid Member
I have used both ac and dc heating with 300B, and finally settled on constant current dc heating as the best compromise between sonics and output noise and ripple at the speaker. (My current Onken based speaker system is about 98dB efficient at 60Hz neglecting room effects) My current 300B SE amplifier has under 0.5mVrms broadband noise, and the ripple/hum component is completely buried in the noise floor.

I currently run a set of JJ 300B at 400V/70mA with fixed bias, over the last three years I have not had an output tube failure.

I had a lot of problems with hum with ac heated 300B in earlier efforts, despite taking a great deal of care to address lead dress, providing a filament balance pot in both cathode biased and fixed bias applications and was not able to get satisfactorily low hum levels with the speakers I was using at the time which were insanely efficient vintage JBL reflexes.

I expect someone using small full range drivers in reasonably sized horns or transmission lines may not have the low end extension to have a big issue with ac heating.
 
I would not recommend the LM317 as a plate supply regulator, there are better 3 pin regs out there for this duty. Its high frequency performance is not so great. Save the LM317 for purely DC tasks such as heaters etc.
 

Attachments

  • lm317 output impedance.jpg
    lm317 output impedance.jpg
    40.3 KB · Views: 649
From TECHNIQUES TO EXTEND THE SERVICE LIFE OF HIGH POWER VACUUM TUBES.

Minimize Filament Operating Voltage This procedure is used to set the filament operating voltage for the power tube to a value that will increase the service life without reducing the power output or the quality of the transmission in any way.


In all cases where filament voltage is mentioned, the parameter to be controlled is actually the filament temperature. The near impossibility of measuring filament temperature in an operating power tube makes the use of filament voltage as a surrogate for temperature a necessity. For the purists, filament power is a somewhat better surrogate than is voltage. However, filament power takes time to calculate. Still, it is important to keep the temperature of the filament in mind, particularly in the case of tubes such as the former Machlett tubes that operate with filaments 50 to 100 degrees Centigrade hotter than most power tubes.


The filament voltage minimization procedure begins by operating the tube at the recommended filament voltage for 200 hours. This preliminary conditioning gets the filament through the initial aging process. During this initial “burn in”, the tube parameters tend to change rapidly as the thoriated and carbonized tungsten filament wires become fully activated. The filament emission will often double during this period.


Next, the filament voltage is reduced in one or two percent decrements until there is a noticeable increase in distortion, a decrease in the grid, screen, or plate currents, or a decrease in power output. Distortion is usually the most sensitive indicator. Grid current is also a very sensitive indicator because the grid conduction angle is very narrow with most of the current drawn during the peak of the plate current. Once the first indication is encountered, the filament voltage is then increased back to the previous setting plus another percent to allow for tube aging. The filament temperature needs to stabilize after each change in voltage.


The above procedure must be performed for an AM transmitter with the modulation at the highest level used during program transmission. In transmitters where several power tubes are operated in parallel, the distortion contribution of a single tube will be diluted by the other tubes and this reading will be desensitized proportionately. If the power tubes have separate grid current metering, the grid current will likely be the most sensitive reading to monitor. Multiple tube class AB1 modulators will therefore have to be monitored with screen and plate currents or power output. Push-pull modulators may, in fact, experience a decrease in distortion as a result of the reduction of the filament voltages. This is probably due to the modulator tubes becoming more nearly alike (balanced dynamically) and canceling the even harmonics more effectively. Obviously, the tube with the highest peak emission will operate at a lower filament voltage than its mating tube, and the final result will be a better dynamic matching of the tubes. This is an additional bonus and will not cause any ill effects so long as the output power and currents have not been reduced. At the VOA Bethany (Ohio) Relay Station, a reduction in distortion of about five dB (from 2.5% to 1.4%) was experienced after reducing the filament voltages of 4CV100, 000C modulator tubes in the Collins 821A -1 transmitters.


A close watch should be kept on the tubes for the next few days to check for changes in the above operating parameters. The procedure should be repeated about once a month or if any adverse changes do occur.


A dramatic increase in the emissive life of the filaments can be realized if the tube has enough excess emission to permit a reduction of five percent or more in the filament voltage. For each three percent reduction in filament voltage, the emissive life of the filament will about double. A five percent filament voltage reduction will therefore more than triple the emissive life.
 
Tweeker said:
I would not recommend the LM317 as a plate supply regulator, there are better 3 pin regs out there for this duty. Its high frequency performance is not so great. Save the LM317 for purely DC tasks such as heaters etc.

Ok but what are the alternatives? I have 1.5 amps (conservatively) to shift so Its not really the best part. But most of the nice simple ones quit at around that amperage

I have a few L200's that are rated at 2A but noise is quoted at 115uV for 5v - is that good?

Output impedance drops with current and is quoted at about 0.1 ohms at 100KHz & 500mA, Surely thats OK?

any comments gratefully received., will try and post a schematic later - :hot: BYOB for the barbecue. ;)

Andy
 
andrew_whitham said:


Ok but what are the alternatives? I have 1.5 amps (conservatively) to shift so Its not really the best part. But most of the nice simple ones quit at around that amperage

...

Andy


How about the LM350 at 3A? Should be fine in current, don't know about noise though. I've used it on a DC heater supply with fine results.
 
Kevin, thanks for the pointer.

LT7805CT - bought! Importantly drop out is less than 1.5v meaning a rectified 6.3Vac winding has sufficient "headroom" for mains variation (also means that the 9v transformer I bought is now unnecessary :whazzat:)

Noise and output impedance are unquoted though....

Andy
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.