G-PULTEC problems, please help!

[gyraf]

2K is fine

 

[qwerty8787]

Hi everyone!

Got a PQD-2 on the bench, almost finished.
I am experiencing volume loss when circuit is engaged, about -1.5 dB, and an overall loss of about -3dB/-4dB (depending on channel, because of tube mismatch I guess).

I tried to put the 1K trimmer above the "cut high" rotary switch and managed to get 0dB overall loss when unit is bypassed. But I still have about the same loss between bypassed and active.

What am I missing?

Thanks to everyone in advance 

 

[gyraf]

managed to get 0dB overall loss when unit is bypassed. But I still have about the same loss between bypassed and active.

Question not entirely clear?

If you're having overall gain loss with the unit in your chain, bypassed or not, it's dependent on actual tube gain. We don't use gain regulated by feedback, so you depend on raw gain being about right.

Jakob E.

 

[qwerty8787]

Hy Jakob!

I wrote my question badly. I'll try to explain again my issue:

Before 1k-trim mod:

Ch1 Bypassed: ≈ -3.dB (got there with the trimmer)
Ch2 Bypassed: ≈ -4 dBdB (same as Ch1)

Ch1 Engaged: about -1.5dB LESS THAN when bypassed
Ch2 Engaged: about -1.5dB LESS THAN when bypassed

After 1k-trim mod:

Ch1 Bypassed: 0 dB (got there with the trimmer)
Ch2 Bypassed: 0 dB (same as Ch1, trimmed manually)

Ch1 Engaged: -1.30 dB
Ch2 Engaged: -1.35dB

So my point is: I managed to get perfectly matched channels when in bypass, but still experiencing some loss when active (actually this loss is pretty close between channels).

I'm quite sure tubes are not matched, cause I'm using random old vintage tubes I found in a drawer. But even if I throw in 2x ECC88 bought from Banzai as matched, the result is the same.

My guess is that this trim mod enables to compensate for tube gain mismatch but not for "eq" loss...or at least this is what I get seeing that mismatch is pretty close between channels after mod.

 

[miszt]

Just finished my stereo g-pultec/pqd2, sounds awsome, rich, smooth and well just awsome

I seem to have a imbalance somewhere though, i'm loosing a few db on one channel (left), the loss is still there in bypass mode

Some things i noticed

The right channel tube glows far more brightly than the left (they are brand new, new production electro-harmonix), coincidence that this channel is also louder? (dont know much about tubes or how the glow reflects its function) - i've read that tubes need to "burn in", but i presume that wouldnt be the cause of that kind of imbalance....could the tubes just be mismatched? (they were sold as a matched pair)

The only 2 components which are not matched less than 1% are the big 4.7k caps near the transformers (they are about 4% out from each other)... would these caps be responsible for (~4.5-6db loss, haven't measured exactly, just running noise and saws thru atm)

Anything else i should check?

Cant work out if this is the same issue as discussed above, using a trimmer to match channels? The problem form me is overall level difference rather than on specific freq bands (which all seem to be matched relative to the overall difference (so far))

 

[gyraf]

miszt - we've discussed this several times before in this thread.

Most part of raw-gain-mismatch comes from tube's gains that vary due to production tolerances (try swapping them)

Most part of bypass/in gain mismatch comes from mismatched ratio ResistanceLowCutPot/ResistanceHiBoostPot

Jakob E.

 

[miszt]

miszt - we've discussed this several times before in this thread.

Most part of raw-gain-mismatch comes from tube's gains that vary due to production tolerances (try swapping them)

Most part of bypass/in gain mismatch comes from mismatched ratio ResistanceLowCutPot/ResistanceHiBoostPot

Jakob E.

Thanks for clarifying a bit; i did read that, but its confused me, because my pots/switches are matched within 0.1% (switched using pcbgrinder switch calculator lists) - the cut and boost all appear to work on every step, so no loose/bad connections afaik

so can i ignore the switches and just focus on tube matching?

Or is the bypass loss only going to be down to the lowcut/highcut switches being somehow mismatched?


Update: i've gone over all the resistors, no mismatch/accidently placed one, i did also use meter to check (i know not very accurate in circuit, but it didnt give me anything to be concerned about) - unfortunetly my filter board is hard wired onto the gain board, so i cant accurately check the resistance of the full signal path of each channel either....waiting for caps to discharge then i'll give it a go tho...

But i'm guessing i just need better matched tubes (although confirmation would be good...

(everything on the filter board does what it should, all changes at each freq seem to be correct relative to the difference in total gain on each channel)


The only other pair of tubes i have are a pair which were sold as matched NOS Siemens, but arrived clearly not even the same brand; will give them a try later tho

Well at least i can use this in m/s mode in the meantime

update 2 - seems i have a dodgy tube in the "NOS" set, gain keeps dropping and climbing *sigh* ...but the EH work well in M/S mode with makeup for the lower gain tube

Can anyone recommend a UK supplier of properly matched tubes for this? Bought a few sets already but reall  cant afford to keep buying pairs and cross my fingers, would rather pay a little extra and know they will be equal gain straight off (i'm sure this must be possible? I cant believe manley for eg supply replacement tubes that are mismatched)

 

[miszt]

I asked the supplier I purchased the EH 6922 how close they could match the tubes; explaining that my two they had supplied are giving me upto 6db difference between left/right, i got this reply:

We match the 6922 on average within 1mA on the Anode and no more than 2mA.  This is normally an accepted good match.
You can request a specific specification on ordering and we would take this into consideration before dispatching.

no idea what that means or how that relates to the balance of left/right?

can anyone shed any light? or better yet recommend a shop that sells them specifically matched for stereo operation - if 6db difference is "acceptable", then I guess we all have different definitions of acceptable

 

[gyraf]

I think we mentioned earlier in this thread that socalled "matching" of small-signal tubes is not worth the paper the figures are written on.

Matching is for power tubes, ensuring approximately the same current vs. grid voltage.

For GPultec you'll want tubes matched for raw gain "Amplification factor" or µ

If you really need matching, the best way is to buy a handful and use the best-matching ones

Jakob E.

 

[miszt]

I think we mentioned earlier in this thread that socalled "matching" of small-signal tubes is not worth the paper the figures are written on.

Matching is for power tubes, ensuring approximately the same current vs. grid voltage.

For GPultec you'll want tubes matched for raw gain "Amplification factor" or µ

If you really need matching, the best way is to buy a handful and use the best-matching ones

Jakob E.

thanks for the info!

i didn't spot that particular post, read thru the whole thread and ran 100's of internet searches on the subject too lol

ah well, i'll stick with make up gain in my DAW for now, noise floor is low enough that it doesn't matter too much the way I'm using the EQ; amazed how quiet this thing actually is tbh, very happy



one last question:  I have a pair of NOS siemens, and I thought that one of them was faulty, the gain on that channel varied wildly and erratically - since I took them out and put in some EH's, I also spotted that one of my inductor connections had come loose (just one! somehow lol), so i fixed it and just carried on

now I've put the NOS tubes back in, I was going to make a note of which one had the varying gain and return them to the shop; but after ~45minutes so far, there has been no variance at all? ...could that dodgy inductor connection have been the cause of the tube gain randomly changing? (it was varying by quite a bit, ~20db up and down randomly (without touching any of the controls - i didn't note whether it was just while in bypass or not))


update: well over an hour playing with the EQ, and both NOS tubes are behaving themselves..so i guess it must have been the dodgy inductor connection?

 

[gyraf]

yup, intermittent inductor can easily mess up gain.

Other than that, it could be slightly dirty tube pins - common on NOS like the ones I use - that cleans up when inserted/extracted a couple of times..

Happy EQ'ing!

Jakob E.

 

[Vac11]

Hi Jakob,

I have one question ... Is it necessary for the capacitor on the output value 4u7F? IS it sufficient even less value? For example 0,47uF? (for build with Lundal ll5402)

I want to use at this point a better paper in oil capacitor from Jensen, and 4u7F value is a bit bigger (about 10x5 cm) and more expensive ...

Thanks for the clarification

M.

EDIT:

I found answer from august 2004:
http://groupdiy.com/index.php?topic=1649.0

"Yes, there is a 4.7uF/250V polyester cap driving the output transformer. The output is somewhat low impedance, so with a lower value capacitor you'll loose low-end......

Jakob E."

 

[darkus]

Hi all.

So ive used my G-Pultecs for a while and i have a list of problems i need to sort out.

Low boost kicks in almost full with a very slight pot turn, like 1-2 on the frontpanel scale. The low end at 2 sounds like its  at 6 and after mid way rotation the low end boost pot has no function.  High boost however works like complete opposite, it sounds like nothing is happening before mid way, its all in the 5-10 of the frontpanel scale.  Both attenuators are soft across the whole turn which is imo how they are supposed to be so i think they are ok.

Both units have very low output, i lose about -10dB in gain. Im using edcor WSM 15k:600 for both trannies, should i try Lundahls instead? Could they fix the boost issues as well?

Also something is going on with the panel light indicator in both units. Ive used Don Audio trafo https://www.don-audio.com/g-pultec-power-trafo-sec-220v-9v-5V and i have a BA9S lamp in the 5v wires withouth a resistor. The lights worked for first few days, then they didnt light up anymore for few weeks, and now it comes and goes. Sometimes when i turn the units on they might light up for few mins and or then they dont.  ???

If anyone could help me fix some of these issues id be so happy.  :-* :-* :-*

 

[gyraf]

1) check response curves of your pots. Linear/Log in the right places?
2) I don't know the edcors, but the price difference alone points towards them not being anything like the LL5402
3) I'm guessing at a loose bulb in the indicator. Screw it back in?

Jakob E.

 

[jonasmaks]

Hello,
I've measured G-Pultec Hi Boost frequency curves and found out that they are quite inconsistent quality wise - from Q=3 at 2.1kHz to Q=2 at 11kHz. I'm using stand alone Murata inductors for my build (R1900 series, as closest substitution to Toko which are practicaly impossible to find). Has anyone measured G-Pultec based on Toko or multitaped inductor frequency responses to compare (some links I found on forum are gone)? Could it be cure for such behaviour - adding some resistor in series with the inductor to compensate  higer Q? But graphs show that biggest difference of Q apears on same inductor (say on 11 and 16 kHz). Any thoughts (and measurement graphs ) would be appreciated!

 

[gyraf]

This is exactly what the math would predict, and how we would expect it to behave.

Does the unit somehow Sound wrong?

Jakob E.

 

[jonasmaks]

Hi Jakob,
Yes, we are powerless against math . I just thought the lower the C, the bigger is Q (http://www.radio-electronics.com/info/formulae/q-quality-factor/rlc-q-factor.php).
Here is nothing wrong with the sound - unit works great (kudos to you!) and compares perfectly with commercial Warm Audio model EQP-WA. Just sharing my own experience and trying to gather more knowledge about the use of different inductors.

Jonas M.

 

[febo2609]

If someone used a coil for G-Pultec eq1 from  Don Audio store.
What are the experiences with them.
Is Sowtek coils better than Don audio coil.
Best regards

 

[Audioman]

yours is a question or a statement?

If someone used a coil for G-Pultec eq1 from  Don Audio store.
What are the experiences with them.
Is Sowtek coils better than Don audio coil.
Best regards

 

[gyraf]

There's not a whole lot of difference between these inductors, even if manufacturers would like you to believe there is.

You may even roll your own, and it will probably be as good as any.

No Angel dust in there.

Jakob E.