Beefing up filtering for Eico HF-22

smokestack

New Member
I know it's been covered in this forum before, but I'm not finding exactly what I'm looking for...

I've got an Eico HF-22 that I am in the process of replacing electrolytic caps in and for which I also purchased a 1.5h 200mA 400v choke to beef things up a little. This is as per the AK thread a while back wherein one of the members lovingly refurbished a pair of these amps and said he was going to add chokes. My question is this: do I keep the first two filter sections (20uf/20uf stock, I've replaced with 32uf/32uf) in parallel and then wire in the choke before the next filter stage, or do I put the choke between the two 32uf/32uf sections. (the first arrangement gives me 64uf, if I don't separate the multicap with the choke, right? Do I want to do that, or do I want to have the choke between the two sections of the multicap?)
 
The 5U4GB spec sheet suggest that 40 uF is the largest cap size that should be used in a "capacitor input" power supply filter without additional plate resistance added.

So if 32 uF is what you've got, I would use that as the single cap right after the rectifier. The reason for the stated max is to keep the surge current low during turn on so the rectifier does not arc. If it arcs once, it will likely arc again which basically renders your rectifier tube unusable.

Typically then in this type of power supply, the choke comes in between the first and second filter capacitors. Because there is some DC resistance in the choke itself, the cap that comes after the choke can be significantly larger than the first cap. The combined ripple rejection of the C-L-C filter section will still be significantly better than the original circuit of just two 20 uF paralleled caps.

At the same time however, the DC resistance in the choke will drop some B+ voltage, maybe 5 to 8 volts depending on circuit conditions. Still, a small price to pay for the way better ripple rejection the choke provides.
 
I know it's been covered in this forum before, but I'm not finding exactly what I'm looking for...

I've got an Eico HF-22 that I am in the process of replacing electrolytic caps in and for which I also purchased a 1.5h 200mA 400v choke to beef things up a little. This is as per the AK thread a while back wherein one of the members lovingly refurbished a pair of these amps and said he was going to add chokes. My question is this: do I keep the first two filter sections (20uf/20uf stock, I've replaced with 32uf/32uf) in parallel and then wire in the choke before the next filter stage, or do I put the choke between the two 32uf/32uf sections. (the first arrangement gives me 64uf, if I don't separate the multicap with the choke, right? Do I want to do that, or do I want to have the choke between the two sections of the multicap?)

You sir, should download PSUD, from Duncanamps. Do a google search for it. It's an awesome tube power supply simulation software, and you can use it to simulate every conceivable arrangement of power supply which comes into your head, and view the ripple voltage. It will even warn you when you do something which can damage your rectifier tube. For me, it has entirely replaced my "old" way, of looking at the rectifier tube charts in the RCA tube manual, making some approximations, and hoping for the best.

One thing about the choke - be careful not to mount it near input tubes, or anywhere it can inductively couple hum into high impedance wiring.

One other idea... if the choke causes too much voltage drop, see what happens when you substitute a GZ34 for the 5U4; the GZ34 is a cathode type tube and has much lower voltage drop than the old timey directly heated 5U4.
 
Last edited:
I was going to suggest a 5V4GA in place of the 5U4. Heathkit uses the 5V4 in their W4 amp, same output tubes. It has a little more drop than the 5AR4, but less than the 5U4 (I suspect the 5AR4 will over-volt things a bit)...

Regards,
Gordon.
 
I installed the choke between the two 32uf sections of the first multi-cap (which was my second option, above) and all seems well. The amp is running well and is quiet. I'm reading 426vdc on the output tube plates, so it doesn't seem like the choke lowered the B+ very much. Now that I have both my HF-22s up and running, I'm noticing that the amp I just worked on is running a little cooler, so I may add the same choke to the other one. Would the power transformer not getting as hot be directly attributable to having this choke in the filter supply?

I'll definitely check out the PSUD tool from Duncanamps when I have the chance -- thanks!
 
BTW, the choke fit nicely under the chassis, right near the power transformer but with the laminates running perpendicular to those of the PT. Drilled two little holes in the chassis and mounted it to the side...
 
Okay... I've gone through both amps and replaced all electrolytic caps, main filter supply and cathode caps, I've installed the previously mentioned chokes on both amps, and they're both working and sounding really good, but one amp is running hotter than the other (the power transformer). I've checked and/or replaced all coupling caps in the amp, checked how much current the power tubes were drawing (similar on both amps, around 40mA), swapped tubes from one amp to the other, and the situation remains unchanged. I took turns plugging each amp into a Kill-a-Watt receptacle/meter and the 'good' amp is idling at around .82A, the one that's getting hotter is idling at 1.04A, which seems a little high. I'm not sure what else to check with this amp. Could it be that the power transformer is compromised internally (windings beginning to short)?

Pics will be forthcoming when I have a chance. At this point, I'm spending all my spare time trying to figure this conundrum out...
 
Amps are identical, and both set up with fresh parts where it counts. Can't figure out for the life of me why the one is running hotter...
 
You didn't mention it, but can we assume that both units are producing the same heater and B+ voltages? If the one running warmer has lower voltages, that can be an indicator of internal strife within the power transformer.

Also, check both amps on your Kill-a-watt meter with all tubes removed, and compare the readings. The current draw will be lower of course, but if a significant difference still exists, then it must be due to the power transformers. You could even let them run for a period that way to see if one runs warmer than the other to verify that the different temps noted is transformer related.

If one has a shorted turn it will definitely draw more current and run hotter than the other. However, check the manufacturer's code to see if the transformers were manufactured by the same manufacturer. Eico used a number of vendors for their transformers, and it may simply be that they are two different transformers from two different manufacturers -- albeit they still look and act the same except for temp rise.

I have seen power transformers from Eico for both their HF-87/HF-89/HF-89A series and ST-70 models that even have obviously and notably different amounts of lamination thicknesses, due to being manufactured by different vendors. Therefore, sometimes use of an alternate vendor is obvious, and sometimes not, but basic construction differences can change current draw readings as well.

Finally, some rough back of napkin calculations indicate that based on the current draw levels from the various windings, an HF-22 with all tubes installed and operating, but not powering any external preamplifier should draw about .77 A from a 117 volt line, which is reasonably close to the one transformer that draws .82 A. If you measured the amp current draw while operating it from a more typical 122 vac line level that exists today, that would account for much of the discrepancy noted. If significantly different current draw levels are in fact noted with all tubes removed, then based on this exercise then, it would appear that the one transformer drawing over an amp during normal operation is outside of the norm.

I hope this helps!

Dave
 
Dave,
Thanks for the input! I've done a little more testing, and here's what I can report:
I read current draw with the Kill-a-Watt meter as follows:

Problem 'hot' amp Normal amp
no tubes in .13A (9.5 w) .13A (8.7 w)
Rect. tube in .2A (21.7 w) .2A (22.1 w)
Rect. & EF86 .21A (23.7 w) .21A (23.7 w)
Rect., EF86 & 6SN7 .26A (29.1 w) .26A (29.2 w)
all tubes in 1.03A (85 w) .8A (88.4 w)

I switched both amps to a 5R4 rectifier to lessen the current draw on the transformer, lowers the B+ a little, but not much. Both amps have IDENTICAL transformers, stamped 'S-30016 798902'. I've re-checked all capacitors and resistors, wiring, checked all solder joints... Tried yet another known good pair of 6L6s... no change.
The current measurements seem to rule out an internal short in the transformer, because, if so, it would be there all the time, regardless of what tubes are in or out, right? I'm still stumped...
 
Opps! Tried to set up that data in columns, but when I submitted the post, it got rid of my extra spacing. Hopefully, it still makes sense.
 
Agreed, the discrepancy comes into play only when the power tubes are installed, with all others already in place.

But now you have got a mystery: You have said that both output stages are drawing the same amount of current from their respective power supplies, based on measurements take at the output stages. However your indications on the primary side of the hot power transformer indicate that (by projection) the output stage in that unit is drawing more than in the other unit. It can't be both ways. Either one of them is inaccurate, or, there is a problem in the hot transformer that doesn't show up until a full load is placed on the transformer.

One combination you did not try is having all tubes in but the rectifier. If the current draw is equal then, but unequal with the rectifier tubes installed, AND both output stages are drawing the same amount of WATTS from the power supply (that is, the same current draw at the SAME B+ voltage), then something would surely seem amiss with the HV winding in the one transformer, which is usually the winding that has the most problems

Have you compared cold DC resistance values of the various windings between the two transformers?

Dave
 
Interesting, this is where a Megger comes in handy. As much old iron as I use, I wouldn't want to do without mine. I can test insulation / varnish up to 1000v for any leakage at all.
 
Yeah, well, definitely don't have access to a Megger...

I didn't check with all tubes except rectifier thinking that without rectifier, no B+. If there's a problem with the 6.3v filament, maybe that would show up then. I will do the DC resistance readings on the power transformer in the next day or so and post results.

Thanks for all advice!...
 
Mystery solved! While poking around inside the amps, taking various DC resistance readings of the power transformers I noticed the rectifier tube socket on the problem amp had been wired up wrong! Been that way since the kit was originally built back in the late '50's/early '60's, I guess! One of the rectifier plate (diode) wires was connected to pin 5 instead of pin 6. Amazingly, the amp still worked! I soldered the wire onto the correct pin and the high current reading on the Kill-a-Watt went away -- all is normal now.

Attached are some pics of the amps, showing the new electrolytic caps I installed and the chokes. Using a Discman to test it out right now, but ultimately I'll be listening to vinyl on my AR XA turntable with this system. One of these days I'll tackle the turntable (needs a new motor)...

Thanks for all the input! System sounds great!
 

Attachments

  • DSCN5637.jpg
    DSCN5637.jpg
    70.2 KB · Views: 47
  • DSCN5636.jpg
    DSCN5636.jpg
    62.8 KB · Views: 49
  • DSCN5630.jpg
    DSCN5630.jpg
    109.4 KB · Views: 52
How does one keep from getting lost in the shuffle here?...

I'm thinking it might be useful information for others -- my quest to get this amp working right. Probably other kits out there from a bygone era that weren't initially assembled quite right... been passed around from junk heap to junk heap... My experience getting these monoblock amps to work right together might serve as inspiration to others... Just wanted to bump this thread, maybe generate more valuable discussion... Been hanging on to the original HF-22 amp for maybe 15 years, finally found a mate for it, paid a few hundred bucks on eBay, been postponing the reconditioning of these amps... 'til now! Finally got them both reconditioned, sounding great! Now I'm going to have to go back to getting my Williamson amp working right! (also discussed on a previous thread...)
 
I installed the choke between the two 32uf sections of the first multi-cap (which was my second option, above) and all seems well. The amp is running well and is quiet. I'm reading 426vdc on the output tube plates, so it doesn't seem like the choke lowered the B+ very much. Now that I have both my HF-22s up and running, I'm noticing that the amp I just worked on is running a little cooler, so I may add the same choke to the other one. Would the power transformer not getting as hot be directly attributable to having this choke in the filter supply?

I'll definitely check out the PSUD tool from Duncanamps when I have the chance -- thanks!

interesting......do you recall what kind of choke was used? ty
 
Back
Top Bottom