Globe valve does not require a bypass as it is designed to open / close / control with a large differential pressure.
Other isolation valves are, in general, not suited for large differential pressures / stopping or starting flow.
Some gate valves are better than others depending on the fluid, differential pressure and location and types of seals.
Bypasses are more common on gas systems which can generate very large velocities on opening / closing but usually only where there is a significant differential pressure or where there is a big volume on the lower pressure side. For short pipe lengths it is often not required.
Remember - More details = better answers
Also: If you get a response it's polite to respond to it.
BUY BRESSURE BALANCED GATE VALVES and you won't need bypasses.
Control valves are often blocked and bypassed using a block valves and a manual throttling valve should the control valve need to be taken out of service temporarly for maintenance purposes.
IMO they don't need as much torque to open at high differential pressure as they are more leak tight across full pressure range, and as they do not rely on differential pressure to make the seal they do not get stuck, or develop bent stems as easily as other types. This type of valve should always be specified for pipeline critical shutoff valves when gates are favoured over trunion mounted ball valves.
I agree - expanding wedge gate valves are much better. However, when you open against high pressure, you still get some impact on the valve as the seals lift off and the gate starts to move.
Remember - More details = better answers
Also: If you get a response it's polite to respond to it.
In your experience do operators let you rely on an expanding (split) gate valve as a double isolation for working with divers? We have specified them in the past but clients/operators have been reluctant to and so you would end up having to install two and so lose any benefit they would offer (fewer valves, cheaper, smaller structure etc).
The double isolation idea IMO is not an acceptable "standard". Variation of the requirements for when double isolation is required or not among operators that use it is too wide to make any sense of why the operator has chosen to make such a thing a requirement. Some operators require double isolation at ANSI#300, some at ANSI#600, some for specific services, some for any service, most operators do not require it at all. Some operators are putting double isolation valves everywhere even if they don't have any experience to show that is necessary. Copy cats? Monkey see monkey do. Makes no sense. I've seen some drawings with some guy trying to double up on very valve on a pig launcher these days. Since thiis double isolation thing is nowhere near a standard practice at all, and actually accepted and used by only a very few operators anywhere outside of the Mideast, or English North Sea, it is hard to say when it sreally hould be used or not with any degree of certainty. Given that, I'd say if it is required for your project, then use two cheap valves and a bypass. Otherwise use a good valve, with internal double block and cavity bleed. Leave the excessive requirements for truely hazardous services and Piper Alfa installations where over-regulation has replaced poor operating standards and good engineering judgement.
Generally you'll put a bypass around a gate valve as a "pressure up" line or "start up" line to have a more controlled flow to the inlet of some type of vessel/facility to slowly start it up. Gate valves do not control flow well, at all, and if you're not careful you could damage them if the fluid velocity is quite high across the valve when cracking it open. And, as was mentioned earlier in the thread, the pressure differential across the valve I describe here would push it against the seat such that it might be very difficult to open and could cause muscle strain/injury when trying to do so.
So, in my experience, what I've seen is a small line as a bypass around a facility inlet to avoid having someone crack open that big inlet valve too much when starting up.
A bypass around a globe valve, haven't seen that except if the globe valve is automated, in which case you would also put isolation valves around that globe control-valve so you can work on it if you need to. Change the trim, etc. So in the meantime, you'd have a bypass around the "valve station", usually using a globe valve also, but smaller, that someone could adjust manually while the control valve is out of service.
Are you looking at a drawing with such a bypass, or did you see one in the field and were curious, or did someone ask you to put a bypass in and you weren't sure why?
1. To fill/pressure the downstream side of an isolation valve before opening it. This means the pressure will equal over the valve gate and making it much easier to open. Only works if the downstream side is a closed system.
2. An open by-pass around a control valve gives a minimum flow when the control valve is closed. Stops having to use the larger valve at small opening.
Gate valve bypass valves are also used a lot in steam systems - for the same reasons CheMatt mentions above. The bypass valve (usually a globe valve) allows a slow, controlled system warm-up, before opening the main valve.
There's an MSS standard that covers relative sizes and orientations. (Sorry - don't have my MSS standards handy)
I think the item everyone forgets when deciding whether to go for double isolation (maybe I'll try it the next time I review or write and isolation philosophy) is that leakage past valves in practice is a function of the number of operations and the type of operations. By this I mean that using valves designed for isolation (gate, ball etc) for starting and stopping flows which generate a high pressure differential on opening or a high velocity / flow rate on closing may well need a second valve for isolation if you're going to need positive double isolation.
Having said that I can't see why the expanding wedge gate should not be classified as a DIB valve and allow operations when you activate the body bleed. Otherwise why bother?
I always fight against bypasses around control valves unless you duplicate it. If it's there to control something then installing a bypass, gate valve or manual valve simply defeats the whole purpose of the valve in the first place. I've fought this many times in HAZOPS and nearly always won the argument.
Remember - More details = better answers
Also: If you get a response it's polite to respond to it.
Right. I've used the same argument many times. Any bypass around an isolation valve is questionable practice, but sometimes unavoidable, such as when its purpose is to allow a bit of fuel to come into a station for a black start. Keep bypasses around isolation valves limited to where they really are needed, and where they are needed, lock them in the closed position.
